vvEPA
United States March 2008
|ng™mental Protection EPA/600/R-07/093aB
Integrated Science Assessment
for Oxides of Nitrogen -
Health Criteria
(Second External Review Draft)
-------
EPA/600/R-07/093aB
March 2008
Integrated Science Assessment
for Oxides of Nitrogen - Health Criteria
National Center for Environmental Assessment-RTF Division
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
-------
DISCLAIMER
This document is a first external review draft being released for review purposes only and
does not constitute U.S. Environmental Protection Agency (EPA) policy. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
PREFACE
National Ambient Air Quality Standards (NAAQS) are promulgated by the United States
Environmental Protection Agency (EPA) to meet requirements set forth in Sections 108 and 109
of the U.S. Clean Air Act (CAA). Sections 108 and 109 require the EPA Administrator (1) to
list widespread air pollutants that reasonably may be expected to endanger public health or
welfare; (2) to issue air quality criteria for them that assess the latest available scientific
information on nature and effects of ambient exposure to them; (3) to set "primary" NAAQS to
protect human health with adequate margin of safety and to set "secondary" NAAQS to protect
against welfare effects (e.g., effects on vegetation, ecosystems, visibility, climate, manmade
materials, etc); and (5) to periodically review and revise, as appropriate, the criteria and NAAQS
for a given listed pollutant or class of pollutants.
The purpose of this revised Integrated Science Assessment (ISA) for Oxides of Nitrogen -
Health Criteria is to critically evaluate and assess the latest scientific information published
since that assessed in the above 1993 Nitrogen Oxides AQCD, with the main focus being on
pertinent new information useful in evaluating health effects data associated with ambient air
nitrogen oxides exposures. A First External Review Draft of this ISA (dated August 2007) was
released for public comment and was reviewed by the Clean Air Scientific Advisory Committee
(CASAC) in October 2007. Public comments and CASAC recommendations have been taken
into account in making revisions to the document for incorporation into this Second External
Review Draft ISA, which is now being released for public comment and CASAC review.
Subsequently, a final ISA will be prepared that addresses comments received. This final ISA
will be drawn on to provide inputs to risk and exposure analyses prepared by EPA's Office of
Air Quality Planning and Standards (OAQPS) to pose options for consideration by the EPA
March 2008 ii DRAFT-DO NOT QUOTE OR CITE
-------
Administrator with regard to proposal and, ultimately, promulgation of decisions on potential
retention or revision, as appropriate, of the current NO2 NAAQS.
Preparation of this document was coordinated by staff of EPA's National Center for
Environmental Assessment in Research Triangle Park (NCEA-RTP). NCEA-RTP scientific
staff, together with experts from other EPA/ORD laboratories and academia, contributed to
writing of document chapters. Earlier drafts of document materials were reviewed by non-EPA
experts in peer consultation workshops held by EPA. The document describes the nature,
sources, distribution, measurement, and concentrations of nitrogen oxides in outdoor (ambient)
and indoor environments. It also evaluates the latest data on human exposures to ambient
nitrogen oxides and consequent health effects in exposed human populations, to support decision
making regarding the primary (health-based) NO2 NAAQS.
NCEA acknowledges the valuable contributions provided by authors, contributors, and
reviewers and the diligence of its staff and contractors in the preparation of this document.
March 2008 iii DRAFT-DO NOT QUOTE OR CITE
-------
Integrated Science Assessment for Oxides of Nitrogen
Health Criteria
(Second External Review Draft)
1. INTRODUCTION 1-1
2. SOURCE TO TISSUE DOSE 2-1
3. INTEGRATED HEALTH EFFECTS OF NO2 EXPOSURE 3-1
4. PUBLIC HEALTH SIGNIFICANCE 4-1
5. INTEGRATIVE SUMMARY AND CONCLUSIONS 5-1
6. REFERENCES 6-1
March 2008 iv DRAFT-DO NOT QUOTE OR CITE
-------
Table of Contents
List of Tables ix
List of Figures xi
Authors, Contributors, and Reviewers xv
U.S. Environmental Protection Agency Project Team xxi
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC) xxiv
Abbreviations and Acronyms xxvii
1. INTRODUCTION 1-1
1.1 LEGISLATIVE REQUIREMENTS 1-1
1.2 HISTORY OF PRIMARY NO2 NAAQS REVIEWS 1-3
1.3 POLICY-RELEVANT QUESTIONS 1-4
1.4 DOCUMENT DEVELOPMENT 1-5
1.5 DOCUMENT ORGANIZATION 1-6
1.6 EPA FRAMEWORK FOR CAUSAL DETERMINATIONS 1-6
1.7 CONCLUSIONS 1-18
2. SOURCE TO TISSUE DOSE 2-1
2.1 INTRODUCTION 2-1
2.2 SOURCES AND ATMOSPHERIC CHEMISTRY 2-2
2.2.1 Sources of NOX 2-3
2.2.2 Chemical Transformations of NOX 2-3
2.2.3 O3 Formation 2-6
2.3 MEASUREMENT METHODS AND ASSOCIATED ISSUES 2-7
2.3.1 Measurement Methods Specific to NO2 2-9
2.3.2 Measurement of Total Oxidized Nitrogen Species
in the Atmosphere 2-9
2.4 AMBIENT CONCENTRATIONS OF NO2 AND ASSOCIATED
OXIDIZED NITROGEN SPECIES AND POLICY-RELEVANT
BACKGROUND CONCENTRATIONS 2-10
2.4.1 Ambient Concentrations 2-10
2.4.2 Historical [NO2] 2-13
2.4.3 Seasonal Variability in NO2 at Urban Sites 2-14
2.4.4 Diurnal Variability inNO2 Concentrations 2-15
2.4.5 Concentrations of NOz Species 2-17
2.4.6 Policy Relevant Background Concentrations of NO2 2-18
2.5 EXPOSURE ISSUES 2-19
2.5.1 Introduction 2-19
2.5.2 Personal Sampling of NO2 2-25
March 2008 v DRAFT-DO NOT QUOTE OR CITE
-------
Table of Contents
(cont'd)
Page
2.5.3 Spatial Variability inNO2 Concentrations 2-26
2.5.4 Traffic as a Source of NO2 2-32
2.5.5 Indoor Sources and Sinks of NO2 and Associated Pollutants 2-34
2.5.6 Relationships of Personal Exposures to
Ambient Concentrations 2-40
2.5.8 NO2 as a Component of Mixtures 2-51
2.6 DOSIMETRY OF INHALED NITROGEN OXIDES 2-59
3. INTEGRATED HEALTH EFFECTS OF NO2 EXPOSURE 3-1
3.1 RESPIRATORY MORBIDITY RELATED TO NO2 SHORT-TERM
EXPOSURE 3-3
3.1.1 Lung Host Defenses and Immunity 3-4
3.1.2 Airways Inflammation 3-10
3.1.3 Airways Hyperresponsiveness 3-15
3.1.4 Effects of Short-Term NO2 Exposure on Respiratory
Symptoms 3-26
3.1.5 Effects of Short-Term NO2 Exposure on Lung Function 3-39
3.1.6 Hospital Admissions and ED Visits for Respiratory
Outcomes 3-46
3.1.7 Summary and Integration—Respiratory Health Effects
with Short-Term NO2 Exposure 3-59
3.2 CARDIOVASCULAR EFFECTS ASSOCIATED WITH
SHORT-TERM NO2 EXPOSURE 3-62
3.2.1 Heart Rate Variability, Repolarization Changes,
Arrhythmia, and Markers of Cardiovascular Function
in Humans and Animals 3-62
3.2.2 Studies of Hospital Admissions and ED Visits for CVD 3-66
3.2.3 Summary of Evidence of the Effect of Short-Term NO2
Exposure on Cardiovascular Morbidity 3-70
3.3 MORTALITY ASSOCIATED WITH SHORT-TERM NO2
EXPOSURE 3-71
3.3.1 Multicity Studies and Meta-Analyses 3-71
3.3.2 Summary of Evidence of the Effect of Short-Term
NO2 Exposure on Mortality 3-77
3.4 RESPIRATORY EFFECTS ASSOCIATED WITH LONG-TERM
NO2 EXPOSURE 3-81
3.4.1 Lung Function Growth 3-81
3.4.2 Asthma Prevalence and Incidence 3-90
3.4.3 Respiratory Symptoms 3-93
3.4.4 Animal Studies of Long-Term Morphological Effects
to the Respiratory System 3-95
March 2008 vi DRAFT-DO NOT QUOTE OR CITE
-------
Table of Contents
(cont'd)
Page
3.4.5 Summary and Integration of Evidence on Long-Term
NO2 Exposure and Respiratory Illness and Lung
Function Decrements 3-96
3.5 OTHER MORBIDITY EFFECTS ASSOCIATED WITH
LONG-TERM NO2 EXPOSURE 3-100
3.5.1 Cancer Incidence Associated with Long-Term NO2
Exposure 3-100
3.5.2 Cardiovascular Effects Associated with Long-Term
NO2 Exposure 3-105
3.5.3 Reproductive and Developmental Effects Associated
with Long-Term NO2 Exposure 3-107
3.5.4 Summary of Other Morbidity Effects Associated with
Long-Term NO2 Exposure 3-111
3.6 MORTALITY ASSOCIATED WITH LONG-TERM EXPOSURE 3-111
3.6.1 U.S. Studies on the Long-Term NO2 Exposure Effects
on Mortality 3-111
3.6.2 European Studies on the Long-Term NO2 Exposure
Effects on Mortality 3-114
3.6.3 Summary of Evidence of the Effect of Long-Term NO2
Exposure on Mortality 3-118
4. PUBLIC HEALTH SIGNIFICANCE 4-1
4.1 DEFINING ADVERSE HEALTH EFFECTS 4-1
4.2 CONCENTRATION-RESPONSE FUNCTIONS AND POTENTIAL
THRESHOLDS 4-4
4.3 POTENTIALLY SUSCEPTIBLE POPULATIONS TO HEALTH
EFFECTS RELATED TO SHORT-TERM AND LONG-TERM
EXPOSURE TO NO2 4-6
4.3.1 Preexisting Disease as a Potential Risk Factor 4-6
4.3.2 Age-Related Variations in Susceptibility 4-9
4.3.3 Gender 4-10
4.3.4 Genetic Factors for Oxidant and Inflammatory Damage
from Air Pollutants 4-10
4.3.5 Populations with Potentially High Exposure 4-12
4.3.6 Socioeconomic Position 4-12
4.4 ESTIMATION OF POTENTIAL NUMBERS OF PERSONS IN
AT-RISK SUSCEPTIBLE POPULATION GROUPS IN THE
UNITED STATES 4-13
4.5 SUMMARY 4-16
5. INTEGRATIVE SUMMARY AND CONCLUSIONS 5-1
5.1 INTRODUCTION 5-1
March 2008 vii DRAFT-DO NOT QUOTE OR CITE
-------
Table of Contents
(cont'd)
Page
5.2 KEY FINDINGS RELATED TO THE SOURCE-TO-DOSE
RELATIONSHIP 5-2
5.2.1 Atmospheric Science and Ambient Concentrations 5-2
5.2.2 Exposure Assessment 5-4
5.3 KEY HEALTH EFFECTS FINDINGS 5-6
5.3.1 Findings from the Previous Review of the National
Ambient Air Quality Standard for Nitrogen Oxides 5-6
5.3.2 New Findings on the Health Effects of Exposure
to Nitrogen Oxides 5-7
5.4 CONCLUSIONS 5-20
APPENDIX 5A 5A-1
6. REFERENCES 6-1
March 2008 viii DRAFT-DO NOT QUOTE OR CITE
-------
List of Tables
Number Page
1.6-1. Decisive Factors to Aid in Judging Causality 1-15
2.5-1. Spatial Variability of NO2 in Selected United States Urban Areas 2-27
2.5-2. NC>2 Concentration Near Indoor Sources: Short-Term Averages 2-37
2.5-3. NC>2 Concentration Near Indoor Sources: Long-Term Averages 2-38
2.5-7. Pearson Correlation Coefficient Between Ambient NC>2 and
Ambient Copollutants 2-52
2.5-8. Pearson Correlation Coefficient Between NOX and
Traffic-Generated Pollutants 2-54
2.5-9. Pearson Correlation Coefficient Between Ambient NC>2 and
Personal Copollutants 2-55
2.5-10. Pearson Correlation Coefficient Between Personal NC>2 and
Ambient Copollutants 2-56
2.5-11. Pearson Correlation Coefficient Between Personal NC>2 and
Personal Copollutants 2-56
2.5-4A. Association Between Personal Exposure and Ambient Concentration 2-62
2.5-4B. Association Between Personal Exposure and Outdoor Concentration 2-65
2.5-5. Summary of Regression Models of Personal Exposure to
Ambient/Outdoor NO2 2-67
2.5-6. Indoor/Outdoor Ratio and the Indoor vs. Outdoor Regression Slope 2-70
3.1-1. Proposed mechanisms whereby NO2 and respiratory virus infections
may exacerbate upper and lower airway symptoms 3-6
3.1-2. Mean rates (SD) per 100 days at risk AND unadjusted rATE ratio
(RR)* for symptoms/activities over 12 weeks during the winter
heating period 3-28
4.1-1. Gradation of Individual Responses to Short-Term NO2 Exposure in
Persons with Impaired Respiratory Systems 4-3
4.4-1. Prevalence of Selected Respiratory Disorders by Age Group and by
Geographic Region in the United States (2004 [U.S. Adults] and 2005
[U.S. Children] National Health Interview Survey) 4-14
5.3-2. Key Human Health Effects of Exposure to
Nitrogen Dioxide—Clinical Studies'1 5-10
5.3-3. Summary of Toxicological Effects from NO2 Exposure
(Lowest-Observed-Effect Level based on category) 5-11
March 2008 ix DRAFT-DO NOT QUOTE OR CITE
-------
List of Tables
(cont'd)
Number
5.3-1. Summary of Evi dence from Epi demi ol ogi cal, Human
Clinical, and Animal Toxicological Studies on the Health
Effects Associated with Short- and Long-Term Exposure to NC>2 5-23
5.3-4. Legend for Figure 5.3-1: Summary of Epi demi ol ogi c Studies
Examining Short-Term Exposures to Ambient NC>2 and
Respiratory Outcomes 5-26
5 A. Effects of Short-Term NC>2 Exposure on Respiratory Outcomes
in the United States and Canada 5A-2
5B. Effects of Short-Term NO2 Exposure on Emergency Department
Visits and Hospital Admissions for Respiratory Outcomes in the
United States and Canada 5 A-11
March 2008 x DRAFT-DO NOT QUOTE OR CITE
-------
List of Figures
Number Page
1.6-1. Exposure-disease-Stress Model for Environmental Health Disparities 1-9
1.6-2. Potential Relationships of NOX With Adverse Health Effects 1-13
2-1. A generalized conceptual model for integrating research on
oxides of nitrogen pollution and human health effects 2-1
2.2-1. Schematic diagram of the cycle of reactive, oxidized N species
in the atmosphere 2-4
2.4-1. Location of ambient NC>2 monitors in the United States
as of November 5, 2007 2-11
2.4-2. Ambient concentrations of NC>2 measured at all monitoring sites
located within Metropolitan Statistical Areas in the United States
from 2003 through 2005 2-12
2.4-3a,b. Monthly average NC>2 concentrations for January 2002 (a)
and July 2002 (b) calculated by CMAQ (36 x 36 km
horizontal resolution) 2-14
2.4-4. Nationwide trend inNO2 concentrations 2-15
2.4-5a,b. Time series of 24-h average NC>2 concentrations at individual sites
in Atlanta, GA from 2003 through 2005 2-16
2.4-6a-d. Mean hourly NO2 concentrations on weekdays and weekends
measured at two sites in Atlanta, GA 2-17
2.4-7. Upper panel: Annual mean NO2 concentrations (in ppb)
in the United States 2-20
2.5-1. Percentage of time persons spend in different environments
in the United States 2-21
2.5-2. NO2 and NOX concentrations normalized to ambient values, plotted
as a function of downwind distance from the freeway 2-30
2.5-3. NO2 concentrations measured at 4 m (Van) and at 15 m at
NY Department of Environmental Conservation ambient monitoring
sites (DEC709406 and DEC709407) 2-31
2.5-4a. Distribution of correlation coefficients (U.S. studies)
between personal NC>2 exposure and ambient NC>2 concentrations
based on Fisher'sZ transform 2-41
2.5-4b. Distribution of correlation coefficients (European studies)
between personal NC>2 exposure and ambient NC>2 concentrations based
on Fisher's Z transform 2-41
2.5-5a-d. Correlations of NC>2 to Os versus correlations of NC>2 to CO
for Los Angeles, CA (2001-2005) 2-53
March 2008 xi DRAFT-DO NOT QUOTE OR CITE
-------
List of Figures
(cont'd)
Number Page
2.5-6. Composite, diurnal variability in 1-h average NO2 in urban areas 2-58
3.1-1. Studies of airways inflammatory responses in relation to the total
exposure toNO2, expressed as ppm-minutes 3-12
3.1-2. Airways responsiveness to allergen challenge in asthmatic subjects
following a single exposure toNC>2 3-18
3.1-3. Geometric mean symptom rates (95% confidence intervals) for
cough with phlegm (panel A) and proportions (95% confidence
intervals) of children absent from school for at least 1 day (panel B)
during the winter heating period grouped by estimated NC>2 exposure
at home and at school (n = number of children atthatNO2 level) 3-30
3.1-4. Adjusted association of increasing indoor NC>2 concentrations with
number of days with persistent cough (panel a) or shortness of breath
(panel b) for 762 infants during the first year of life 3-32
3.1-5. Odds ratios (95% confidence interval [CI]) for daily asthma symptoms
(panel A) and rate ratios (95% CI) for daily rescue inhaler use (panel B)
associated with shifts in within-subject concentrations of NO2 for
single- and joint (with PMi0)-pollutant models from the Childhood
Asthma Management Program (November 1993-September 1995) 3-36
3.1-6. Odds ratios (95% CI) for associations between asthma symptoms
and24-h average NO2 concentrations (per 20 ppb) 3-38
3.1-7. Odds ratios and 95% confidence intervals for associations between
asthma symptoms and 24-h average NO2 concentrations (per 20 ppb) from
multipollutant models 3-39
3.1-8. Relative Risks (95% CI) for hospital admissions or ED visits for all
respiratory disease stratified by all ages or children 3-47
3.1-9. Relative Risks (95% CI) for hospital admissions or ED visits for all
respiratory disease stratified by adults and older adults (>65 years) 3-48
3.1-10. Relative Risks (95% CI) for hospital admissions or emergency
department visits for all respiratory causes, standardized from
two-pollutant models adjusted for particle concentration 3-51
3.1-11. Relative Risks (95% CI) for hospital admissions or emergency
department visits for all respiratory causes, standardized from
two-pollutant models adjusted for gaseous pollutant concentration 3-52
3.1-12. Relative Risks (95% CI) for hospital admissions or emergency
department visits for asthma stratified by all ages or children 3-54
March 2008 xii DRAFT-DO NOT QUOTE OR CITE
-------
List of Figures
(cont'd)
Number Page
3.1-13. Relative Risks (95% CI) for hospital admissions or emergency
department (ED) visits for asthma stratified by adults and
older adults (>65 years) 3-55
3.2-1. Relative risks (95% CI) for associations of 24-h NO2 (per 20 ppb)
and daily 1 hour maximum* NO2 (per 30 ppb) with hospitalizations
or emergency department visits for cardiac diseases 3-68
3.2-2. Relative risks (95% CI) for associations of 24-h NO2 exposure
(per 20 ppb) and daily 1 h maximum NO2* (per 30 ppb) with
hospitalizations for all cerebrovascular disease 3-69
3.3-1. Posterior means and 95% posterior intervals of national average
estimates for NO2 effects on total mortality from nonexternal causes
at lags 0, 1, and 2 within sets of the 90 cities with pollutant
data available 3-73
3.3-2. Combined NO2 mortality risk estimates from multicity and
meta-analysis studies 3-78
3.3-3. Combined NO2 mortality risk estimates for broad cause-specific
categories from multicity studies 3-80
3.4-1. Decrements in forced expiratory volume in 1 s (FEVi) associated
with a 20-ppb increase in NO2 (A) and a 20-|ig/m3 increase in PMi0
(B) in children, standardized per year of follow-up 3-82
3.4-2. Decrements in forced vital capacity (FVC) associated with a
20-ppb increase in NO2 (A) and a 20-|ig/m3 increase in PMio
(B) in children, standardized per year of follow-up 3-83
3.4-3. Proportion of 18-year olds with a FEVi below 80% of the predicted
value plotted against the average levels of pollutants from 1994
through 2000 in the 12 southern California communities of the
Children's Health Study 3-85
3.4-4. Estimated annual growth in FEVi, of long-term ozone (63), parti culate
matter < 10 jim in diameter (PMio), and nitrogen dioxide (NO2) in girls
and boys 3-87
3.4-5. Odds ratios for within-community bronchitis symptoms associations
with NO2, adjusted for other pollutants in two-pollutant models for the
12 communities of the Children's Health Study 3-94
3.4-6. Biologic pathways of long-term NO2 exposure on morbidity 3-97
3.6-1. Age-adjusted, nonparametric smoothed relationship between NO2
and mortality from all causes in Oslo, Norway, 1992 through 1995 3-117
March 2008 xiii DRAFT-DO NOT QUOTE OR CITE
-------
List of Figures
(cont'd)
Number Page
3.6-2. Total mortality relative risk estimates from long-term studies 3-119
4.1-1. The frequency distribution of hypothetical health outcome (A) and the
consequence of a shift in the population mean on the tails of the
distribution (B) 4-2
4.4-1. Fraction of the population living within a specified distance from
roadways 4-17
5.3-1. Summary of Epidemiologic Studies Examining Short-term
Exposures to Ambient NC>2 and Respiratory Outcomes 5-9
March 2008 xiv DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
Authors
Dr. Dennis J. Kotchmar (NOX Team Leader)—National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Mary Ross (Branch Chief)—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Kathleen Belanger, Yale University, Epidemiology and Public Health, 60 College Street,
New Haven, CT 06510-3210
Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Douglas Bryant—Cantox Environmental Inc., 1900 Minnesota Court, Mississauga, Ontario
L8S IPS
Dr. Ila Cote—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
Dr. Mark Frampton—Strong Memorial Hospital, 601 Elmwood Ave., Box 692, Rochester, NY
14642-8692
Dr. Janneane Gent—Yale University, CPPEE, One Church Street, 6th Floor, New Haven, CT
06510
Dr. Vic Hasselblad—Duke University, 29 Autumn Woods Drive, Durham, NC 27713
Dr. Kazuhiko Ito—New York University School of Medicine, 57 Old Forge Road, Tuxedo, NY
10987
Dr. Jee Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Ellen Kirrane—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Thomas Long—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Thomas Luben—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
March 2008 xv DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
(cont'd)
Authors
(cont'd)
Dr. Andrew Maier—Toxicology Excellence for Risk Assessment, 2300 Montana Avenue,
Suite 409, Cincinnati, OH 45211
Dr. Qingyu Meng—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Joseph Pinto—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Paul Reinhart—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Lori White—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Contributors
Dr. Dale Allen, University of Maryland, College Park, MD
Dr. Jeffrey Arnold—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Barbara Buckley—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Louise Camalier, U.S. EPA, OAQPS, Research Triangle Park, NC
Ms. Rebecca Daniels, MSPH—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Russell Dickerson, University of Maryland, College Park, MD
Dr. Tina Fan, EOHSI/UMDNJ, Piscataway, NJ
Dr. Arlene Fiore, NOAA/GFDL, Princeton, NJ
March 2008 xvi DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
(cont'd)
Contributors
(cont'd)
Dr. Panos Georgopoulos, EOHSI/UMDNJ, Piscataway, NJ
Dr. Larry Horowitz, NOAA/GFDL, Princeton, NJ
Dr. William Keene, University of Virginia, Charlottesville, VA
Dr. Randall Martin, Dalhousie University, Halifax, Nova Scotia
Dr. Maria Morandi, University of Texas, Houston, TX
Dr. William Munger, Harvard University, Cambridge, MA
Mr. Charles Piety, University of Maryland, College Park, MD
Dr. Jason Sacks—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Sandy Sillman, University of Michigan, Ann Arbor, MI
Dr. Jeffrey Stehr, University of Maryland, College Park, MD
Dr. Helen Suh, Harvard University, Boston, MA
Ms. Debra Walsh—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Charles Wechsler, EOHSI/UMDNJ, Piscataway, NJ
Dr. Clifford Weisel, EOHSI/UMDNJ, Piscataway, NJ
Dr. Jim Zhang, EOHSI/UMDNJ, Piscataway, NJ
Reviewers
Dr. Tina Bahadori—American Chemistry Council, 1300 Wilson Boulevard, Arlington, VA
22209
March 2008 xvii DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
(cont'd)
Reviewers
(cont'd)
Dr. Tim Benner—Office of Science Policy, Office of Research and Development, Washington,
DC 20004
Dr. Daniel Costa—National Program Director for Air, Office of Research and Development,
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Robert Devlin—National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Chapel Hill, NC
Dr. Judy Graham—American Chemistry Council, LRI, 1300 Wilson Boulevard, Arlington, VA
22209
Dr. Stephen Graham—Office of Air and Radiation, U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711
Ms. Beth Hassett-Sipple—U.S. Environmental Protection Agency (C504-06), Research Triangle
Park, NC 27711
Dr. Gary Hatch—National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
27711
Dr. Scott Jenkins—Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency (C504-02), Research Triangle Park, NC 27711
Dr. David Kryak—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Mr. John Langstaff—U.S. Environmental Protection Agency (C504-06), Research Triangle Park,
NC27711
Dr. Morton Lippmann—NYU School of Medicine, 57 Old Forge Road, Tuxedo, NY 10987
Dr. Thomas Long—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Karen Martin—Office of Air and Radiation, U.S. Environmental Protection Agency
(C504-06), Research Triangle Park, NC 27711
March 2008 xviii DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
(cont'd)
Reviewers
(cont'd)
Dr. William McDonnell—William F. McDonnell Consulting, 1207 Hillview Road, Chapel Hill,
NC27514
Dr. Dave McKee—Office of Air and Radiation/Office of Air Quality Planning and Standards,
U.S. Environmental Protection Agency (C504-06), Research Triangle Park, NC 27711
Dr. Lucas Neas—National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Chapel Hill, NC 27711
Dr. Russell Owen—National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
27711
Dr. Haluk Ozkaynak—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jennifer Peel—Colorado State University, 1681 Campus Delivery, Fort Collins, CO 80523-
1681
Mr. Harvey Richmond—Office of Air Quality Planning and Standards/Health and
Environmental Impacts Division, U.S. Environmental Protection Agency (C504-06), Research
Triangle Park, NC 27711
Mr. Joseph Somers—Office of Transportation and Air Quality, U.S. Environmental Protection
Agency, 2000 Traverwood Boulevard, Ann Arbor, MI 48105
Ms. Susan Stone—U.S. Environmental Protection Agency (C504-06), Research Triangle Park,
NC27711
Dr. John Vandenberg—National Center for Environmental Assessment (B243-01), Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
27711
Dr. Alan Vette—National Exposure Research Laboratory, Office of Research and Development,
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Mr. Ron Williams—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
March 2008 xix DRAFT-DO NOT QUOTE OR CITE
-------
Authors, Contributors, and Reviewers
(cont'd)
Reviewers
(cont'd)
Dr. William Wilson—Office of Research and Development, National Center for Environmental
Assessment (B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC
27711
March 2008 xx DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency Project Team
for Development of Integrated Scientific Assessment
for Oxides of Nitrogen
Executive Direction
Dr. Ila Cote (Acting Director)—National Center for Environmental Assessment-RTF Division,
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Debra Walsh (Deputy Director)—National Center for Environmental Assessment-RTF
Division, (B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Scientific Staff
Dr. Dennis Kotchmar (NOX Team Leader)—National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jeff Arnold—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Barbara Buckley—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Rebecca Daniels—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jee Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Ellen Kirrane—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Tom Long—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Thomas Luben—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Qingyu Meng—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
March 2008 xxi DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency Project Team
for Development of Integrated Scientific Assessment
for Oxides of Nitrogen
(cont'd)
Scientific Staff
(cont'd)
Dr. Joseph Pinto—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Paul Reinhart—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Mary Ross—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jason Sacks—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Lori White—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. William Wilson—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Technical Support Staff
Ms. Ella King—Executive Secretary, National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Emily R. Lee—Management Analyst, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Ellen F. Lorang—Information Manager, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Christine Searles—Management Analyst, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
March 2008 xxii DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency Project Team
for Development of Integrated Scientific Assessment
for Oxides of Nitrogen
(cont'd)
Technical Support Staff
(cont'd)
Mr. Richard Wilson—Clerk, National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Document Production Staff
Ms. Barbra H. Schwartz—Task Order Manager, Computer Sciences Corporation,
2803 Slater Road, Suite 220, Morrisville, NC 27560
Mr. John A. Bennett—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852
Mr. David Casson—Publication/Graphics Specialist, TekSystems, 1201 Edwards Mill Road,
Suite 201, Raleigh, NC 27607
Mrs. Melissa Cesar—Publication/Graphics Specialist, Computer Sciences Corporation,
2803 Slater Road, Suite 220, Morrisville, NC 27560
Mr. Eric Ellis—Records Management Technician, InfoPro, Inc., 8200 Greensboro Drive, Suite
1450, McLean, VA 22102
Ms. Kristin Hamilton—Publication/Graphics Specialist, TekSystems, 1201 Edwards Mill Road,
Suite 201, Raleigh, NC 27607
Ms. Stephanie Harper—Publication/Graphics Specialist, TekSystems, 1201 Edwards Mill Road,
Suite 201, Raleigh, NC 27607
Ms. Sandra L. Hughey—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852
Dr. Barbara Liljequist—Technical Editor, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560
Ms. Molly Windsor—Graphic Artist, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560
March 2008 xxiii DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency
Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
CASAC NOX and SOX Primary NAAQS Review Panel
Chair
Dr. Rogene Henderson*, Scientist Emeritus, Lovelace Respiratory Research Institute,
Albuquerque, NM
Members
Mr. Ed Avol, Professor, Preventive Medicine, Keck School of Medicine, University of Southern
California, Los Angeles, CA
Dr. John R. Balmes, Professor, Department of Medicine, Division of Occupational and
Environmental Medicine, University of California, San Francisco, CA
Dr. Ellis Cowling*, University Distinguished Professor At-Large, North Carolina State
University, Colleges of Natural Resources and Agriculture and Life Sciences, North Carolina
State University, Raleigh, NC
Dr. James D. Crapo [M.D.]*, Professor, Department of Medicine, National Jewish Medical and
Research Center, Denver, CO
Dr. Douglas Crawford-Brown*, Director, Carolina Environmental Program; Professor,
Environmental Sciences and Engineering; and Professor, Public Policy, Department of
Environmental Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel
Hill, NC
Dr. Terry Gordon, Professor, Environmental Medicine, NYU School of Medicine, Tuxedo, NY
Dr. Dale Hattis, Research Professor, Center for Technology, Environment, and Development,
George Perkins Marsh Institute, Clark University, Worcester, MA
Dr. Patrick Kinney, Associate Professor, Department of Environmental Health Sciences,
Mailman School of Public Health, Columbia University, New York, NY
Dr. Steven Kleeberger, Professor, Laboratory Chief, Laboratory of Respiratory Biology,
NIH/NIEHS, Research Triangle Park, NC
Dr Timothy Larson, Professor, Department of Civil and Environmental Engineering, University
of Washington, Seattle, WA
March 2008 xxiv DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency
Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
CASAC NOX and SOX Primary NAAQS Review Panel
(cont'd)
Members
(cont'd)
Dr. Kent Pinkerton, Professor, Regents of the University of California, Center for Health and
the Environment, University of California, Davis, CA
Mr. Richard L. Poirot*, Environmental Analyst, Air Pollution Control Division, Department of
Environmental Conservation, Vermont Agency of Natural Resources, Waterbury, VT
Dr. Edward Postlethwait, Professor and Chair, Department of Environmental Health Sciences,
School of Public Health, University of Alabama at Birmingham, Birmingham, AL
Dr. Armistead (Ted) Russell*, Georgia Power Distinguished Professor of Environmental
Engineering, Environmental Engineering Group, School of Civil and Environmental
Engineering, Georgia Institute of Technology, Atlanta, GA
Dr. Richard Schlesinger, Associate Dean, Department of Biology, Dyson College, Pace
University, New York, NY
Dr. Christian Seigneur, Vice President, Atmospheric and Environmental Research, Inc., San
Ramon, CA
Dr. Elizabeth A. (Lianne) Sheppard, Research Professor, Biostatistics and Environmental &
Occupational Health Sciences, Public Health and Community Medicine, University of
Washington, Seattle, WA
Dr. Frank Speizer [M.D.]*, Edward Kass Professor of Medicine, Channing Laboratory,
Harvard Medical School, Boston, MA
Dr. George Thurston, Associate Professor, Environmental Medicine, NYU School of Medicine,
New York University, Tuxedo, NY
Dr. James Ultman, Professor, Chemical Engineering, Bioengineering Program, Pennsylvania
State University, University Park, PA
Dr. Ronald Wyzga, Technical Executive, Air Quality Health and Risk, Electric Power Research
Institute, P.O. Box 10412, Palo Alto, CA
March 2008 xxv DRAFT-DO NOT QUOTE OR CITE
-------
U.S. Environmental Protection Agency
Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
CASAC NOX and SOX Primary NAAQS Review Panel
(cont'd)
SCIENCE ADVISORY BOARD STAFF
Dr. Angela Nugent, CASAC Designated Federal Officer, 1200 Pennsylvania Avenue, N.W.,
Washington, DC, 20460, Phone: 202-343-9981, Fax: 202-233-0643 (nugent.angela@epa.gov)
(Physical/Courier/FedEx Address: Angela Nugent, Ph.D, EPA Science Advisory Board Staff
Office (Mail Code 1400F), Woodies Building, 1025 F Street, N.W., Room 3614, Washington,
DC 20004, Telephone: 202-343-9981)
* Members of the statutory Clean Air Scientific Advisory Committee (CASAC) appointed by the EPA
Administrator
March 2008 xxvi DRAFT-DO NOT QUOTE OR CITE
-------
Abbreviations and Acronyms
a
ACS
ADP
a;
AIRE
AM
APEX
APHEA
AQCD
AQS
ATS
BAL
BALF
BHPN
BHR
Br
Cx T
Ca++
CAA
CALINE4
CAMP
CAPS
CAPs
CARS
CASAC
CC16
CDC
CHAD
CHF
CHS
CI
CMAQ
CO
C02
COD
brackets signifying concentration(s)
alpha; the ratio of a person's exposure to a pollutant of ambient
origin to the pollutant's ambient concentration
American Cancer Society
adenosine dinucleotide phosphate
air exchange rate for microenvironment /'
Asma Infantile Ricerca (Italian study)
alveolar macrophage
Air Pollution Exposure (model)
Air Pollution on Health: a European Approach (study)
Air Quality Criteria Document
Air Quality System (database)
American Thoracic Society
bronchoalveolar lavage
bronchoalveolar lavage fluid
7V-bis(2-hydroxyl-propyl)nitrosamine
bronchial hyperresponsiveness
bromine
concentration x time; concentration times duration of exposure
calcium ion
Clean Air Act
California line source dispersion (model)
Childhood Asthma Management Program
cavity attenuated phase shift (monitor)
concentrated ambient particles
California Air Resources Board
Clean Air Scientific Advisory Committee
Clara cell 16-kDa protein
Centers for Disease Control and Prevention
Consolidated Human Activity Database
congestive heart failure
Children's Health Study
confidence interval
Community Multiscale Air Quality (model)
carbon monoxide
carbon dioxide
coefficient of divergence
March 2008
XXVll
DRAFT-DO NOT QUOTE OR CITE
-------
CoH
COPD
CRP
CTM
CVD
DEPcCBP
DHHS
DMA
DMN
DNA
DOAS
Ea
EC
ECP
ED
ELF
Ena
EPA
EPO
ER
ETS
FEF25
FEF75
F£NO
FEVo.s
FEVi
Finfi
FRM
FVC
GAM
GEE
GEOS-CHEM
GIS
GM-CSF
GSH
GST
H+
HCHO
HDL
coefficient of haze
chronic obstructive pulmonary disease
C-reactive protein
Chemistry-transport model
cardiovascular disease
diesel exhaust particulates extract-coated carbon black particles
U.S. Department of Health and Human Services
dimethylamine
dimethylnitrosamine
deoxyribonucleic acid
differential optical absorption spectroscopy
a person's exposure to pollutants of ambient origin
elemental carbon
eosinophil cationic protein
emergency department
epithelial lining fluid
a person's exposure to pollutants that are not of ambient origin
U.S. Environmental Protection Agency
eosinophil peroxidase
emergency room
environmental tobacco smoke
forced expiratory flow at 25% of vital capacity
forced expiratory flow at 25 to 75% of vital capacity
forced expiratory flow at 75% of vital capacity
fractional exhaled nitric oxide
forced expiratory volume in 0.5 second
forced expiratory volume in 1 second
the infiltration factor for microenvironment /'
Federal Reference Method
forced vital capacity
Generalized Additive Model(s)
generalized estimating equation(s)
three-dimensional, global model of atmospheric chemistry driven by
assimilated Goddard Earth Orbiting System observations
Geographic Information System
granulocyte-macrophage colony stimulating factor
glutathione
glutathione S-transferase (e.g., GSTM1, GSTP1, GSTT1)
hydrogen ion
formaldehyde
high-density lipoprotein cholesterol
March 2008
XXVlll
DRAFT-DO NOT QUOTE OR CITE
-------
HNO3
HNO4
HONO
HR
HRV
HS
H2SO4
hv
ICAM-1
ICD, ICD9
id
Ig
fflD
IIASA
IL
He
IN
IOM
IQR
IS
ISA
ISAAC
ki
LDH
LIF
LOESS
LRD
LT
MEF25
MEF50
MEF75
MENTOR
MI
MMEF
MoOx
MOZART
MPO
MPP
MSA
N
n
nitric acid
pernitric acid
nitrous acid
heart rate
heart rate variability
hemorrhagic stroke
sulfuric acid
solar ultraviolet proton
intercellular adhesion molecule-1
International Classification of Diseases, Ninth Revision
identification
immunoglobulin (e.g., IgA, IgE, IgG)
ischemic heart disease
International Institute for Applied Systems Analysis
interleukin (e.g., IL-6, IL-8)
isoleucine
inorganic particulate species
Institute of Medicine
interquartile range
ischemic stroke
Integrated Science Assessment
International Study of Asthma and Allergies in Children
pollutant specific decay rate in microenvironment /'
lactate dehydrogenase
laser-induced fluorescence
locally estimated smoothing splines
lower respiratory disease
leukotriene (e.g., LTB4, LTC4, LTD4, LTE4)
maximal expiratory flow at 25%
maximal expiratory flow at 50%
maximal expiratory flow at 75%
Modeling Environment for Total Risk
myocardial infarction
maximal midexpiratory flow
molybdenum oxide
Model for Ozone and Related Chemical Tracers
myeloperoxidase
multiphase processes
metropolitan statistical area
nitrogen
number of observations
March 2008
XXIX
DRAFT-DO NOT QUOTE OR CITE
-------
Na+
NAAQS
NaAsO2
NAL
NAMS
NAS
NCo.oi-o.io
NCHS
NCICAS
NDMA
NEI
NERL
2NF
NHAPS
NHIS
NHX
NK
NLCS
NMMAPS
NMOR
NN
NO
NO2
NO2
NO3
NO3
NOX
NOY
NOZ
NOAANCEP
1NP
2NP
NR,N/R
NRC
NSA
03
oc
OH
sodium ion
National Ambient Air Quality Standards
sodium arsenite
nasal lavage
National Air Monitoring Stations
National Academy of Sciences
particle number concentration for particle aerodynamic diameter
between 10 and 100 nm
National Center for Health Statistics
National Cooperative Inner-City Asthma Study
7V-nitrosodimethylamine
National Emissions Inventory
National Exposure Research Laboratory
2-nitrofluoranthene
National Human Activity Pattern Survey
National Health Interview Survey
reduced nitrogen compounds (NH3, NH4+)
natural killer (lymphocytes)
the Netherlands Cohort Study on Diet and Cancer
National Morbidity, Mortality, and Air Pollution Study
7V-nitrosomorpholine
nitronapthalene
nitric oxide
nitrogen dioxide
nitrite ion
nitrate radical
nitrate ion
sum of NO and NO2
sum of NOX and NOZ, total oxidized nitrogen
sum of all inorganic and organic reaction products of NOX (HONO,
HNO3, HNO4, organic nitrates, particulate nitrate, nitro-PAHS, etc.)
U.S. National Oceanic and Atmospheric Administration's National
Center for Environmental Prediction
1-nitropyrene
2-nitropyrene
not reported
National Research Council
nitrosating agent
ozone
organic carbon
hydroxyl radical
March 2008
XXX
DRAFT-DO NOT QUOTE OR CITE
-------
OR
OVA
P,P
P90
PAARC
PAF
PAHs
PAMS
PAN
PANs
PaO2
Pb
PD20-FEVi
PD100
PEACE
PEF
PEFR
Pi
PIH
PM
PMiQ-2.5
PM2.5
PMN
pN03
POM
ppb
ppm
ppt
PRB
PT
PUFA
R
r
R2
RAPS
RCS
RONO2
odds ratio
ovalbumin
probability value
90th percentile
French air pollution and chronic respiratory diseases study
paroxysmal atrial fibrillation
polycyclic aromatic hydrocarbons
Photochemical Aerometric Monitoring System
peroxyacetyl nitrate
peroxyacyl nitrates
pressure of arterial oxygen
lead
provocative dose that produces a 20% decrease in FEVi
provocative dose that produces a 100% increase in SRaw
Pollution Effects on Asthmatic Children in Europe (study)
peak expiratory flow
peak expiratory flow rate
pollutant specific penetration coefficient for microenvironment /
primary intracerebral hemorrhage
particulate matter
particulate matter with an aerodynamic diameter of < lOjim
coarse particulate matter
fine particulate matter
polymorphonuclear leukocytes
particulate nitrate
particulate organic matter
parts per billion (by volume)
parts per million (by volume)
parts per trillion (by volume)
Policy Relevant Background
prothombin time
polyunsaturated fatty acids
intraclass correlation coefficient; organic radical
correlation coefficient
coefficient of determination
Pearson's correlation coefficient
Spearman's rank correlation coefficient
Regional Air Pollution Study
random component superposition
organic nitrates
March 2008
XXXI
DRAFT-DO NOT QUOTE OR CITE
-------
ROS
RR
RSV
S
SAPALDIA
SAR
SCE
SD
SE
SEP
SES
SGA
SHEDS
SIDS
SLAMS
SO2
SO42"
SRaw
STN
T
TEA
Th2
TNF
TSP
TVOCs
TX
UFP
URI
V
Val
VOCs
VWF
WBC
Yi
reactive oxygen species
relative risk
respiratory syncytial virus
microenvironmental source strength
Study of Air Pollution and Lung Diseases in Adults
Site Audit Report
sister chromatid exchange
standard deviation
standard error
social-economic position
social-economic status
small for gestational age
Simulation of Human Exposure and Dose System
sudden infant death syndrome
State and Local Air Monitoring Stations
sulfur dioxide
sulfate ion
specific airways resistance
Speciation Trends Network
tau; atmospheric lifetime
triethanolamine
T-derived helper 2 lymphocyte
tumor necrosis factor (e.g., TNF-a)
total suspended particulates
total volatile organic compounds
thromboxane (e.g., TXA2, TXB2)
ultrafine particles; <0.1 jim diameter
upper respiratory infections
volume of the microenvironment
valine
volatile organic compounds
von Willibrand Factor
white blood cell
the fraction of time people spend in microenvironment /
the fraction of time people spend outdoors
Fisher's transform of the correlation coefficient
March 2008
XXXll
DRAFT-DO NOT QUOTE OR CITE
-------
i 1. INTRODUCTION
2
O
4 The draft Integrated Science Assessment (ISA) presents a concise review, synthesis, and
5 evaluation of the most policy-relevant science and communicates critical science judgments
6 relevant to the review of national ambient air quality standards (NAAQS). In doing so, the
7 evaluation focuses on the studies published since the most recent review, and builds upon key
8 conclusions presented in previous U.S. Environmental Protection Agency (EPA) reports. This
9 strategy of building on past findings is more efficient than starting with a new review of the
10 pertinent literature and more effectively addresses the large body of work since the previous
11 reviews. This draft ISA forms the scientific foundation for the review of the primary (health-
12 based) NAAQS for nitrogen dioxide (NO2).1 The ISAs are accompanied by a series of Annexes
13 that provide more detailed summaries of the most pertinent scientific literature.
14 The draft ISA is intended to "accurately reflect the latest scientific knowledge useful in
15 indicating the kind and extent of identifiable effects on public health which may be expected
16 from the presence of [a] pollutant in ambient air" (Clean Air Act, Section 108 [U.S. Code,
17 2003a]). Scientific research is incorporated from atmospheric sciences, air quality analyses,
18 exposure assessment, dosimetry, toxicology, clinical studies, and epidemiology. Annexes to the
19 draft ISA also provide more details of the most pertinent scientific literature. The draft ISA and
20 the Annexes serve to update and revise the information included in the 1993 Air Quality Criteria
21 Document (AQCD) for Nitrogen Oxides (U.S. Environmental Protection Agency, 1993).
22 In this document, the terms "oxides of nitrogen" or "nitrogen oxides" refer to all forms
23 of oxidized nitrogen compounds, including nitric oxide (NO), NO2, and all other oxidized
24 nitrogen-containing compounds transformed from NO and NO2 (defined further in Chapter 2,
25 Section 2.1).
26
27
28 1.1 LEGISLATIVE REQUIREMENTS
29 Two sections of the Clean Air Act (CAA) govern the establishment and revision of the
30 NAAQS. Section 108 (U.S. Code, 2003a) directs the Administrator to identify and list "air
1 The secondary NAAQS for NO2 is being reviewed independently, in conjunction with the review of the secondary
NAAQS for sulfur dioxide (SO2). A review of the primary NAAQS for SO2 is also underway.
March 2008 1-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 pollutants" that "in his judgment, may reasonably be anticipated to endanger public health and
2 welfare" and whose "presence in the ambient air results from numerous or diverse mobile or
3 stationary sources" and to issue air quality criteria for those that are listed. Air quality criteria
4 are intended to "accurately reflect the latest scientific knowledge useful in indicating the kind
5 and extent of identifiable effects on public health or welfare which may be expected from the
6 presence of [a] pollutant in ambient air."
7 Section 109 (U.S. Code, 2003b) directs the Administrator to propose and promulgate
8 "primary" and "secondary" NAAQS for pollutants listed under Section 108. Section 109(b)(l)
9 defines a primary standard as one "the attainment and maintenance of which in the judgment of
10 the Administrator, based on such criteria and allowing an adequate margin of safety, are requisite
11 to protect the public health."2 A secondary standard, as defined in Section 109(b)(2), must
12 "specify a level of air quality the attainment and maintenance of which, in the judgment of the
13 Administrator, based on such criteria, is required to protect the public welfare from any known
14 or anticipated adverse effects associated with the presence of [the] pollutant in the ambient air."3
15 The requirement that primary standards include an adequate margin of safety was
16 intended to address uncertainties associated with inconclusive scientific and technical
17 information available at the time of standard setting. It was also intended to provide a reasonable
18 degree of protection against hazards that research has not yet identified. See Lead Industries
19 Association v. EPA, 647 F.2d 1130, 1154 (D.C. Cir 1980), cert, denied, 449 U.S. 1042 (1980);
20 American Petroleum Institute v. Costle, 665 F.2d 1176, 1186 (D.C. Cir. 1981), cert, denied, 455
21 U.S. 1034(1982). Both kinds of uncertainties are components of the risk associated with
22 pollution at levels below those at which human health effects can be said to occur with
23 reasonable scientific certainty. Thus, in selecting primary standards that include an adequate
24 margin of safely, the Administrator seeks to limit pollution levels demonstrated to be harmful as
The legislative history of Section 109 indicates that a primary standard is to be set at "the maximum permissible
ambient air level.. .which will protect the health of any [sensitive] group of the population" and that, for this
purpose, "reference should be made to a representative sample of persons comprising the sensitive group rather
than to a single person in such a group" [U.S. Senate, 1970].
3 Welfare effects as defined in Section 302(h) [U.S. Code, 2005] include, but are not limited to, "effects on soils,
water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate, damage to and
deterioration of property, and hazards to transportation, as well as effects on economic values and on personal
comfort and well-being."
March 2008 1 -2 DRAFT-DO NOT QUOTE OR CITE
-------
1 well as lower pollutant levels that may pose an unacceptable risk of harm, even if the nature or
2 degree of risk is not precisely identified.
3 In selecting a margin of safety, EPA considers such factors as the nature and severity of
4 the health effects involved, the size of sensitive population(s) at risk, and the kind and degree of
5 the uncertainties that must be addressed. The selection of any particular approach to providing
6 an adequate margin of safety is a policy choice left specifically to the Administrator's judgment.
7 See Lead Industries Association v. EPA, supra, 647 F.2d at 1161-62.
8 In setting standards that are "requisite" to protect public health and welfare, as provided
9 in Section 109(b), EPA's task is to establish standards that are neither more nor less stringent
10 than necessary for these purposes. In so doing, EPA may not consider the costs of
11 implementing the standards. See generally Whitman v. American Trucking Associations, 531
12 U.S. 457, 465-472, and 475-76 (U.S. Supreme Court, 2001).
13 Section 109(d)(l) requires that "not later than December 31, 1980, and at 5-year intervals
14 thereafter, the Administrator shall complete a thorough review of the criteria published under
15 Section 108 and the national ambient air quality standards and shall make such revisions in such
16 criteria and standards and promulgate such new standards as may be appropriate..." Section
17 109(d)(2) requires that an independent scientific review committee "shall complete a review of
18 the criteria... and the national primary and secondary ambient air quality standards... and shall
19 recommend to the Administrator any new standards and revisions of existing criteria and
20 standards as may be appropriate..." Since the early 1980s, this independent review function has
21 been performed by the Clean Air Scientific Advisory Committee (CASAC) of EPA's Science
22 Advisory Board.
23
24
25 1.2 HISTORY OF PRIMARY NO2 NAAQS REVIEWS
26 On April 30, 1971, EPA promulgated identical primary and secondary NAAQS for NC>2,
27 under Section 109 of the Act, set at 0.053 parts per million (ppm), annual average (Federal
28 Register, 1971). In 1982, EPA published Air Quality Criteria for Oxides of Nitrogen (U.S.
29 Environmental Protection Agency, 1982), which updated the scientific criteria upon which the
30 initial NC>2 standards were based. On February 23, 1984, EPA proposed to retain these standards
31 (Federal Register, 1984). After taking into account public comments, EPA published the final
32 decision to retain these standards on June 19, 1985 (Federal Register, 1985).
March 2008 1 -3 DRAFT-DO NOT QUOTE OR CITE
-------
1 On July 22, 1987, EPA announced it was undertaking plans to revise the 1982 air quality
2 criteria for oxides of nitrogen (Federal Register, 1987). In November 1991, EPA released an
3 updated draft AQCD for CASAC and public review and comment (Federal Register, 1991). The
4 draft document provided a comprehensive assessment of the available scientific and technical
5 information on health and welfare effects associated with NO2 and other oxides of nitrogen. The
6 CASAC reviewed the document and concluded in a closure letter to the Administrator that the
7 document "provides a scientifically balanced and defensible summary of current knowledge of
8 the effects of this pollutant and provides an adequate basis for EPA to make a decision as to the
9 appropriate NAAQS for NO2" (Wolff, 1993).
10 The EPA also prepared a draft Staff Paper that summarized and integrated the key studies
11 and scientific evidence contained in the revised AQCD and identified the critical elements to be
12 considered in the review of the NO2 NAAQS. The draft Staff Paper was reviewed by CASAC
13 and revised by EPA staff in response to CASAC comments and recommendations. CASAC
14 reviewed the final draft of the Staff Paper in June 1995 and responded by written closure letter
15 (Wolff, 1996). In September 1995, EPA finalized the document entitled, Review of the National
16 Ambient Air Quality Standards for Nitrogen Dioxide Assessment of Scientific and Technical
17 Information (U.S. Environmental Protection Agency, 1995).
18 Based on that review, the Administrator announced her proposed decision not to revise
19 either the primary or the secondary NAAQS for NO2 (Federal Register, 1995). The decision not
20 to revise the NO2 NAAQS was finalized after careful evaluation of the comments received on the
21 proposal (October 11, 1995). The level for both the existing primary and secondary NAAQS for
22 NO2 is 0.053 ppm annual arithmetic average, calculated as the arithmetic mean of the 1-h NO2
23 concentrations.
24
25
26 1.3 POLICY-RELEVANT QUESTIONS
27 The Integrated Plan for the Review of the Primary National Ambient Air Quality
28 Standard for Nitrogen Dioxide (U.S. Environmental Protection Agency, 2007) identifies a set of
29 key policy-relevant questions. These questions frame this review of the scientific evidence that
30 provides the scientific basis for a decision on whether the current primary NAAQS for NO2
31 (0.053 ppm, annual average) should be retained or revised. The questions are:
March 2008 1 -4 DRAFT-DO NOT QUOTE OR CITE
-------
1 • Has new information altered the scientific support for the occurrence of health effects
2 following short- and/or long-term exposure to levels of nitrogen oxides found in the
3 ambient air?
4 • What do recent studies focused on the near-roadway environment tell us about health
5 effects of nitrogen oxides?
6 • At what levels of nitrogen oxides exposure do health effects of concern occur?
7 • Has new information altered conclusions from previous reviews regarding the
8 plausibility of adverse health effects caused by exposure to nitrogen oxides?
9 • To what extent have important uncertainties identified in the last review been reduced
10 and/or have new uncertainties emerged?
11 • What are the air quality relationships between short- and long-term exposures
12 to nitrogen oxides?
13
14
15 1.4 DOCUMENT DEVELOPMENT
16 The EPA formally initiated the current review of the NC>2 NAAQS by announcing the
17 commencement of the review in the Federal Register with a call for information (Federal
18 Register, 2005). In addition to the call for information, publications are identified through an
19 ongoing literature search process. Literature search strategies include extensive computer
20 database mining on specific topics; reviewing previous EPA reports; reviewing peer reviewed
21 publications reporting results from observational studies, clinical studies, and animal studies with
22 information related to exposure-response relationships, mechanism(s) of action, or susceptible
23 subpopulations; and review of reference lists from important publications. Additional evidence
24 related to exposure is taken from published studies or EPA's analyses of air quality data and
25 emissions data and the atmospheric chemistry, transport, and fate of these emissions.
26 Information is also acquired from consultation with content and area experts and the public. The
27 search strategies used in the draft ISA development are detailed in Annex AX1. The focus of
28 this draft ISA is on literature published since the 1993 AQCD for nitrogen oxides. Key findings
29 and conclusions from the 1993 review are discussed in conjunction with recent findings.
30 Generally, only information that has undergone scientific peer review and that has been
March 2008 1 -5 DRAFT-DO NOT QUOTE OR CITE
-------
1 published (or accepted for publication) in the open literature is considered. Details of the criteria
2 for study selection for this draft ISA are found in Annex AX1.
3
4
5 1.5 DOCUMENT ORGANIZATION
6 This draft ISA includes five chapters. This introductory chapter (Chapter 1) presents an
7 overview, including the framework for the evaluation of causality used in this review. Chapter 2
8 highlights key concepts or issues relevant to understanding the sources, atmospheric chemistry,
9 exposure, and dosimetry of nitrogen oxides, following a "source-to-dose" paradigm. Chapter 3
10 evaluates and integrates health information relevant to the review of the primary NAAQS for
11 NC>2. Chapter 4 provides information relevant to the public health impact of exposure to ambient
12 nitrogen oxides, including identification of potentially susceptible population groups. Finally,
13 Chapter 5 articulates findings and conclusions regarding the health evidence and makes
14 recommendations pertinent to exposure and risk assessments.
15 In addition, a series of Annexes provides additional details of information in the ISA.
16 Annex 1 is an introduction to the Annex series, and detailed discussions of the study selection
17 process for the ISA and Annexes. Annex 2 contains evidence related to the physical and
18 chemical processes controlling the production, destruction, and levels of reactive nitrogen
19 compounds in the atmosphere, including both oxidized and reduced species. Annex 3 presents
20 information on environmental concentrations, patterns, and human exposure to ambient nitrogen
21 oxides. Annex 4 presents results from toxicological studies as well as information on dosimetry
22 of nitrogen oxides. Annex 5 presents results from controlled human exposure studies, and
23 Annex 6 presents evidence from epidemiologic studies. Annex tables for health studies are
24 generally organized to include information about (1) concentrations of nitrogen oxides levels or
25 doses and exposure times, (2) description of study methods employed, (3) results and comments,
26 and (4) quantitative outcomes for nitrogen oxides measures.
27
28
29 1.6 EPA FRAMEWORK FOR CAUSAL DETERMINATIONS
30 It is important to have a consistent and transparent basis for the critical decisions on the
31 causal nature of air pollution induced health effects. The framework described below establishes
32 uniform language concerning causality and brings more specificity to the findings. It draws
March 2008 1 -6 DRAFT-DO NOT QUOTE OR CITE
-------
1 normalizing language from across the Federal government and wider scientific community,
2 especially from the recent National Academy of Sciences (NAS) Institute of Medicine (IOM)
3 document, Improving the Presumptive Disability Decision-Making Process for Veterans (IOM,
4 2007), the most recent comprehensive work on evaluating the causality of health effects. This
5 section:
6 • describes the kinds of scientific evidence used in establishing a general causal
7 relationship between exposure and health effects,
8 • defines cause in contrast to statistical association,
9 • discusses the sources of evidence necessary to reach a conclusion about the existence
10 of a causal relationship,
11 • highlights the issue of multifactorial causation,
12 • identifies issues and approaches related to uncertainty, and
13 • provides a framework for classifying and characterizing the weight of evidence in
14 support of a general causal relationship.
15 Approaches to assessing the separate and combined lines of evidence from epidemiology,
16 controlled human exposure studies, animal toxicology, and in vitro studies have been formulated
17 by a number of regulatory and science agencies, including the Institute of Medicine of the
18 National Academies of Science (IOM, 2007), the International Agency for Research on Cancer
19 (IARC, 2006), the National Toxicology Programs (NTP, 2005), the EPA (U.S. Environmental
20 Protection Agency, 2005), the Centers for Disease Control and Prevention (CDC, 2004), and the
21 National Acid Precipitation Assessment Program (NAPAP, 1991). Highlights or excerpts from
22 the various decision framework documents are included in Annex AX1. These formalized
23 approaches offer guidance for assessing the relative weights of those lines of evidence. The
24 frameworks are similar in nature, although adapted to different purposes, and have proven to be
25 effective in providing a uniform structure and language for causal determinations. Moreover,
26 these frameworks must support decision-making under conditions of great uncertainty.
27
28 Scientific Evidence Used in Establishing Causality
29 The most compelling evidence of a causal relationship between pollutant exposures and
30 human health effects comes from controlled human exposure (i.e., clinical) studies. This type of
March 2008 1 -7 DRAFT-DO NOT QUOTE OR CITE
-------
1 study experimentally evaluates the effects of administered exposures under highly controlled
2 laboratory conditions.
3 In observational, or epidemiologic studies of humans, the investigator does not control
4 exposures or intervene with the study population. Broadly speaking, observational studies can
5 describe associations between exposures and effects and fall into several categories: cross-
6 sectional, prospective cohort, time-series, and panel studies. "Natural experiments" occur
7 occasionally in epidemiology; these include comparisons of epidemiologic results before and
8 after a change in population exposures (e.g., closure of a pollution source).
9 The clinical and observational data are complemented by experimental animal data,
10 which can support the biological plausibility of causation. In the absence of clinical or
11 observational data, animal data alone may be sufficient to support a likely causal determination,
12 assuming that humans respond similarly to the experimental species.
13
14 Association and Causation
15 Association and causation are not the same. The word cause conveys the notion of a
16 significant, effectual relationship between an agent and an associated disorder or disease in the
17 population. In contrast, association is the statistical dependence between two or more events,
18 characteristics, or other variables. An association is prima facie evidence for causation, but not
19 sufficient by itself for proving a causal relationship between exposure and disease. Unlike
20 associations, causal claims support making counterfactual claims; that is, claims about what the
21 world would have been like under different or changed circumstances (IOM, 2007). Currently,
22 much of the newly available health information evaluated in the draft ISA comes from
23 epidemiologic studies that report a statistical association between exposure and health outcome.
24 It would be naive to insist upon mono-etiology in pathological processes or in vital
25 phenomena. Epidemiologists have long recognized that most chronic diseases (e.g., cancer or
26 coronary heart disease) result from a complex web of causation, whereby one or more external
27 agents (exposures) taken into the body initiate a disease process, the outcome of which could
28 depend upon many factors including age, genetic susceptibility, nutritional status, immune
29 competence, social factors, and others (IOM, 2007; Gee and Payne Sturges, 2004). Figure 1.6-1
30 shows a diagram of a variety of etiologic factors that contribute to disease.
March 2008 1 -8 DRAFT-DO NOT QUOTE OR CITE
-------
(The combined risks
Race/Ethnicity *•
^
1 * — *s"^ 1 —
Cumulative Risks
from aggregate exposures to multiple agents or Stressors)
Residential Location
^ 1 \\
^ —
Neighborhood « » Community 4
Resources Stressors
^
\
-^s 1
+ Structural
Factors
z
Community «- — "
Stress
j
i
Stress/Coping, ^^.
Life Stage/Style • —
+
Individual Stressors
t-
Modified from Gee & Payne-Sturges, 2004
-• —
5>I1
^ ^ Enviror
Haza
Pollu
~~~~^
mental
rds &
tants
Exposure
__^^-— — """"^
Internal dose
r
i —
Biologically
effective dose
i
Health effects/
disparities
Community
Level
Vulnerability
Individual
Level
Vulnerability
Figure 1.6-1. Exposure-disease-stress model for environmental health disparities.
Source: Modified from Gee and Payne-Sturges (2004).
1 Additionally, various exposure profiles can be important. Exposures may occur over an
2 extended period of time with some cumulative effect; repetitive acute exposures may produce
3 both episodic and chronic illness; exposure to multiple agents together could result in synergistic
4 or antagonistic effects different from what might result from exposure to each separately.4 The
5 end results are the net effect of many actions and counteractions. Epidemiologists use the term
6 interaction (or effect modification) to denote the departure of the observed joint risk from what
7 might be expected based on the separate effects of the factors.
8
For example, one could define a multiplicative interaction relative risk (RR) as: RRint(muit) = RRjomt/RRs x RRS, or
an additive interaction RR as RRin^add)= RRjoint ~ RRs ~ RRs + 1.
March 2008
1-9
DRAFT-DO NOT QUOTE OR CITE
-------
1 Evidence for Going Beyond Association to Causation
2 Developing evidence for going beyond association to causation involves experimental
3 control, statistical control, and models. Controlled human exposure studies are experiments in
4 which subjects in a population are randomly allocated into groups, usually called study and
5 control groups, and exposed to a pollutant or a sham. The results are assessed by rigorous
6 comparison of rates of appropriate outcome between the study and control groups. Randomized
7 controlled human exposure studies are generally regarded as the most scientifically rigorous
8 method of hypothesis testing available. By assigning exposure randomly, the study design
9 attempts to remove the effect of any factor that might influence exposure, and any possible effect
10 of the outcome on exposure. Done properly, and setting aside randomness, only a causal
11 relationship from exposure to health outcome should produce observed associations in
12 randomized clinical trials. In another type of controlled human exposure study, the same subject
13 is exposed to a pollutant and a sham at different time points and the responses to the two types of
14 exposures are compared. This study design is also effective at controlling for any potential
15 confounders since the subject is serving as his/her own control. A lack of observation of effects
16 from controlled human exposure studies does not mean that a causal relationship does not occur.
17 Controlled human exposure studies are often limited because the study population is generally
18 small. This restricts the power to discern statistically significant findings. In addition, the most
19 susceptible individuals may be explicitly excluded (for ethical reasons), and more susceptible
20 individuals or groups (e.g., those with nutritional deficits) may not be included.
21 Inferring causation from observational (epidemiologic) studies requires consideration of
22 potential confounders. When associations are found in observational studies, the first approach
23 for removing spurious associations from possible confounders is statistical control of the
24 difference between characteristics of exposed and unexposed persons, frequently termed
25 adjustment. Multivariable regression models constitute one tool for estimating the association
26 between exposure and outcome after adjusting for characteristics of participants that might
27 confound the results. Another way to adjust for potential confounding is through stratified
28 analysis, i.e., examining the association within homogeneous groups with the confounding
29 variable. Stratified analyses have the secondary benefit of allowing examination of effect
30 modification through comparison of the effect estimates across different groups. If investigators
31 have successfully measured characteristics that distort the results, then adjustment of these
March 2008 1-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 factors will help separate a spurious from a true causal association. Appropriate statistical
2 adjustment for confounders requires identifying and measuring all reasonably expected
3 confounders. Deciding which variables to control for in a statistical analysis of the association
4 between exposure and disease depends upon knowledge about the possible mechanisms
5 connecting them. Identifying mechanisms allows us to identify and control for potential sources
6 of spurious association.
7 Measurement error is another problem when adjusting for spurious associations. There
8 are several components to exposure measurement error in epidemiologic studies, including the
9 use of average population exposure rather than individual exposure estimates, the difference
10 between average personal exposure to ambient pollutants and ambient concentrations at central
11 monitoring sites, and the difference between true and measured ambient concentrations. In
12 multivariate analyses, the effects of a well-measured covariate may be overestimated in
13 comparison to a more poorly measured covariate.
14 It is important to recognize the difficulties of identifying and measuring all potential
15 confounders. However, if observational studies are repeated in different settings with different
16 subjects having different eligibility criteria and/or different exposure opportunities, each of
17 which might eliminate another source of confounding from consideration, then confidence that
18 unmeasured confounders are not producing the findings is increased. The number and degree of
19 diversity of such studies as well as their interpretation for relevance to the potential confounders
20 remain matters of scientific judgment. Multicity studies use a consistent method to analyze data
21 from across locations with different levels of covariates and, thus, can provide insights on
22 potential confounding in associations.
23 In addition to clinical and epidemiologic studies, the tools of experimental biology have
24 been extraordinarily valuable for developing insights into human physiology and pathology.
25 Such laboratory tools have been extended to explore the effects of putative toxicants on human
26 health, especially through the study of model systems in other species. Background knowledge
27 about the biological mechanisms by which an exposure might or might not cause disease can
28 prove crucial in establishing, or negating, a causal claim. At the same time, species can differ in
29 fundamental aspects of physiology and anatomy (e.g., metabolism, airway branching, hormonal
30 regulation) that may limit extrapolation from one species to another. Testable hypotheses about
31 the causal nature of the proposed mechanisms or modes of action are central to utilizing
March 2008 1-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 experimental data in causal deteminations. Principles for evaluating mechanisms or modes of
2 action as part of causal determinations should be developed.
3
4 Multifactorial Causation
5 Scientific judgments are needed regarding the likely sources and magnitude of
6 confounding, together with judgments about how well the existing constellation of study designs,
7 results, and analyses address this potential threat to inferential validity. One key consideration in
8 this review is evaluation of the potential contribution to health effects of NO2 when it is a
9 component of a complex air pollutant mixture. There are multiple ways by which NO2 might
10 cause or be associated with adverse health effects: (1) as a direct causal effect, (2) as an indirect
11 causal effect mediated by other pollutants formed in the atmosphere including paniculate matter
12 (PM) and ozone (63), and (3) by acting as a surrogate for emissions from the same sources that
13 emit NO2 that are actually responsible for the adverse health effects observed; these relationships
14 are illustrated in Figure 1.6-2. Moreover, these possibilities are not necessarily exclusive.
15 Confounding, as usually defined, would refer to the production of an association between NO2
16 and adverse health effects, by the actions of one or more other exposures, themselves associated
17 with NO2 in a particular study. Multivariate models are the most widely used strategy to address
18 confounding in epidemiologic studies, but such models are not readily interpreted when the
19 potential confounders such as PM may be mediating effects possibly attributable to NO2.
20
21 Uncertainty
22 The science of estimating the causal influence of an exposure on disease is uncertain.
23 Formal statistical descriptions provide one means for dealing with uncertainty; however, they do
24 this in two distinct ways:
25 • Model uncertainty—uncertainty regarding gaps in scientific theory required to make
26 predictions on the basis of causal inferences and
27 • Parameter uncertainly—uncertainty as to the statistical estimates within each model.
28 The uncertainty concerning the correct causal model involves uncertainty about (1)
29 whether exposure causes the health outcome, (2) the set of confounders associated with exposure
30 and disease, (3) which parametric forms for describing the relations of exposure and confounders
31 with outcome are correct, and (4) whether other forms of bias could be affecting the evidence.
March 2008 1-12 DRAFT-DO NOT QUOTE OR CITE
-------
Direct Causal Effect
NOV
>t
Risk for outcome
Mediated Effect
PM
t
Risk for outcome
Source
Surrogate
NOy
Other
Pollutants
Risk for outcome
Confoyndjng
NOV
Confounder
Risk for outcome
Figure 1.6-2. Potential relationships of NOx with adverse health effects.
1 Uncertainty about the model is not limited to the qualitative causal structure: it also
2 involves uncertainty about the parametric form of the model specified, the variables included,
3 whether or not measurement error is modeled, and so on. When mechanistic knowledge exists,
4 this sort of uncertainty can be reduced. Nevertheless, model uncertainty is perhaps the more
5 important source of uncertainty. In contrast, uncertainty about the parameter estimates
6 (regression coefficients) for a given model is a well-studied problem. The important point is that
7 these reports of uncertainty are conditional on the model providing a sufficiently adequate
8 approximation of reality so that inferences are valid. The overall scientific inference involves
March 2008
1-13
DRAFT-DO NOT QUOTE OR CITE
-------
1 evaluating uncertainty about the model and uncertainty about the parameter estimates given each
2 model.
3 There are systematic, quantitative approaches for including uncertainty about the model
4 in an assessment of overall uncertainty about a causal inference. These approaches include
5 sensitivity analysis and model averaging. Sensitivity analysis attempts to quantify the sensitivity
6 of the parameter estimate to assumptions about the model. Uncertainty ranges can be estimated
7 using classical analysis (Robinson, 1989) or the Monte Carlo technique (Eggleston, 1993).
8 Model averaging attempts to provide an overall uncertainty to the estimate by calculating the
9 estimate of a common parameter or target and its uncertainty for each model considered to be
10 plausible, and weighting the estimates and the uncertainties by the likelihood of each model.
11
12 Application of Framework
13 In EPA's framework for evaluation, a two-step approach is used to judge the scientific
14 evidence about exposure to criteria pollutants and risks to public health. The first step is to
15 determine the weight of evidence in support of causation and characterize the strength of any
16 resulting causal classification. The second step includes further evaluation of the quantitative
17 evidence regarding the shape of concentration-response or dose-response relationships and the
18 levels at which effects are observed.
19 To aid judgment, decisive factors for the determination of a cause have been proposed by
20 many philosophers and scientists. The most widely cited decisive factors in epidemiology and
21 public health more generally were set forth by Sir Austin Bradford Hill in 1965. The nine "Hill
22 criteria" were also incorporated in the EPA Guidelines for Carcinogen Risk Assessment (U.S.
23 Environmental Protection Agency, 2005). These nine decisive factors for determination of
24 causality are described in Table 1.6-15 (adapted from Hill, 1965, and U.S. Environmental
25 Protection Agency, 2005). A number of these decisive factors are judged to be particularly
26 salient in evaluating the body of evidence available in this review, including the factors
27 described by Hill as strength, experiment, consistency, plausibility, and coherence. Other factors
28 identified by Hill, including temporality and biological gradient, are also relevant and considered
29 here (e.g., in characterizing lag structures and concentration-response relationships).
1 We have chosen to use the words "decisive factors" in this document, as opposed to the commonly used term
"criteria," in order to avoid confusion with criteria as characterized by the Clean Air Act.
March 2008 1-14 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 1.6-1. DECISIVE FACTORS TO AID IN JUDGING CAUSALITY
1. Consistency of the observed association. An inference of causality is strengthened when a pattern of
elevated risks is observed across several independent studies. The reproducibility of findings constitutes
one of the strongest arguments for causality. If there are discordant results among investigations, possible
reasons such as differences in exposure, confounding factors, and the power of the study are considered.
2. Strength of the observed association. The finding of large, precise risks increases confidence that the
association is not likely due to chance, bias, or other factors. A modest risk, however, does not preclude a
causal association and may reflect a lower level of exposure, an agent of lower potency, or a common
disease with a high background level.
3. Specificity of the observed association. As originally intended, this refers to increased inference of
causality if one cause is associated with a single effect or disease (Hill, 1965). Based on our current
understanding this is now considered one of the weaker guidelines for causality; for example, many
agents cause cancer at multiple sites and that many cancers have multiple causes. Thus, although the
presence of specificity may support causality, its absence does not exclude it.
4. Temporal relationship of the observed association. A causal interpretation is strengthened when
exposure is known to precede development of the disease.
5. Biological gradient (exposure-response relationship). A clear exposure-response relationship (e.g.,
increasing effects associated with greater exposure) strongly suggests cause and effect, especially when
such relationships are also observed for duration of exposure (e.g., increasing effects observed following
longer exposure times). There are many possible reasons that an epidemiologic study may fail to detect
an exposure-response relationship. Thus, the absence of an exposure-response relationship does not
exclude a causal relationship.
6. Biological plausibility. An inference of causality tends to be strengthened by consistency with data from
experimental studies or other sources demonstrating plausible biological mechanisms. A lack of
mechanistic data, however, is not a reason to reject causality.
7. Coherence. An inference of causality may be strengthened by other lines of evidence that support a
cause-and-effect interpretation of the association. Information is considered from animal bioassays,
toxicokinetic studies, and short-term studies. The absence of other lines of evidence, however, is not a
reason to reject causality.
8. Experimental evidence (from human populations). Experimental evidence is generally available from
human populations for the criteria pollutants. The strongest evidence for causality can be provided when
a change in exposure brings about a change in adverse health effect or disease frequency in either clinical
or observational studies.
9. Analogy. Structure activity relationships (SARs) and information on the agent's structural analogs can
provide insight into whether an association is causal. Similarly, information on mode of action for a
chemical, as one of many structural analogs, can inform decisions regarding likely causality.
1 While these decisive factors frame considerations weighed in assessing the evidence, they
2 do not lend themselves to being considered in terms of simple formulas or hard-and-fast rules of
3 evidence leading to conclusions about causality (Hill, 1965). For example, one cannot simply
4 count the number of studies reporting statistically significant results or statistically
5 nonsignificant results for health effects and reach credible conclusions about the relative weight
6 of the evidence and the likelihood of causality. Rather, these important considerations are taken
March 2008 1-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 into account throughout the assessment with the goal of producing an objective appraisal of the
2 evidence (informed by peer and public comment and advice), which includes the weighing of
3 alternative views on controversial issues. Additionally, it is important to note that principles
4 listed in Table 1.6-1 cannot be used as a strictly quantitative checklist. Rather, these principles
5 should be used to determine the weight of the evidence for inferring causality. In particular, the
6 absence of one or more of the principles does not automatically exclude a study from
7 consideration (e.g., see discussion in CDC, 2004).
8
9 First Step—Determination of Causality
10 This draft ISA uses a five-level hierarchy that classifies the weight of evidence for
11 causation, not just association; that is, whether the weight of scientific evidence makes causation
12 at least as likely as not in the judgment of the reviewing group.6 In developing this hierarchy,
13 EPA has drawn upon the work of previous evaluations, most prominently the lOM's Improving
14 the Presumptive Disability Decision-Making Process for Veterans (2007), EPA's Guidelines for
15 Carcinogen Risk Assessment (U.S. Environmental Protection Agency, 2005), and the U.S.
16 Surgeon General's smoking reports (CDC, 2004). These efforts are presented in more detail in
17 Annex AX1. In the draft ISA, causality of association was placed into one of five categories
18 with regard to the weight of the evidence. These conclusions are based on EPA's evaluation of
19 the weight of evidence from epidemiologic studies, animal studies, or other mechanistic,
20 lexicological, or biological sources. These separate judgments are integrated into a qualitative
21 statement about the overall weight of the evidence and causality. The five descriptors are:
22 • Sufficient to infer a causal relationship,
23 • Sufficient to infer a likely causal relationship (i.e., more likely than not),
24 • Suggestive but not sufficient to infer a causal relationship,
25 • Inadequate to infer the presence or absence of a causal relationship, and
26 • Suggestive of no causal relationship.
27
6 It should be noted that the CDC and IOM frameworks use a four-category hierarchy for the strength of the
evidence. A five-level hierarchy is used here to be consistent with the EPA Guidelines for Carcinogen Risk
Assessment (U.S. Environmental Protection Agency, 2005).
March 2008 1-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 Second Step—Evaluation of Population Response
2 Beyond judgments regarding causality are questions relevant to characterizing exposure
3 and risk to populations. Such questions include:
4 • At what doses or concentrations are effects observed?
5 • What is the shape of the concentration-response or dose-response relationship?
6 • What population groups appear to be affected or more susceptible to effects?
7 • With what exposure time periods (e.g., peak, long-term average) are effects seen?
8 On the population level, causal and likely causal claims typically proceed to characterize how
9 risk (the probability of health effects) changes in response to exposure. Initially, the response is
10 evaluated within the range of observation. Approaches to analysis of the range of observation of
11 epidemiologic and clinical studies are determined by the type of study and how dose and
12 response are measured in the study. Extensive human data for concentration-response analyses
13 exists for all criteria pollutants, unlike most other environmental pollutants. Animal data also
14 can inform concentration-response, particularly relative to dosimetry, mechanisms of action, and
15 characteristics of sensitive subpopulations.
16 An important consideration in characterizing the public health impacts associated with
17 pollutant exposure is whether the concentration-response relationship is linear across the full
18 concentration range encountered or if nonlinear departures exist along any part of this range. Of
19 particular interest is the shape of the concentration-response curve at and below the level of the
20 current standards. The complex molecular and cellular events that underlie cancer and
21 noncancer toxicity are likely to be both linear and dose-transitional. At the human population
22 level, however, various sources of both variability and uncertainty tend to smooth and "linearize"
23 the concentration-response function, obscuring any thresholds that may exist. (This does not
24 presume that the dose-response relationship will be linear for individuals.) There are limitations
25 to identifying possible "thresholds" in epidemiologic studies, including difficulties related to the
26 low data density in the lower concentration range, possible influence of measurement error, and
27 individual differences in susceptibility to air pollution health effects. These attributes of
28 population dose-response may explain why the available human data at ambient concentrations
29 for some environmental pollutants (e.g., PM, secondhand tobacco smoke, lead, radiation) do not
30 exhibit evident thresholds for cancer or noncancer health effects even though likely mechanisms
31 of action include nonlinear processes for some key events. These attributes of human population
March 2008 1-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 dose-response relationships have been extensively discussed in the broader epidemiologic
2 literature (e.g., Rothman and Greenland, 1998).
3
4
5 1.7 CONCLUSIONS
6 The scientific assessment of air pollution-related health effects involves reviewing
7 evidence from clinical, epidemiologic, and animal studies, including mechanistic evidence from
8 basic biological science. Clinical studies can provide the strongest evidence for causation.
9 Epidemiologic studies that are reasonably free of bias and confounding provide evidence that can
10 support determination of causation, but may not provide proof of causation. Mechanistic
11 knowledge of how particular agents might produce adverse health effects provides further
12 evidence. For example, animal mechanism-of-action studies may provide further evidence by
13 showing that an agent may induce the same effect as observed in human studies, using a
14 mechanism that is conserved across species with key features of the mechanism observed.
15 Uncertainty surrounding a causal claim can arise because of uncertainty about which among a set
16 of plausible models is correct, uncertainty about study design and execution, uncertainty caused
17 by simple sampling variability, or uncertainty in the basic science required to analyze other
18 evidence. The overall uncertainty is some combination of all of these uncertainties.
19 The draft Integrated Science Assessment (ISA) presents a concise review, synthesis, and
20 evaluation of the most policy-relevant science, and communicates critical science judgments
21 relevant to the NAAQS review. Those judgments include determinations of causality. The draft
22 ISA relies on widely accepted principles for determinations of causality based on decisive factors
23 such as those put forth by Hill in 1965 and, subsequently, generally adopted by numerous
24 agencies. Inferences, whether about causality or statistical associations, always carry some
25 degree of uncertainty.
26 The draft ISA uses standardized language to express the evaluation of the evidence
27 bearing on causality. This approach helps clarify the assessment and makes it possible for
28 subsequent groups to measure progress by comparing their judgments with those expressed here.
29 This structure also encourages the description of the sources of uncertainty in the evidence,
30 which hopefully will stimulate necessary research. The framework used in this report should
31 assist EPA and others, now and in the future, to accurately represent what is presently known and
March 2008 1-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 what remains unknown concerning the effects of these environmental air pollutants on human
2 health.
March 2008 1-19 DRAFT-DO NOT QUOTE OR CITE
-------
i 2. SOURCE TO TISSUE DOSE
2
3
4 This chapter provides concepts and findings relating to emissions sources, atmospheric
5 science, human exposure assessment, and human dosimetry. The order of these topics
6 essentially follows that given in the National Research Council paradigm for integrating air
7 pollutant research (National Research Council, 2004) as shown in Figure 2-1. This chapter is
8 meant to serve as a prologue for detailed discussions on the evidence on health effects that
9 follow in Chapters 3 and 4.
Sources
of Airborne
Oxides
of Nitrogen
Emissions
Indicator
in Ambient
(Outdoor) Air
Personal
Exposure
Dose to
Target
Tissues
Human
Health
Response
Mechanisms determining Human time-activity Deposition, clearance, Mechanisms
emissions, chemical patterns, indoor retention, and of damage and repair
transformation, and (or microcenvironmental) disposition of oxides
transport in air sources, and sinks of of nitrogen presented
oxides of nitrogen to an individual
Figure 2-1. A generalized conceptual model for integrating research on oxides of
nitrogen pollution and human health effects.
Source: Adapted from National Research Council (2004).
10 2.1 INTRODUCTION
11 As noted in Chapter 1, the definition of "nitrogen oxides" as it appears in the enabling
12 legislation related to the national ambient air quality standard (NAAQS) differs from the one
13 commonly used in the air pollution research and control communities. In this document, the
14 terms "oxides of nitrogen" and "nitrogen oxides" (NOx) refer to all forms of oxidized nitrogen
15 (N) compounds, including nitric oxide (NO), nitrogen dioxide (NOz), and all other oxidized
March 2008
2-1
DRAFT-DO NOT QUOTE OR CITE
-------
1 N-containing compounds formed from NO and NOz.1 In the Federal Register Notice for the
2 previous Air Quality Criteria Document (AQCD) for Oxides of Nitrogen (Federal Register,
3 1995), the term "nitrogen oxides" was used to "describe the sum of NO, N02, and other oxides
4 of nitrogen."
5 NO and NOz, along with volatile organic compounds (VOCs; anthropogenic and biogenic
6 hydrocarbons, aldehydes, etc.) and carbon monoxide (CO), are precursors in the formation of
7 ozone (Os) and photochemical smog. N02 is an oxidant and can react to form other
8 photochemical oxidants, including organic nitrates (RONOz) like the peroxyacyl nitrates (PANs).
9 N02 can also react with toxic compounds such as polycyclic aromatic hydrocarbons (PAHs) to
10 form nitro-PAHs, some of which are more toxic than either reactant alone. NOz and sulfur
11 dioxide (802), another U.S. Environmental Protection Agency (EPA) criteria air pollutant, can
12 also be oxidized to form the strong mineral acids nitric acid (HNOs) and sulfuric acid (F^SO^,
13 respectively, thereby contributing to the acidity of cloud-, fog-, and rainwater and of ambient
14 particles.
15
16
17 2.2 SOURCES AND ATMOSPHERIC CHEMISTRY
18 The role of NOx in Os formation was reviewed in Chapter 2 (Section 2.2) of the latest Air
19 Quality Criteria for Ozone and Related Photochemical Oxidants (2006 AQCD for Os; U.S.
20 Environmental Protection Agency, 2006) and has been presented in numerous texts (see, e.g.,
21 Seinfeld and Pandis, 1998; Jacob, 1999; Jacobson, 2002). Mechanisms for transporting Os
22 precursors including NOx, the factors controlling the efficiency of Os production from NOx,
23 methods for calculating Os from its precursors, and methods for measuring total oxidized
24 nitrogen (NOy) were all reviewed in Section 2.6 of 2006 AQCD for Os. The main points from
25 that 2006 AQCD for 03 will be presented here along with updates based on new material. The
26 overall chemistry of reactive, oxidized N compounds in the atmosphere is summarized in Figure
27 AX2.2-1 and described in greater detail in Annex AX2.
1 This follows usage in the Clean Air Act Section 108(c): "Such criteria [for oxides of nitrogen] shall include a
discussion of nitric and nitrous acids, nitrites, nitrates, nitrosamines, and other carcinogenic and potentially
carcinogenic derivatives of oxides of nitrogen." By contrast, within the air pollution research and control
communities, the terms "oxides of nitrogen" and "nitrogen oxides" are restricted to refer only to the sum of NO
and N02, and this sum is commonly abbreviated as NOX. The category label used by this community for the sum
of all forms of oxidized nitrogen compounds including those listed in Section 108(c) is NOY.
March 2008 2-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.2.1 Sources of NOX
2 Both anthropogenic and natural (biogenic) processes emit NOx. NOx is emitted by
3 combustion sources mainly as NO with smaller quantities of N02. The major combustion
4 sources of NOx in the United States are listed in Annex Table AX 2-3. Figure 2.2-1 shows
5 schematically the on-road motor vehicles and electric utilities sources, the two largest NOx
6 sources in the United States, along with NOx species and some reaction pathways. Stationary
7 engines, off-road vehicles, and industrial facilities also emit NOx, but because they are fewer in
8 number or burn less fuel, their mass contribution is relatively smaller. The ratios of N02 to NOx
9 in emissions are variable with typical values being less than 0.1. However, ratios in emissions
10 from retrofitted diesel engines range from 0.3 to 0.6 as shown in a study of public transit buses in
11 New York City (Shorter et al., 2005). Sources of NOx are distributed across various heights,
12 some are at or near ground level (e.g., motor vehicles) and others aloft (e.g., electric utilities
13 stacks), as indicated in Figure 2.2-1. Because the prevailing winds aloft are generally stronger
14 than those at the surface, emissions from elevated sources can be distributed over a wider area
15 than those emitted at the surface.
16 Biomass burning also produces NOx. Apart from these anthropogenic sources, there are
17 also smaller natural sources which include microbial activity in soils (particularly fertilized soils)
18 and lightning. Wildfires can be large but epidosic and highly variable sources of NOx. NOx
19 sources and emissions are described in greater detail in Annex Section AX2.6.
20
21 2.2.2 Chemical Transformations of NOX
22 NO and N02 are often grouped together and given the category label "NOx" because they
23 are emitted together and can rapidly interconvert as shown in the inner box in Figure 2.2-1. N02
24 reacts with 03 and various free radicals in the gas phase and on surfaces in multiphase processes
25 to form the oxidation products shown in Figure 2.2-1. These products include inorganic species
26 (shown on the left side of the outer box in Figure 2.2-1) and organic species (shown on the right
27 side of the outer box in Figure 2.2-1). The oxidized N species in the outer box are often
28 collectively termed NOZ; thus, NOX + NOZ = NOY.
29 The concentrations and atmospheric lifetimes (T) of inorganic and organic products from
30 reactions of NOx vary widely in space and time. Inorganic reaction products include nitrous acid
31 (HONO), HN03, pernitric acid (HN04), and particulate nitrate (pN03~). While a broad range of
March 2008 2-3 DRAFT-DO NOT QUOTE OR CITE
-------
Long range transport to remote
regions at low temperatures
NH
nitro-PAHs
NO, R-C=C-% RONO
nitrosamines.
nitro-phenols, etc.
RO,
—*•—^- HU • =—»-RONO
'..... MO ••' 2
I f0-X -I"
i \oo& ?
deposition
deposition
emissions
Figure 2.2-1. Schematic diagram of the cycle of reactive, oxidized N species in the
atmosphere. IN refers to inorganic particulate species (e.g., sodium [Na+],
calcium [Ca++]), MPP to multiphase processes, hv to a solar photon and R to
an organic radical. Particle-phase RONO2 are formed from the species
shown on the right side.
1 organic N compounds are emitted by combustion sources (e.g., nitrosamines and nitro-PAHs),
2 they are also formed in the atmosphere from reactions of NO and NOz. These include PANs and
3 isoprene nitrates, other nitro-PAHs, and the more recently identified nitrated organic compounds
4 in the quinone family. Most of the mass of products shown in the outer box of Figure 2.2-1 is in
5 the form of peroxyacetyl nitrate (PAN) and HNOs, although other organic nitrates, e.g., isoprene
6 nitrates and specific biogenic PANs can be important at locations closer to biogenic sources
7 (Horowitz et al., 2007; Singh et al., 2007).
8 In addition to gas-phase reactions, reactions occurring on surfaces or occurring in
9 multiple phases (MPP) are important for the formation of HONO and pNCV. These reactions
10 can occur on the surfaces of suspended particles, soil, and buildings, and within aqueous media.
11 The T of PAN is strongly temperature dependent and is long enough at low temperatures so that
March 2008
2-4
DRAFT-DO NOT QUOTE OR CITE
-------
1 PAN can be transported tens or hundreds of kilometers (depending on meteorological conditions)
2 before decomposing to release N02, which can then participate in 03 formation in these regions
3 which are remote from the original NOX source. HN03 can act similarly to some extent, but its
4 high solubility and high deposition rate imply that it is removed from the gas phase faster than
5 PAN, and thus would not be as important as a source of NOx in remote regions. Characteristic
6 concentrations of many of the NOx species are given in Annex AX3.2.
7 The timescale for reactions of NOx to form NOz products like PAN and HNOs typically
8 ranges from a few hours during summer to about a day during winter. As a result, morning rush
9 hour emissions of NOx from motor vehicles can be converted almost completely to NOz
10 products by late afternoon during warm, sunny conditions. Because the time required for mixing
11 emissions down to the surface is similar to or longer than the time for oxidation of NOx,
12 emissions of NOx from elevated sources like the stacks of electric utilities tend to be transformed
13 to NOz before they reach the surface. However, people live closer to emissions from on-road
14 and off-road motor vehicles fixed-site combustion engines (e.g., generators), and indoor sources,
15 and so are more likely to be exposed to NO and NOz from these sources. Hence, because
16 atmospheric dispersion and chemical reactions in this way determine the partitioning of a
17 person's exposure to N02 and its reaction products from multiple different sources, a person's
18 total exposure to NOx cannot be judged solely by the NO and N02 source strengths given in the
19 national emissions inventories (NEI).
20 Ultimately, oxidized N compounds are lost from the atmosphere by deposition to the
21 earth's surface. Soluble species are taken up by aqueous aerosols and cloud droplets that can
22 then be removed by either wet or dry deposition. Insoluble species are lost by dry deposition and
23 washout. Discussions of the reactions in particles are beyond the scope of this review, but once
24 in particles, a variety of organic and inorganic nitrates can be formed, which are then removed
25 either by dry deposition to the surface or by rainout or washout.
26
27 2.2.2.1 Formation of Nitro-PAHs
28 Nitro-PAHs are produced either by direct emissions or by atmospheric reactions. Among
29 combustion sources, diesel emissions have been identified as the major source of nitro-PAHs in
30 ambient air (Bezabeh et al., 2003; Gibson, 1983; Schuetzle, 1983; Tokiwa and Ohnishi, 1986).
31 Direct emissions of nitro-PAHs vary with fuel type, vehicle maintenance, and ambient conditions
32 (Zielinska et al., 2004). In addition to direct emission, nitro-PAHs are formed from both gaseous
March 2008 2-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 and heterogeneous reactions of PAHs with gaseous N-containing pollutants in the atmosphere;
2 reactions of hydroxyl (OH) and nitrate (NOs) radicals with PAHs are the major sources of nitro
3 PAHs, (Arey et al., 1986, 1989, 1998; Perrini, 2005; Pitts, 1987; Sasaki et al, 1997; Zielinska
4 et al., 1989; Bamford and Baker, 2003; Reisen and Arey, 2005, and references therein).
5 Reactions involving OH radicals occur mainly during the day, while reactions with NOs radicals
6 occur mainly during the night. The major loss process of nitro-PAHs is photodecomposition
7 (Fan et al., 1996; Feilberg et al., 1999; Feilberg and Nielsen, 2001) with lifetimes on the order of
8 hours, followed by reactions with OH and NOs radicals. The reaction mechanisms for forming
9 and destroying nitro-PAHs in the atmosphere are described in Annex AX2.2.3.
10 In ambient particulate organic matter (POM), 2-nitrofluoranthene (2NF) is the dominant
11 compound, followed by 1-nitropyrene (1NP) and 2-nitropyrene (2NP) (Arey et al., 1989;
12 Bamford et al., 2003; Reisen and Arey, 2005; Zielinska et al., 1989). 2NF and 2NP are not
13 directly emitted from primary combustion emissions, but are formed in the atmosphere. 1NP is
14 generally regarded as a tracer of primary combustion sources, in particular, diesel exhaust. After
15 formation, nitro-PAHs with low vapor pressures (such as 2NF and 2NP) immediately migrate to
16 particles under ambient conditions (Fan et al., 1995; Feilberg et al., 1999). More volatile nitro-
17 PAHs, such as nitronapthalene (NN), remain mainly in the gas phase.
18 The concentrations for most nitro-PAHs found in ambient air are typically lower than
19 1 pg/m3, except NNs, 1NP, and 2NF, which can be present at levels up to several tens or
20 hundreds of pg/m3. These levels are from ~2 to -1000 times lower than those of their parent
21 PAHs. However, nitro-PAHs are much more toxic than PAHs (Durant et al., 1996; Grosovsky
22 et al., 1999; Salmeen et al., 1982; Tokiwa et al., 1998; Tokiwa and Ohnishi, 1986). Moreover,
23 most nitro-PAHs are present in particles with a mass median diameter of <0.1 pm.
24
25 2.2.3 O3 Formation
26 As mentioned earlier, NO and NOz are important precursors of Os formation. However,
27 because Os changes in a nonlinear way with the concentrations of its precursor NOx and VOCs,
28 it is unlike many other secondarily formed atmospheric species whose rates of formation vary
29 directly with emissions of their precursors. At the low NOx concentrations found in most
30 environments (ranging from remote continental areas to rural and suburban areas downwind of
31 urban centers) the net production of 03 increases with increasing NOx. At the high NOx
March 2008 2-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations found in downtown metropolitan areas, and especially near busy streets and
2 roadways and in power plant plumes, net destruction of 03 is initiated with the excess NO found
3 there. In the high NOX regime, N02 scavenges OH radicals that would otherwise oxidize VOCs
4 to produce peroxy radicals, which would in turn oxidize NO to NOz. In the low NOx regime,
5 oxidation of VOCs generates excess free radicals; hence Os production is more nearly linear with
6 NOx. Between these two regimes, there is a transition zone in which Os shows only a weak
7 dependence on [NOx] •
8
9
10 2.3 MEASUREMENT METHODS AND ASSOCIATED ISSUES
11 NO is routinely measured using the principle of gas-phase chemiluminescence induced
12 by the reaction of NO with Os at low pressure. The Federal Reference Method (FRM) for NOz
13 makes use of this technique of NO detection with a prerequisite step to reduce NOz to NO on the
14 surface of a molybdenum oxide (MoOx) substrate, heated to between 300 and 400 °C. Because
15 the FRM monitor cannot detect NOz specifically, the concentration of NOz is determined as the
16 difference between the air sample passed over the heated MoOx substrate (the nitrogen oxides
17 total) and the air sample that has not passed over the substrate (the NO).
18 Reduction of N02 to NO on the MoOx substrate is not specific to N02; hence, the
19 chemiluminescence analyzers are subject to unknown and varying interferences produced by the
20 presence in the sample of the other oxidized N compounds, the NOz species shown in the outer
21 box of Figure 2.2-1. This interference by NOz compounds has long been known (Fehsenfeld
22 et al., 1987; Rodgers and Davis, 1989; U.S. Environmental Protection Agency, 1993, 2006;
23 Crosley, 1996; Nunnermacker et al., 1998; Parrish and Fehsenfeld, 2000; McClenny et al., 2002;
24 Dunlea et al., 2007; Steinbacher et al., 2007). These studies have relied on intercomparisons of
25 measurements using the FRM and other techniques for measuring NOz. The sensitivity of the
26 FRM to potential interference by individual NOZ compounds is variable and also depends in part
27 on characteristics of individual monitors, such as the design of the instrument inlet, the
28 temperature and composition of the reducing substrate, and on the interactions of atmospheric
29 species with the reducing substrate.
30 Only recently have attempts been made to systematically quantify the magnitude and
31 variability of the interference by NOz species in ambient measurements of N02. Dunlea et al.
32 (2007) found an average of -22% of ambient N02 (~9 to 50 parts per billion [ppb]) measured in
March 2008 2-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mexico City was due to interference from NOz compounds. Comparable levels of N02 are
2 found in many locations in the United States. Dunlea et al. (2007) compared N02 measured
3 using the conventional chemiluminescent instrument with other (optical) techniques. The main
4 sources of interference were HNOs and various RON02. Efficiency of conversion was estimated
5 to be -38% for HN03; for PAN, -95% and - 95% for other RON02. Peak interference of up to
6 50% was found during afternoon hours and was associated with Os and NOz compounds such as
7 HNOs and the alkyl and multifunctional alkyl nitrates.
8 In a study in rural Switzerland, Steinbacher et al. (2007) compared measurements of N02
9 continuously measured using a conventional NOx monitor and measurements in which N02 was
10 photolyzed to NO. They found the conventional technique using catalytic reduction (as in the
11 FRM) overestimated the N02 measured using the photolytic technique on average by 10%
12 during winter and 50% during summer.
13 Another approach to estimating the measurement interference is to use model
14 calculations in conjunction with data for the efficiency of reduction of NOz species on the
15 catalytic converters. Lamsal et al. (2007) used satellite data along with output from the GEOS-
16 CHEM global chemical transport model (CTM) to derive seasonal correction factors across the
17 United States. These factors range from <10% in winter in the East to >80%, with the highest
18 values found during summer in relatively unpopulated areas. These correction factors are based
19 on data collected during satellite overpass in early afternoon and, thus, are applicable only for
20 that time of overpass. Calculations using EPA's Community Multiscale Air Quality (CMAQ)
21 modeling system for the Mid-Atlantic region in a domain extending from Virginia to southern
22 New Jersey were made at much higher spatial resolution than the GEOS-CHEM simulations (see
23 http://www.mde.state.md.us/Programs/AirPrograms/air_planning/index.asp). The daily average
24 interference for an episode during the summer of 2002 ranged from -20% in Baltimore to -80%
25 in Madison, VA. Highest values were found during the afternoon, when photochemical activity
26 is highest, and lowest values during the middle of the night. The model calculations showed
27 episode averages of the N0z/N02 ratio ranging from 0.26 to 3.6 in rural Virginia; the highest
28 ratios were in rural areas, and lowest were in urban centers closer to sources of fresh NOX
29 emissions. (The capabilities of three-dimensional CTMs such as GEOS-CHEM and CMAQ and
30 issues associated with their use are presented in Annex AX2.7.) It appears that interference is
31 likely to be on the order of 10% or less during most or all of the day during winter, but much
March 2008 2-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 larger interference is likely to be found during summer in the afternoon. In general, the
2 interference in the measurement of NOz is greater downwind of urban source areas and in
3 relatively remote areas away from concentrated sources as compared to the level of interference
4 at measurements in urban cores with fresh NOx emissions.
5
6 2.3.1 Measurement Methods Specific to NO2
7 There are approaches to measuring N02 not affected by the artifacts mentioned above.
8 For example, NOz can be photolytically reduced to NO with an efficiency of -70%, as used in
9 the Steinbacher et al. (2007) study. This method requires additional development to ensure its
10 cost effectiveness and reliability for extensive field deployment. The relatively low and variable
11 conversion efficiency of this technique would necessitate more frequent calibration. Optical
12 methods such as those using differential optical absorption spectroscopy (DOAS) or laser
13 induced fluorescence (LIF) are also available, as described in Annex AX2.8. However, these
14 particular methods are more expensive than either the FRM monitors or photolytic reduction
15 technique and require specialized expertise to operate. Moreover, the DOAS obtains an area-
16 integrated measurment rather than a point measurement. Cavity attenuated phase shift (CAPS)
17 monitors are an alternative optical approach that is potentially less costly than DOAS or LIF
18 (Kebabian et al., 2007). However, this technique is not highly specific to NOz and is subject to
19 interference by other species absorbing at 440 nm, such as the 1,2-dicarbonyls. The extent of
20 this interference and the potential of the CAPS technique for extensive field deployment have not
21 been evaluated.
22
23 2.3.2 Measurement of Total Oxidized Nitrogen Species in the Atmosphere
24 Commercially available NOx monitors have been converted to NOy monitors by moving
25 the MoOx convertor to interface directly with the sample inlet. Because of losses on inlet
26 surfaces and differences in the efficiency of reduction of NOz compounds on the heated MoOx
27 substrate, NOX cannot be considered as a universal surrogate for NOY. However, in settings
28 close to relatively high-concentration fresh emissions like those during urban rush hour, most of
29 the NOy is present as NOx. To the extent that all the major oxidized N species can be reduced
30 quantitatively to NO, measurements of NOy should be more reliable than those of NOx,
31 particularly at typical ambient levels of N02. It is worth reiterating that with the current FRM
32 monitors, the direct measurements of NO are the most specific. Measurements of total NOy
March 2008 2-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 characterize the entire suite of oxidized N compounds to which humans are exposed. Reliable
2 measurements of NOy and N02, especially at the low concentrations observed in many areas
3 remote from sources are also crucial for evaluating the performance of three-dimensional,
4 chemical transport models of oxidant and acid production in the atmosphere (described in Annex
5 AX2.7).
6
7
8 2.4 AMBIENT CONCENTRATIONS OF NO2 AND ASSOCIATED
9 OXIDIZED NITROGEN SPECIES AND POLICY-RELEVANT
10 BACKGROUND CONCENTRATIONS
11 This brief overview of ambient concentrations of NO 2 and associated oxidized N
12 compounds in the United States provides estimates of Policy-Relevant Background (PRB)
13 concentrations, i.e., background concentrations used to inform risk and policy assessments for
14 the review of the NAAQS.
15
16 2.4.1 Ambient Concentrations
17 Figure 2.4-1 shows the distribution of monitoring sites for NOz across the United States.
18 As can be seen from Figure 2.4-1, there are large areas of the United States for which data for
19 ambient N02 are either not collected or are collected at very few sites. N02 is monitored mainly
20 in several large urban areas. Few cities have more than two monitors and several large cities,
21 including Seattle, WA, have none. Note that the number of NOz monitors has been decreasing in
22 the United States as ambient average concentrations have fallen to a few tenths of the level of the
23 NAAQS. There were, for example, 375 N02 monitors identified in mid 2006, but only 280 in
24 November 2007.
25 Criteria for siting ambient monitors for NAAQS pollutants are given in the SLAMS /
26 NAMS / PAMS Network Review Guidance (U.S. Environmental Protection Agency, 1998). As
27 might be expected, criteria for siting monitors differ by pollutant. NOz monitors are meant to be
28 representative of several scales: middle (several city blocks, 300 to 500 m), neighborhood (0.5
29 to 4 km), and urban (4 to 50 km). Middle- and neighborhood-scale monitors are used to
30 determine highest concentrations and source impacts, while neighborhood- and urban-scale
31 monitors are used for monitoring population exposures. As can be seen, there is considerable
March 2008 2-10 DRAFT-DO NOT QUOTE OR CITE
-------
Monitor Locator Map - Criteria Air Pollutants
United States
Shaded slates have monitors
AirData
Monitor Location: A N02 (280]
Source: US EPA Office of Air and Radiation, AQS Database
.' • >, • •- ••• - • ' ;
Figure 2.4-1. Location of ambient NOi monitors in the United States as of November 5,
2007. Shaded states have NOi monitors; unshaded states have none.
1 overlap between monitoring objectives and scales of representativeness. The distance of
2 neighborhood- and urban-scale monitor inlets from roadways increases with traffic volume and
3 can vary from 10 to 250 m away from roadways as traffic volume increases. Where the distance
4 of an inlet to a road is shorter than the value in this range for the indicated traffic volume on that
5 road, that monitor is classified as middle scale. Vertically, the inlets to NOz monitors can be set
6 at a height from 2 to 15m.
7 Figure 2.4-2 shows box plots of ambient concentrations of NOz measured at all
8 monitoring sites located within Metropolitan Statistical Areas (MSAs) or urbanized areas in the
9 United States from 2003 through 2005. As can be seen from Figure 2.4-2, mean [NOz] are
March 2008
2-11
DRAFT-DO NOT QUOTE OR CITE
-------
100
90
80
3" 70
Q.
CL
c 60
O
"| 50
•4->
§ 40
0
0 30
20
10
n
r T "T T
* 1201 * 1201 * M29
-
_
*
O
-
0
O
OK • » %*
^ MAX
-p 0 99
""" ^ I Q'S
I — — I ^^
I » • • * 50 * Mean
9*
I TT~^ i
^ _L _L ^ S
1- h max
1-h
24-h
2 week
1-year
Figure 2.4-2. Ambient concentrations of NOi measured at all monitoring sites located
within Metropolitan Statistical Areas in the United States from 2003
through 2005.
1 -15 ppb for averaging periods ranging from a day to a year, with an interquartile range (IQR) of
2 10 to 25 ppb. However, the average of the daily 1 h maximum [N02] over this 3-year period is
3 -30 ppb. These values are about twice as high as the 24-h average. The highest maximum
4 hourly concentration (-200 ppb) found during the period of 2003 to 2005 was more than a factor
5 of ten greater than the overall mean 24-h concentrations. The ratio of the 99th percentile
6 concentration to the mean ranges from 2.1 for the 1-year averages to 3.5 for the 1-h averages.
7 Because ambient N02 monitoring data are so sparse across the United States (see Figure
8 2.4-1) and particularly so in rural areas, it would not be appropriate to use these data in
9 constructing a map of N02 concentrations across the continental United States. The short T of
10 N02 with respect to conversion to NOz species and the concentrated nature of N02 emissions
11 result in steep gradients and low concentrations away from major sources that are not adequately
12 captured by the existing monitoring networks. Model predictions might be more useful for
13 showing large-scale features in the distribution of N02 and could be used in conjunction with the
March 2008
2-12
DRAFT-DO NOT QUOTE OR CITE
-------
1 values shown in Figure 2.4-2 to provide a more complete picture of the variability of N02 across
2 the United States. Monthly average N02 concentrations for July and December 2002 calculated
3 using EPA's CMAQ model are shown in Figures 2.4-3a,b. (A description of the capabilities of
4 CMAQ and other three-dimensional CTMs is given in Annex AX2.7) The high variation in N02
5 concentrations of at least a factor of 10 is apparent in these model estimates. As expected, the
6 highest N02 concentrations are seen in large urban regions, such as the Northeast Corridor, and
7 lowest values are found in sparsely populated regions located mainly in the West. N02
8 concentrations tend to be higher in December than in July.
9
10 2.4.2 Historical [NO2]
11 Trends in N02 concentrations across the United States from 1980 to 2006 are shown in
12 Figure 2.4-4. The white line shows the mean values and the upper and lower borders of the blue
13 (shaded) areas represent the 10th and 90th percentile values. Information on trends at individual,
14 local air monitoring sites can be found at www.epa.gov/airtrends/nitrogen.html.
15 Concentrations were substantially higher during earlier years in selected locations and
16 contributed in those years to the "brown clouds" observed in many cities. Residents in
17 Chattanooga, TN, for example, were exposed more than 30 years ago to high levels of N02 from
18 a munitions plant (Shy and Love, 1980). Annual mean N02 concentrations there declined from
19 -102 ppb in 1968 to -51 ppb in 1972. There was a strike at the munitions plant in 1973 and
20 levels declined to -32 ppb. With the implementation of control strategies, values dropped
21 further. In 1988, the annual mean N02 concentration varied from -20 ppb in Dallas, TX and
22 Minneapolis, MN to 61 ppb in Los Angeles, CA. However, in New York City, the city with the
23 second-highest annual mean concentration in the United States in 1988, the mean N02
24 concentration was 41 ppb.
25 In contrast to most urban areas in the United States, in other countires, N02
26 concentrations have increased. For example, annual mean N02 concentrations in central London
27 increased during the 1980s from -25 ppb in 1978 to -40 ppb in 1989 at the background
28 measurement site and from -35 to -45 ppb at the roadside site. Corresponding NO
29 concentrations increased from -20 ppb to -40 ppb at the background site and from -125 to
30 -185 ppb at the roadside site (Elsom, 1992).
March 2008 2-13 DRAFT-DO NOT QUOTE OR CITE
-------
20,000 112
15.000
10.000
5.000
0.000
January 2002
Min = 0.019 at (1,1), Max = 45.966 at (23,46)
148
20.000 112
15.000
10.000
5.000
0.000
July 2002
Min = 0.012 at (6,4), Max = 40.802 at (23,46)
Figure 2.4-3a,b. Monthly average NOi concentrations for January 2002 (a) and July
2002 (b) calculated by CMAQ (36 x 36 km horizontal resolution).
1 2.4.3 Seasonal Variability in NOi at Urban Sites
2 The month-to-month variability in 24-h average NOz concentrations at two sites in
3 Atlanta, GA is shown in Figure 2.4-5; variability at other individual sites in selected urban areas
4 is shown in Annex 3, Figures AX3.4 to AX3.10. As might be expected from an atmospheric
5 species that behaves essentially like a primary pollutant emitted from surface sources, there is
6 strong seasonal variability in NOz concentrations in the data shown in Figures 2.4-5a-b. Higher
7 concentrations are found during winter, consistent with the lowest mixing layer heights found
8 during the year. Lower concentrations are found during summer, consistent with higher mixing
9 layer heights and increased rates of photochemical oxidation of N02 to NOZ. Note also the day-
10 to-day variability in NOz concentration, which also tends to be larger during the winter. There
11 appears to be a somewhat regular pattern for the other southern cities examined with their winter
12 maxima and summer minima. Monthly maxima tend to be found from late winter to early spring
13 in Chicago, IL, and New York, NY, with minima occurring from summer through the fall.
14 However, in Los Angeles and Riverside, CA, monthly maxima tend to occur from autumn
15 through early winter, with minima occurring from spring through early summer.
March 2008
2-14
DRAFT-DO NOT QUOTE OR CITE
-------
NO2 Air Quality, 1980 - 2006
(Based on Annual Arithmetic Average)
National Trend based on 87 Sites
U.UU i i i i i i i i i i i j i i r r r r i i i i i i i
111111111111111111112222222
999999999999999999990000000
888888888899999999990000000
012345678901234567890123456
1980 to 2006: 41% decrease in National Average
Figure 2.4-4. Nationwide trend in NOi concentrations. The white line shows the mean
values, and the upper and lower borders of the blue (shaded) areas represent
the 10th and 90th percentile values. Information on trends at individual,
local air monitoring sites can be found at www.epa.gov/airtrends/
nitrogen.html
1 Mean and peak N02 concentrations during winter can be up to a factor of two greater than those
2 during the summer at sites in Los Angeles.
3
4 2.4.4 Diurnal Variability in NOi Concentrations
5 The diurnal variability in N02 concentrations at the same two sites in the Atlanta
6 metropolitan area shown in Figures 2.4-5a,b is illustrated in Figures 2.4-6a-d. As can be seen
7 from these figures, N02 typically exhibits daily maxima during the morning rush hours, although
8 they can occur at other times of day. In addition, there are differences between weekdays and
9 weekends. At both sites, N02 concentrations are generally lower and the diurnal cycles more
10 compressed on weekends than on weekdays. The diurnal variability of N02 at these sites is
11 typical of that observed at other urban sites. Monitor siting plays a role in determining diurnal
March 2008
2-15
DRAFT-DO NOT QUOTE OR CITE
-------
a. Atlanta, GA.
SUBURBAN
a.
a.
2
4-1
s
u
c
O
o
0.09:
0,08^
0.07-
0,06:
0.05-
0.04-
0.03:
0.02-
0.01:
0.00:
site id=130890002 poc=1
= Natural Spline Fitw/ 9 df
01/01/2003 07/01/2003 01/01/2004 07/01/2004 01/01/2005 07/01/2005 01/01/2006
Sample Date (mm/dd/yyyy)
b. Atlanta, GA.
URBAN and CENTER CITY
5.
c
o
u
c
o
o
siteid=131210048poc=1
0.09:
0.08;
0.07-
0.06-
0.05-
0.04:
0.03^
0.02-
0.01-
0.00-
I I I I T
01/01/2003 07/01/2003 01/01/2004 07/01/2004 01/01/2005 07/01/2005 01/01/2006
Sample Date (mm/dd/yyyy)
Figure 2.4-5a,b. Time series of 24-h average NOi concentrations at individual sites in
Atlanta, GA from 2003 through 2005. A natural spline function (with
9 degrees of freedom) was fit and overlaid to the data (dark solid line).
March 2008
2-16
DRAFT-DO NOT QUOTE OR CITE
-------
A. Atlanta, GA Suburban Weekday B. Atlanta, GA Suburban Weekend
.15-
a.
5
.2 .10
2
8
£ .05
o
0
.10
.05
* I X X
HIM
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Hour Hour
Q.
C. Atlanta, GA Urban & City Center Weekday D. Atlanta, GA Urban & City Center Weekday
.15,
.Q .10
o>
o
O
10 ppb
10 in general, with the highest HNOs concentrations and the highest ratio of HNOs/NOz found
March 2008
2-17
DRAFT-DO NOT QUOTE OR CITE
-------
1 downwind from central Los Angeles in the San Bernadino Valley during summer, as one would
2 expect for this more oxidized N product.
3 Measurements of HONO in urban areas are very limited; however, data from Stutz et al.
4 (2004) and Wang and Lu (2006) indicate that levels of HONO are <1 ppb even under heavily
5 polluted conditions, with the highest levels found during the night and just after dawn and the
6 lowest values found in the afternoon. Several field studies conducted at ground level (Hayden
7 et al., 2003, in rural Quebec; Williams et al.,1987, near Boulder, CO) and aircraft flights (Singh
8 et al., 2007, over eastern North America) have also found much higher [NOz] than [NOx] in
9 relatively unpolluted rural air.
10
11 2.4.6 Policy Relevant Background Concentrations of NOi
12 Background N02 concentrations used for purposes of informing decisions about NAAQS
13 are referred to as PRB concentrations. PRB concentrations are those that would occur in the
14 United States in the absence of anthropogenic emissions in continental North America (defined
15 here as the United States, Canada, and Mexico). PRB concentrations include contributions from
16 natural sources everywhere in the world and from anthropogenic sources outside these three
17 countries. Background levels defined in this way facilitate separation of pollution levels that can
18 be controlled by U.S. regulations (or through international agreements with neighboring
19 countries) from levels that are generally uncontrollable by the United States. These levels may
20 also be used in quantitative risk assessments of human health and environmental effects.
21 Contributors to PRB concentrations include natural emissions of NO, NOz, and reduced
22 nitrogen compounds, as well as their long-range transport from outside North America. Natural
23 sources of N02 and its precursors include biogenic emissions, wildfires, lightning, and the
24 stratosphere. Biogenic emissions from agricultural activities, such as emissions of NO from
25 fertilized soils, are not considered to be contributing to the formation of PRB concentrations.
26 Discussions of the sources and estimates of emissions are given in Annex AX2.6.2.
27
28 2.4.6.1 Analysis of Policy Relevant Background Contribution to NOi Concentrations
29 over the United States
30 The MOZART-2 global model of tropospheric chemistry (Horowitz et al., 2003) is used
31 to estimate the PRB contribution to [NOz]. The model setup for the present-day simulation has
32 been published in a series of papers from a recent model intercomparison (Dentener et al.,
March 2008 2-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 2006a,b; Shindell et al., 2006; Stevenson et al., 2006; van Noije et al., 2006). MOZART-2 is
2 driven by the U.S. National Oceanic and Atmospheric Administration's National Center for
3 Environmental Prediction (NOAA NCEP) meteorological fields using 2001 data and using 2000
4 emissions from the International Institute for Applied Systems Analysis (IIASA). The model
5 was run at a resolution of 1.9° x 1.9° with 28 sigma levels in the vertical dimension with both
6 gas-phase and aerosol chemistry.
7 Figure 2.4-7 shows the annual mean [NO2] in surface air in the base case simulation (top
8 panel) and the PRB simulation (middle panel), along with the percentage contribution of the
9 background to the total base case N02 (bottom panel). Maximum concentrations in the base case
10 simulation occur along the Ohio River Valley and in the Los Angeles basin. While total surface
11 [N02] are often >5 ppb, PRB is <300 parts per trillion (ppt) over most of the continental United
12 States and <100 ppt in the eastern United States. The distribution of PRB (middle panel of
13 Figure 2.4-7) largely reflects the distribution of soil NO emissions, with some local increases like
14 those in western Montana due to biomass burning. In the northeastern United States, where
15 present-day [NO2] are highest, PRB contributes <1% to the total. Thus, it appears that PRB
16 levels of NOz are much smaller than observed levels.
17
18
19 2.5 EXPOSURE ISSUES
20
21 2.5.1 Introduction
22 Human exposure to an airborne pollutant consists of contact between the human and the
23 pollutant at a specific concentration for a specified period of time. People spend various
24 amounts of time in different microenvironments characterized by different pollutant
25 concentrations. The integrated exposure of a person to a given pollutant is the sum of the
26 exposures over all time intervals for all microenvironments in which the individual spends time.
27 Figure 2.5-1 represents a composite average of activity patterns across all age groups in the
28 United States based on data collected in the National Human Activity Pattern Survey (NHAPS).
29 The demographic distribution of the respondents was designed to be similar to that of overall
March 2008 2-19 DRAFT-DO NOT QUOTE OR CITE
-------
rota
flO*W
280 510
Background
970 I'QS
PPb
001
0. 06 O.U 0.1 S 0.20
ppb
Figure 2.4-7. Upper panel: Annual mean NOi concentrations (in ppb) in the United
States. Middle panel: Annual mean PRB concentrations (in ppb) for NOi in
the United States. These simulations were made using the MOZART-2
global, chemical transport model. The lower panel shows PRB
concentrations expressed as a percentage of total NOi concentrations shown
in the upper panel. See text in Annex AX2.9 for details.
March 2008
2-20
DRAFT-DO NOT QUOTE OR CITE
-------
NHAPS - Nation, Percentage Time Spent
Total n = 9,196
In a risidence (68.7%)
Total time spent
indoors (86.9%)
Office-factory (5.4%)
Outdoors (7.6%)
In a vehicle (5.5%)
Other indoor location (11%%)
Bar-restaurant (1.8%)
Figure 2.5-1. Percentage of time persons spend in different environments in the United
States.
Source: Klepeis et al. (2001).
1 U.S. Census data. Different cohorts, e.g., the elderly, young and middle-aged working adults,
2 and children exhibit different activity patterns.2
3 The personal exposure concentration to a pollutant, such as NOz, can be represented by
4 the following equation:
(2.5-1)
6 where Et is the time-weighted average personal exposure concentration over a certain period of
7 time, n is the total number of microenvironments that a person encounters,/ is the (fractional)
2 For example, the cohort of working adults between the ages of 18 and 65 represents -50% of the population. Of
this total, about 60% work outside the home, spending -24% (40 h/168 h) of their time in factory/office
environments. Thus, this cohort is likely to spend considerably more time in offices and factories than shown in
the figure (5.4 %), which reflects the entire population, and is also likely to spend less time in a residence
compared to small children or the elderly.
March 2008
2-21
DRAFT-DO NOT QUOTE OR CITE
-------
1 time spent in the /th microenvironment, and d is the average concentration in the /'th
2 microenvironment during the time fraction,/. The exposure a person experiences can be
3 characterized as an instantaneous exposure, a peak exposure such as might occur during cooking,
4 an average exposure, or an integrated exposure over all environments a person encounters.
5 These distinctions are important because health effects caused by long-term, low-level exposures
6 may differ from those caused by short-term, peak exposures.
7 An individual's total exposure (ET) can also be represented by the following equation:
ET = Ea + Ena = {y0 + I v/ [PiOi/fai + kj)j } Ca + Ena = {y0 + !» Finf} Ca + Ena
8 ' ' (2.5-2)
9 subject to the constraint,
y0 + Iji = 1
10 i (2.5-3)
1 1 where Ea is the person's exposure to pollutants of ambient origin; Ena is the person's exposure to
1 2 pollutants that are not of ambient origin; y0 is the fraction of time people spend outdoors and y, is
13 the fraction of time they spend in microenvironment /'; Finf Pt, at, and kt are the infiltration
14 factor, penetration coefficient, air exchange rate, and decay rate for microenvironment /'. In the
1 5 case where microenvironmental exposures occur mainly in one microenvironment, Equation
16 2.5-2 may be approximated by Equation 2.5-4:
17
Ena= {>' + (l~y)[Pa/(a + k)]}Ca + Ena = aCa + E
m
18 where y is the fraction of time persons spend outdoors, and a is the ratio of a person's exposure
19 to a pollutant of ambient origin to the pollutant's ambient concentration. Other symbols have the
20 same definitions in Equation 2.5-2 and 2.5-3. If microenvironmental concentrations are
21 considered, then Equation 2.5-4 can be recast as:
22 Cme=Ca+Cnona - [Pa/(a+k)]Ca
23 where Cme is the concentration in a microenvironment; Ca and Cna are the contributions to Cme
24 from ambient and nonambient sources; S is the microenvironmental source strength; and Fis the
25 volume of the microenvironment. The symbols in brackets have the same meaning as in
March 2008 2-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 Equation 2.5-4. In this equation, it is assumed that microenvironments do not exchange air with
2 each other, but only with the ambient environment.
3 Microenvironments in which people are exposed to air pollutants such as N02 typically
4 include residential indoor environments, other indoor locations, near-traffic outdoor
5 environments, other outdoor locations, and in vehicles, as shown in Figure 2.5-1. Indoor
6 combustion sources such as gas stoves and space heaters need to be considered when evaluating
7 exposures to N02. Exposure misclassification may result when total human exposure is not
8 disaggregated between various microenvironments, and this may obscure the true relationship
9 between ambient air pollutant exposure and health outcome.
10 In a given microenvironment, the ambient component of a person's microenvironmental
11 exposure to a pollutant is determined by the following physical factors:
12 • The ambient concentration, Ca
13 • The air exchange rate, a,
14 • The pollutant specific penetration coefficient, P,
15 • The pollutant specific decay rate, k,
16 • The fraction of time an individual spends in the microenvironment, yf
17
18 These factors are in turn affected by the following exposure factors (see Annex AX3.5):
19 • Environmental conditions, such as weather and season
20 • Dwelling conditions, such as house location, which determines proximity to sources
21 and geographical features that can modify transport from sources; the amount of
22 natural ventilation (e.g., open windows and doors, and the "draftiness" of the
23 dwelling) and ventilation system (e.g., filtration efficiency and operation cycle)
24 • Personal activities (e.g., the time spent cooking or commuting)
25 • Indoor sources and sinks of a pollutant
26 • Microenvironmental line and point sources (e.g., lawn equipment)
27 Microenvironmental exposures can also be influenced by the individual-specific factors
28 such as age, gender, health, or socioeconomic status.
29 Time-activity diaries, completed by study participants, are often used in exposure models
30 and assessments. The EPA's National Exposure Research Laboratory (NERL) has consolidated
31 the majority of the most significant human activity databases into one comprehensive database
March 2008 2-23 DRAFT-DO NOT QUOTE OR CITE
-------
1 the Consolidated Human Activity Database (CHAD). Eleven different human activity pattern
2 studies were evaluated to obtain over 22,000 person-days of 24-h human activities in CHAD
3 (McCurdy et al., 2000). These data can be useful in assembling population cohorts to be used in
4 exposure modeling and analysis.
5 In general, the relationship between personal exposures and ambient concentrations can
6 be modified by microenvironments. During infiltration, ambient pollutants can be lost through
7 chemical and physical loss processes; therefore, the ambient component of a pollutant's
8 concentration in a microenvironment is not the same as its ambient concentration but the product
9 of the ambient concentration and the infiltration factor (Fmfor a if people spend 100% of their
10 time indoors). In addition, exposure to nonambient, microenvironmental sources modifies the
11 relationship between personal exposures and ambient concentrations.
12 In practice, it is extremely difficult to characterize community exposure by individual
13 personal exposure. Instead, the distribution of personal exposure in a community, or the
14 population exposure, is characterized by extrapolating measurements of personal exposure using
15 various techniques or by stochastic, deterministic, or hybrid exposure modeling approaches such
16 as APEX, SHEDS, and MENTOR (see AX3.7 for a description of modeling methods).
17 Variations in community-level personal exposures are determined by cross-community
18 variations in ambient pollutant concentrations and the physical and exposure factors mentioned
19 above. These factors also determine the strength of the association between population exposure
20 to N02 of ambient origin and ambient N02 concentrations.
21 Of major concern is the ability of NOz as measured by ambient monitors to serve as a
22 reliable indicator of personal exposure to NOz of ambient origin. The key question is what errors
23 are associated with using NOz measured by ambient monitors as a surrogate for personal
24 exposure to ambient N02 and/or its oxidation products in epidemiologic studies. There are three
25 aspects of this issue: (1) ambient and personal sampling issues; (2) the spatial variability of
26 ambient N02 concentrations; (3) the associations between ambient concentrations and personal
27 exposures as influenced by exposure factors, e.g., proximity to traffic, indoor sources and sinks,
28 and the time people spend indoors and outdoors. These issues are treated individually in the
29 following subsections.
March 2008 2-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.5.2 Personal Sampling of NO2
2 Personal exposures in human exposure and panel studies of N02 health effects are
3 monitored by passive samplers. Their performance is evaluated by comparison to the
4 chemiluminescence monitoring method. Some form of evaluation is crucial for determining
5 measurement errors associated with exposure estimates. However, measurements of N02 are
6 subject to artifacts both at the ambient level and at the personal level. As discussed in Section
7 2.3, measurements of ambient N02 are subject to an unknown and variable level of interference
8 caused by other NOY compounds, in particular HN03, PANs, HONO, and RON02.
9 The most widely used passive samplers are Palmes tubes (Palmes et al., 1976),
10 Yanagisawa badges (Yanagisawa and Nishimura, 1982), Ogawa samplers (Ogawa and
11 Company, http://www.ogawausa.com), and radial diffusive samplers (Cocheo et al., 1996). The
12 methodology and application of Palmes tubes and Yanagisawa badges were described in the last
13 AQCD for Oxides of Nitrogen (U.S. Environmental Protection Agency, 1993). Descriptions of
14 other commercialized samplers is in Annex AX3.3. These samplers do not use a pump to bring
15 air into contact with the sampling substrate; rather, they rely on diffusion or small scale
16 turbulence to transport N02 to a sorbent (Krupa and Legge, 2000). The sorbent can be either
17 physically sorptive (e.g., active carbon) or chemisorptive (e.g., triethanolamine [TEA], KI,
18 sodium arsenite [NaAs02]); passive samplers for N02 are chemisorptive, i.e., a reagent coated on
19 a support (e.g., metal mesh, filter) chemically reacts with and captures N02. The sorbent is
20 extracted and analyzed for one or more reactive derivatives; the mass of N02 collected is derived
21 from the concentration of the derivative (s) based on the stoichiometry of the reaction.
22 The effect of environmental conditions (e.g., temperature, wind speed, humidity) on the
23 performance of passive samplers is a concern when used for residential indoor, outdoor, and
24 personal exposure studies because of sampling rates that deviate from the ideal and can vary
25 throughout the sampling period. Overall, field test results of passive sampler performance are
26 not consistent, and they have not been extensively studied over a wide range of concentrations,
27 wind velocities, temperatures, and relative humidities (Varshney and Singh, 2003).
28 Another concern with the passive sampling method is interference from other pollutants.
29 Interference from other NOy species can contribute to N02 exposure monitoring errors, but the
30 kinetics and stoichiometry of interferent compound reactions have not been well established,
31 especially for passive samplers; an N02 monitoring plan to use tube-type TEA passive samplers
March 2008 2-25 DRAFT-DO NOT QUOTE OR CITE
-------
1 has been proposed and implemented throughout Great Britain, for example. However, in a
2 comparison of N02 concentrations measured outdoors by the passive samplers with those
3 measured by the chemiluminescence method, N02 concentrations measured by the passive
4 samplers were -30% higher than those measured by the chemiluminescence method (Campbell
5 etal.,1994).
6 Although most studies indicate that passive samplers have very good precision, generally
7 within 5% (Gair et al, 1991; Gair and Penkett, 1995; Plaisance et al., 2004; Kirby et al, 2001),
8 field evaluation studies showed that the overall average N02 concentrations calculated from
9 diffusion tube measurements were likely to be within 10% of chemiluminescent measurement
10 data (Bush et al., 2001; Mukerjee et al., 2004). As mentioned before, TEA-based diffusive
11 sampling methods tend to overestimate N02 concentrations in field comparisons with
12 chemiluminescence analyzers (Campbell et al., 1994). This could be due in part to chemical
13 reactions between Os and NO occurring in the diffusion tube or to differential sensitivity to other
14 forms of NOy, such as HONO, PAN, and HNOs, between the passive samplers and the
15 chemiluminescence analyzers (Gair et al., 1991). Due to spatial and temporal variability of NO
16 and N02 concentrations, especially at roadsides where NO concentrations are relatively high and
17 when sufficient 03 is present for interconversion between the species, the lack of agreement
18 between the passive samplers and ambient monitors can represent differences in sampler
19 response (Heal et al., 1999; Cox, 2003).
20 A third aspect of passive sampler performance is that, compared with ambient
21 chemiluminescence monitors, passive samplers give relatively longer time-averaged
22 concentrations (from days to weeks). Consequently, diffusive samplers including those used for
23 N02 monitoring provide integrated but not high time-resolution concentration measurements.
24 Hourly fluctuations in N02 concentrations may be important to the evaluation of exposure-health
25 effects relationships, and continuous monitors, such as the chemiluminescent monitors, remain
26 the only approach for estimating short-term, peak exposures.
27
28 2.5.3 Spatial Variability in NO2 Concentrations
29
30 2.5.3.1 Variability of NOi Concentrations Across Ambient Monitoring Sites
31 Summary statistics for the spatial variability in several urban areas across the United
32 States are shown in Table 2.5-1. Data were obtained from EPA's Air Quality System (AQS).
March 2008 2-26 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-1. SPATIAL VARIABILITY OF NO2 IN SELECTED
UNITED STATES URBAN AREAS
New York, NY
(5)
Atlanta, GA
(5)
Chicago, IL
(7)
Houston, TX
(7)
Los Angeles, CA
(14)
Riverside, CA
(9)
Mean
Concentration (ppb)
29
(25-37)
11
(5-16)
22
(6-30)
13
(7-18)
25
(14-33)
21
(5-32)
r
0.77-0.90
0.22-0.89
-0.05-0.83
0.31-0.80
0.01-0.90
0.03-0.84
P90 (ppb)
7-19
7-24
10-39
6-20
8-32
10-40
COD
0.08-0.23
0.15-0.59
0.13-0.66
0.13-0.47
0.08-0.51
0.14-0.70
1 These areas were chosen because they are the major urban areas with at least five monitors
2 operating from 2003 to 2005. Values in parentheses below the city name indicate the number of
3 monitoring sites in that particular city. The second column shows the 3-year mean concentration
4 across all sites and the range in these means at individual sites. Metrics for characterizing spatial
5 variability include the use of Pearson correlation coefficients (r; column 3), the 90th percentile
6 (P90) of the absolute difference in concentrations (column 4), and coefficient of divergence
7 (COD; column 5).
8 These three metrics are calculated based on measurements of daily average
9 concentrations at individual site pairs. The COD provides an indication of the variability across
10 the monitoring sites in each city and is defined in Equation 2.5-6, as follows
11 • i=i u « (2.5-6)
12 where Xtj and ^ represent observed concentrations averaged over some measurement averaging
13 period (hourly, daily, etc.), for measurement period / at sitey and site k, and/? is the number of
March 2008 2-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 observations. A COD of 0 indicates there are no differences between concentrations at paired
2 sites (spatial homogeneity), while a COD approaching 1 indicates extreme spatial heterogeneity.
3 The same statistics shown in Table 2.5-1 have been used to describe the spatial variability of
4 PM2.5 (U.S. Environmental Protection Agency, 2004; Pinto et al., 2004) and 03 (U.S.
5 Environmental Protection Agency, 2006).
6 As can be seen from Table 2.5-1, mean concentrations at individual sites vary by factors
7 of 1.5 to 6 in the MSAs examined. The sites in New York City tend to be the most highly
8 correlated and show the highest mean levels, reflecting their proximity to traffic, as evidenced by
9 the highest mean concentration of all the entries. They are also located closer to each other than
10 sites in western cities. Correlations between individual site pairs range from slightly negative to
11 highly positive in all of the urban areas except for New York City. However, correlation
12 coefficients are not sufficient for describing spatial variability, as daily average concentrations at
13 two sites may be highly correlated but show differences in levels. Thus, the range in mean
14 concentrations is given. Even in New York City, the spread in mean concentrations is -40% of
15 the citywide mean (12 ppb / 29 ppb). The relative spread in 3-year mean concentrations is larger
16 in the other urban areas shown in Table 2.5-1. As might be expected, the 90th percentile
17 concentration ranges are even larger than the ranges in the means.
18 Because of relative sparseness in data coverage for N02, spatial variability in all cities
19 considered for PM2.5 and 03 could not be considered here. Thus, the number of cities included
20 here is much smaller than for either Os (24 urban areas) or PM2.s (27 urban areas). Even in those
21 cities where there were monitors for all three pollutants, data may not have been collected at the
22 same locations, and even if they were, there will be different responses to local sources. For
23 example, concentrations of N02 collected near traffic will be highest in an urban area, but
24 concentrations of Os will tend to be lowest there because of titration by NO forming N02.
25 However, some general observations can still be made. Mean concentrations of N02 at
26 individual monitoring sites are not as highly variable as for Os but are more highly variable than
27 PM2.5. Lower bounds on intersite correlation coefficients for PM2.5 and for 03 tend to be much
28 higher than for N02 in the same areas shown in Table 2.5-1. CODs for PM2.5 are much lower
29 than for Os, whereas CODs for N02 tend to be the largest among these three pollutants.
March 2008 2-28 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.5.3.2 Small-Scale Horizontal Variability
2 N02 monitors are sited for compliance with air quality standards rather than for capturing
3 small-scale variability in NOz concentrations near sources such as roadway traffic. Significant
4 gradients in NOz concentrations near roadways have been observed in several studies, and NOz
5 concentrations have been found to be correlated (or inversely correlated) with distance from
6 roadway, traffic volume, season, road length, open space, and population density (Gilbert et al.,
7 2007; Bignal et al., 2007; Singer et al., 2004; Cape et al., 2004; Pleijel et al., 2004; Maruo et al.,
8 2003; Roorda-Knape et al., 1998, 1999; Monn et al., 1997; Gauderman et al., 2005). A sample
9 gradient is shown in Figure 2.5-2.
10 Singer et al. (2004) found a strong gradient for concentrations downwind of freeways
11 within the first 230 m. Gilbert et al (2007) found that associations remained robust when sites
12 within 200 m of roadways were removed from the analysis, indicating that traffic influences
13 concentrations as far as 2000 to 3000 m from roadways. Small-scale spatial variations in NOz
14 concentrations are more pronounced during spring and summer seasons due to meteorology and
15 increased photochemical activity (Monn, 2001).
16 Localized effects of roadway sources lead to variability in N02 concentrations that is not
17 captured by the regulatory monitoring network. This variation affects population-level exposure
18 estimates and adds exposure error to time-series epidemiologic studies relying on ambient
19 concentrations as indicators of exposure. Elevated concentrations near roadways also increase
20 exposure of vulnerable populations residing, working, or attending school in the vicinity.
21
22 2.5.3.3 Small-Scale Vertical Variability
23 Inlets to instruments for monitoring gas-phase criteria pollutants can be located from 3 to
24 15m above ground level (Code of Federal Regulations, 2002). Depending on the pollutant, there
25 can be a positive, negative, or no vertical gradient from the surface to the monitor inlet. Positive
26 gradients (i.e., concentrations increase with height) result when pollutants are formed over large
27 areas by atmospheric photochemical reactions (i.e., secondary pollutants such as 03) and
28 destroyed by deposition to the surface or by reaction with pollutants emitted near the surface.
29 Pollutants that are emitted by sources at or just above ground level show negative vertical
30 gradients. Pollutants with area sources (widely dispersed surface sources) and that have minimal
31 deposition velocities show little or no vertical gradient. Restrepo et al. (2004) compared data for
March 2008 2-29 DRAFT-DO NOT QUOTE OR CITE
-------
Cll
O 1.5
•o
N 1.0
15
o 0.5
Z
0.0
2.5
X
0 2.0
•o
O 4 C
.N I-O
15
E 1.0
o
0.5
n n
5
n i
I
I I I
i
n
n
* 1-880
| D 1 - 580
- CA92
*!
x i
I I I
1
s
I *
-2000 -1500 -1000 -500 0 500 1000 1500
Downwind distance to nearest freeway (m)
2000
Figure 2.5-2. NOi and NOx concentrations normalized to ambient values, plotted as a
function of downwind distance from the freeway. Symbols indicate freeway
closest to each monitor.
Source: Singer et al. (2004).
1 criteria pollutants collected at fixed monitoring sites at 15 m above the surface on a school
2 rooftop to those measured by a van whose inlet was 4 m above the surface at monitoring sites in
3 the South Bronx during two sampling periods in November and December 2001. They found
4 that CO, S02, and N02 showed negative vertical gradients, whereas 03 showed a positive
5 vertical gradient and PM2.s showed no significant vertical gradient. As shown in Figure 2.5-3,
6 N02 mixing ratios obtained at 4 m (mean -74 ppb) were about a factor of 2.5 higher than at 15 m
7 (mean -30 ppb). Because tail pipe emissions occur at lower heights, N02 values could have
March 2008
2-30
DRAFT-DO NOT QUOTE OR CITE
-------
01 .
0 08 -
n fifi .
0 04 -
n n9 -
0 -
1
•••
, .
» * *
1 /( T *
\ 1 \ f
jf__" • /
t
* **
*
' * * .
* "F
V A
*">ij ''''AM! ^ A W^^
m
*
*» *
.
•s^A^-Cv
Van — ---DEC709406---A---DEC709407
Figure 2.5-3. NOi concentrations measured at 4 m (Van) and at 15 m at NY Department of
Environmental Conservation ambient monitoring sites (DEC709406 and
DEC709407).
Source: Restrepo et al. (2004).
1 been much higher nearer to the surface and the underestimation of NOz values by monitoring at
2 15m even larger. Restrepo et al. (2004) noted that the use of the N02 data obtained by the
3 stationary monitors underestimates human exposures to NOz in the South Bronx. This situation
4 is not unique to the South Bronx and could arise in other large urban areas in the United States
5 with similar settings.
6 The magnitude of the vertical gradient of NOz in "street canyons" depends strongly on
7 the configuration of the buildings forming the canyons and the meteorological conditions; in
8 particular, static stability in the lower planetary boundary layer, local wind direction and speed,
9 and differential solar heating all affect turbulence in street canyons. These meteorological
10 factors also help determine the relative importance of turbulence induced by traffic, in addition
11 to traffic volume and speed. Descriptions of the effects for many of these factors are available
1 2 only from complex numerical models such as large eddy simulations and very fine grid
March 2008
2-31
DRAFT-DO NOT QUOTE OR CITE
-------
1 resolution computational fluid dynamics models. Thus the quantitative extrapolation of these
2 results to other situations even at the same location at different times is highly problematic.
3 Weak associations might be found between concentrations at ambient monitors and other
4 outdoor locations and between concentrations in indoor microenvironments and personal
5 exposures in part because of the spatial (horizontal and vertical) variability in NOz. This
6 variability is itself location- and time-dependent, and can lead to either over- or underestimates
7 of exposure, depending on the siting of monitors and location of the exposed population. N02
8 ambient monitors may be less representative of community or personal exposures than are
9 ambient monitors for 03 or PM2.s for their respective exposures. This conclusion is based on a
10 comparison of metrics of spatial variability for 03 or PM2.s used in the last AQCD for Particulate
11 Matter (U.S. Environmental Protection Agency, 2004) and AQCD for Os (U.S. Environmental
12 Protection Agency, 2006), indicating generally lower correlations and larger relative spreads in
13 concentrations than for Os or PM2.5. As mentioned earlier, there are far fewer monitors for NOz
14 than for Os or PM2.5, making estimation of the spatial variability in NOz levels more difficult
15 than for 03 or PM2.5.
16
17 2.5.4 Traffic as a Source of NO2
18 Lee et al. (2000) reported that NOz concentration in heavy traffic (-60 ppb) can be more
19 than double that of the residential outdoor level (-26 ppb) in North America. Westerdahl et al.
20 (2005) reported on-road N02 concentrations in Los Angeles ranging from 40 to 70 ppb on
21 freeways, compared to 20 to 40 ppb on residential or arterial roads. NOx concentrations
22 measured at the Caldecott Tunnel in San Francisco in 1999 (Kean et al., 2001) were
23 approximately 7-fold higher at the tunnel exit than at the entrance (1500 ppb versus 200 ppb).
24 People in traffic can potentially experience high concentrations of NOz as a result of the high air
25 exchange rates in vehicles. Park et al. (1998) observed that the air exchange in cars varied from
26 1 to 3 times per hour, with windows closed and no mechanical ventilation, to 36 to 47 times per
27 h, with windows closed and the fan set on fresh air. These results imply that the N02
28 concentration inside a vehicle could rapidly approach the level outside the vehicle during
29 commuting.
30 While driving, concentrations for personal exposure in a vehicle cabin could be
31 substantially higher than ambient concentrations measured nearby. Sabin et al. (2005) reported
March 2008 2-32 DRAFT-DO NOT QUOTE OR CITE
-------
1 that N02 concentrations in the cabins of school buses in Los Angeles ranged from 24 to 120 ppb,
2 which were typically factors of 2 to 3 (max, 5) higher than at ambient monitors in the area.
3 Lewne et al. (2006) reported work hour exposures to N02 for taxi drivers (25.1 ppb), bus drivers
4 (31.4 ppb), and truck drivers (35.6 ppb). These levels were 1.8, 2.7, and 2.8 times the ambient
5 concentrations. Riediker et al. (2003) studied the exposure to N02 inside patrol cars. The
6 authors found that the mean and maximum N02 concentrations in a patrol car were 41.7 ppb and
7 548.5 ppb compared to 30.4 ppb and 69.5 ppb for the ambient sites. These studies suggest that
8 people in traffic can be exposed to much higher levels of N02 than are measured at ambient
9 monitoring sites. Due to high peak exposures while driving, total personal exposure could be
10 underestimated if exposures while commuting are not considered, and sometimes exposure in
11 traffic can dominate personal exposure to N02 (Lee et al., 2000; Son et al., 2004). Variations in
12 traffic-related exposure could be attributed to time spent in traffic, type of vehicle, ventilation in
13 the vehicle, and distance from major roads (Sabin et al., 2005; Son et al., 2004; Chan et al.,
14 1999). Sabin et al. (2005) reported that the intrusion of the vehicle's own exhaust into the
15 passenger cabin is another N02 source contributing to personal exposure while commuting, but
16 that the fraction of air inside the cabin from a vehicle's own exhaust was small, ranging from
17 0.02 to 0.28% and increasing with the age of the vehicle (CARB, 2007a,b).
18 Distance to major roadways could be another factor affecting indoor and outdoor N02
19 concentration and personal N02 exposure. Many studies show that outdoor N02 levels are
20 strongly associated with distance from major roads (i.e., the closer to a major road, the higher the
21 N02 concentration) (Gilbert et al., 2005; Roorda-Knape et al., 1998; Lai and Patil, 2001;
22 Kodama et al., 2002; Gonzales et al., 2005; Cotterill and Kingham, 1997; Nakai et al., 1995).
23 Meteorological factors (wind direction and wind speed) and traffic density are also important in
24 interpreting measured N02 concentrations (Gilbert et al., 2005; Roorda-Knape et al., 1998;
25 Rotko et al., 2001; Aim et al., 1998; Singer et al., 2004; Nakai et al., 1995). For example,
26 Roorda-Knape et al. (1998) reported that N02 concentrations in classrooms were significantly
27 correlated with car and total traffic density (r = 0.68), percentage of time downwind (r = 0.88),
28 and distance of the school from the roadway (r = -0.83). Singer et al. (2004) reported results of
29 the East Bay Children's Respiratory Health Study. The authors found that N02 concentrations
30 increased with decreasing downwind distance for school and neighborhood sites within 350 m
March 2008 2-33 DRAFT-DO NOT QUOTE OR CITE
-------
1 downwind of a freeway, and schools located upwind or far downwind of freeways were
2 generally indistinguishable from one another or by regional pollution levels.
3 Personal exposure is associated with traffic density and proximity to traffic, although
4 personal exposure is also influenced by indoor sources. Aim et al. (1998) reported that weekly
5 average N02 exposures (geometric mean) were higher (p = 0.0001) for children living in the
6 downtown area of Helsinki (13.8 ppb) than in the suburban area (9.1 ppb). Within the urban area
7 of Helsinki, Rotko et al. (2001) observed that the N02 exposure was significantly associated with
8 traffic volume near homes. The average exposure level of 138 subjects having low or moderate
9 traffic near their homes was 12.3 ppb, while the level was 15.8 ppb for the 38 subjects having
10 high traffic volume near home. Gauvin et al. (2001) reported that the ratio of traffic density to
11 distance from a roadway was one of the significant predictors of personal exposure in Grenoble,
12 Toulouse, and Paris. After controlling for indoor source impacts on personal exposure, Kodama
13 et al. (2002) and Nakai et al. (1995) observed that personal exposure decreased with increasing
14 distance from residence to major road.
15 Although traffic is a major source of ambient N02, industrial point sources are also
16 contributors to ambient N02. Nerriere et al. (2005) measured personal exposures to PM2.5, PM
17 with an aerodynamic diamter of < 10 pm (PMio), and N02 in traffic-dominated, urban
18 background, and industrial settings in four French cities (Paris, Grenoble, Rouen, and
19 Strasbourg). Ambient concentrations and personal exposures for N02 were generally highest in
20 the traffic-dominated sector. It should be remembered that there can be high traffic emissions
21 (including shipping traffic) in industrial zones, such as in the Ship Channel in Houston, TX, and
22 in the Port of Los Angeles, CA. In rural areas where traffic is sparse, other sources could
23 dominate. Martin et al. (2003) found that pulses of N02 released from agricultural areas occur
24 after rainfall. Other rural contributors to N02 include wildfires and residential wood burning.
25
26 2.5.5 Indoor Sources and Sinks of NO2 and Associated Pollutants
27 Indoor sources and indoor air chemistry of N02 are important, because they influence the
28 indoor N02 concentrations to which humans are exposed and contribute to total personal
29 exposures. These indoor source and sink terms must be characterized in an exposure assessment
30 if the fraction of a person's exposure to N02 of ambient origin is to be determined.
March 2008 2-34 DRAFT-DO NOT QUOTE OR CITE
-------
1 Penetration of outdoor N02 and indoor combustion in various forms are the major
2 sources of N02 to indoor environments, e.g., homes, schools, restaurants, theaters. As might be
3 expected, indoor concentrations of N02 in the absence of combustion sources are determined by
4 the infiltration of outdoor N02 (Spengler et al., 1994; Weschler et al., 1994; Levy et al., 1998a).
5 Contributions to indoor N02 from the reaction of NO in exhaled breath with Os could potentially
6 be important in certain circumstances (see AX3.4.2 for sample calculations). Indoor sources of
7 nitrogen oxides have been characterized in several reviews, namely the last AQCD for Oxides of
8 Nitrogen (U.S. Environmental Protection Agency, 1993); the Review of the Health Risks
9 Associated with Nitrogen Dioxide and Sulfur Dioxide in Indoor Air for Health Canada (Brauer
10 et al., 2002); and the Staff Recommendations for revision of the N02 standard in California
11 (CARB, 2007a). Mechanisms by which NOx is produced in the combustion zones of indoor
12 sources were reviewed in the last AQCD for Oxides of Nitrogen (U.S. Environmental Protection
13 Agency, 1993). It should be noted that indoor sources can affect ambient N02 levels,
14 particularly in areas in which atmospheric mixing is limited, such as in valleys.
15 Combustion of fossil and biomass fuels is the major indoor source of nitrogen oxides.
16 Combustion of fossil fuels occurs in appliances used for cooking, heating, and drying clothes,
17 e.g., coal stoves, oil furnaces, kerosene space heaters. Motor vehicles and various types of
18 generators in structures attached to living areas also contribute N02 to indoor environments.
19 Indoor sources of N02 from combustion of biomass include wood-burning fireplaces and wood
20 stoves and tobacco.
21 Many studies have noted the importance of gas cooking appliances as sources of N02
22 emissions. Depending on geographical location, season, other sources of N02, and household
23 characteristics, homes with gas cooking appliances have approximately 50% to over 400%
24 higher N02 concentrations than homes with electric cooking appliances (Gilbert et al., 2006; Lee
25 et al., 2000; Garcia-Algar et al., 2003; Raw et al., 2004; Leaderer et al., 1986). Gas cooking
26 appliances remain significantly associated with indoor N02 concentrations after adjusting for
27 several factors that influence exposures, including season, type of community, socioeconomic
28 status, use of extractor fans, household smoking, and type of heating (Garcia-Algar et al., 2004;
29 Garrett, 1999). Homes with gas appliances with pilot lights emit more N02, resulting in N02
30 concentrations -10 ppb higher than in homes with gas appliances with electronic ignition
31 (Spengler et al., 1994; Lee et al., 1998).
March 2008 2-35 DRAFT-DO NOT QUOTE OR CITE
-------
1 Secondary heating appliances are additional sources of NOz in indoor environments,
2 particularly if the appliances are unvented or inadequately vented. As heating costs increase, the
3 use of these secondary heating appliances tends to increase. Gas heaters, particularly when
4 unvented or inadequately vented, produce high levels of indoor NOz (Kodoma et al., 2002).
5 Results summarized by Brauer et al. (2007) indicate that concentrations of NOz in homes with
6 unvented gas hot water heaters were 10 to 21 ppb higher than in homes with vented heaters,
7 which in turn, had N02 concentrations 7.5 to 38 ppb higher than homes without gas hot water
8 heaters. On the other hand, mean concentrations of N02 were all < 10 ppb in a study of Canadian
9 homes with vented gas and oil furnaces and electric baseboard heaters (Weichenthal et al., 2007),
10 suggesting that these are not likely to be significant sources of NOz to indoor environments.
11 Table 2.5-2 shows average concentrations of N02 in homes while combustion sources
12 (mainly gas fired) were in operation. Averaging periods ranged from minutes to hours in the
13 studies shown. Table 2.5-3 shows 24-h to 2-week-long average concentrations of NOz in homes
14 with primarily gas combustion sources.
15 As can be seen from Tables 2.5-2 and 2.5-3, average concentrations while appliances are
16 in operation tend to be much higher than longer-term averages. As Triche et al. (2005) indicated,
17 the 90th percentile concentrations can be substantially greater than the medians, even for 2-week
18 samples. This finding illustrates the high variability of indoor NOz found among homes,
19 reflecting differences in ventilation of emissions from sources, air exchange rates, the size of
20 rooms, etc. The concentrations for short averaging periods listed in Table 2.5-2 correspond to
21 -10 to 30 ppb on a 24-h average basis. As can be seen from inspection of Table 2.5-3, these
22 sources would contribute significantly to the longer-term averages reported if operated daily on a
23 similar schedule. This implies measurements made with long averaging periods may not capture
24 the nature of the diurnal pattern of indoor concentrations of N02 in homes with strong indoor
25 sources, a problem that becomes more evident as ambient N02 levels decrease with more
26 efficient controls on outdoor sources.
27 The emissions of NOz from burning biomass fuels indoors have not been characterized as
28 extensively as those from burning gas. A main conclusion from the 1993 AQCD for Oxides of
29 Nitrogen was that properly vented wood stoves and fireplaces would make only minor
30 contributions to indoor NOz levels, and several studies have concluded that using wood-burning
March 2008 2-36 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-2. NO2 CONCENTRATION NEAR INDOOR SOURCES:
SHORT-TERM AVERAGES
Avg
Concentration (ppb)
191 kitchen
195 living room
184 bedroom
400 kitchen
living room
bedroom
90 (low setting)
350 (med setting)
360 (high setting)
N/R
N/R
180 to 650
Peak
Concentration (ppb)
375 kitchen
401 living room
421 bedroom
673 bedroom
N/R
1000
1500
N/R
Comment
Cooked full meal with gas range for 2 h,
20 min; 7 h TWA.
Self-cleaning gas range. Avg's are over the entire
cycle.
Natural gas unvented fireplace, 0.5 h TWA in
main living area of house (177 m3).
Room concentration with kerosene heater
operating for 46 min.
Room concentration with gas heater operating for
10 min.
Calculated steady-state concentration from
specific unvented gas space heaters1 operating in a
1400 ft2 house, 1.0 h"1 for air exchange rate.
Reference
Fortmann
etal.
(2001)
Fortmann
etal.
(2001)
Dutton et al.
(2001)
Girman et al.
(1982)
Girman et al.
(1982)
Girman et al.
(1982)
N/R = not reported
TWA = time-weighted avg
1 Unvented appliances are not permitted in many areas including California.
1 appliances does not increase indoor N02 concentrations (Levesque et al., 2001; Triche et al.,
2 2005).
3 Other indoor combustion sources of NOz are candle burning and smoking. In a study of
4 students living in Copenhagen, S0rensen et al. (2005) found that personal exposures to N02 were
5 significantly associated with time exposed to burning candles in addition to other sources (data
6 not reported). Results of studies relating NO 2 concentrations and exposures to environmental
7 tobacco smoke (ETS) have been mixed. Several studies found positive associations between
8 N02 levels and ETS (e.g., Linaker et al., 1996; Farrow et al., 1997; Aim et al., 1998; Levy,
9 1998b; Monn et al., 1998; Cyrys et al., 2000; Lee et al., 2000; Garcia Algar, 2004), whereas
10 others have not (e.g., Hackney et al., 1992; Kawamoto et al., 1993).
March 2008
2-37
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-3. NO2 CONCENTRATION NEAR INDOOR SOURCES:
LONG-TERM AVERAGES
Avg Concentration (ppb)
Comment
Reference
30 to 33
22
6 to 11
55 (Median)
41 (90th percentile)
80 (90th percentile)
84 (90th percentile)
147 (90th percentile)
52 (90th percentile)
18
19
15
Gas stoves with pilot lights
Gas stoves without pilot lights
Electric ranges
Study conducted in 517 homes in Boston
Values represent 2-wk avgs
Gas space heaters
No indoor combustion sources
Fireplaces
Kerosene heaters
Gas space heaters
Wood stoves
All values represent 2-wk avgs in living rooms
Bedrooms
Living rooms
Outdoors
Almost all homes had gas stoves
Values represent 2-wk avgs
Leeetal. (1998)
Triche et al. (2005)
Zipprich et al. (2002)
1 2.5.5.1 Indoor Air Chemistry
2 Chemistry in indoor settings can be both a source and a sink for NO 2 (Weschler and
3 Shields, 1997). N02 is produced by reactions of NO with Os or peroxy radicals, while N02 is
4 removed by gas-phase reactions with Os and assorted free radicals and by surface-promoted
5 hydrolysis and reduction reactions. The concentration of indoor N02 also affects the
6 decomposition of PAN.
7 Indoors, NO can be oxidized to N02 by reacting with 03 or peroxy radicals. The latter
8 are generated by indoor air chemistry involving Os and unsaturated hydrocarbons such as
9 terpenes found in air fresheners and other household products (Sawar et al., 2002a,b; Nazaroff
10 and Weschler, 2004; Carslaw, 2007).
11 At an indoor Os concentration of 10 ppb and an indoor NO concentration that is
12 significantly smaller than that of Os, the half-life of NO is 2.5 min (using kinetic data contained
13 in Jet Propulsion Laboratory, 2006). This reaction is sufficiently fast to compete with even
14 relatively fast air exchange rates. Hence, the amount of N02 produced from NO tends to be
15 limited by the amount of 03 available (Weschler et al., 1994).
March 2008
2-38
DRAFT-DO NOT QUOTE OR CITE
-------
1 N02 reacts with 03 to produce nitrate radicals (N03). To date, there have been no indoor
2 measurements of the concentration of NOs radicals in indoor settings. Modeling studies by
3 Nazaroff and Cass (1986), Weschler et al. (1992), Sarwar et al. (2002b), and Carslaw (2007)
4 estimate indoor NOs radical concentrations in the range of 0.01 to 5 ppt, depending on the indoor
5 levels of Os and N02. Once formed, NOs can oxidize organic compounds by either adding to an
6 unsaturated carbon bond or abstracting a hydrogen atom (Wayne et al., 1991). In certain indoor
7 settings, the NOs radical may be a more important indoor oxidant than either Os or the OH
8 radical (Nazaroff and Weschler, 2004; Wayne et al., 1991). Thus, N03 radicals and the products
9 of N03 radical chemistry could contribute to uncertainty in N02 exposure-health outcome studies
10 Reactions between N02 and various free radicals can be an indoor source of organo-
11 nitrates, analogous to the chain-terminating reactions observed in photochemical smog
12 (Weschler and Shields, 1997). Additionally, based on laboratory measurements and
13 measurements in outdoor air (Finlayson Pitts and Pitts, 2000), one would anticipate that N02, in
14 the presence of trace amounts of HNOs, can react with PAHs sorbed onto indoor surfaces to
15 produce mono- and dinitro-PAHs. N02 can also be reduced on certain surfaces, forming NO.
16 Spicer et al. (1989) found that as much as 15% of the N02 removed on various indoor surfaces
17 was reemitted as NO. Weschler and Shields (1996) found that the amount of N02 removed by
18 charcoal filters used in buildings were almost equally matched by the amount of NO
19 subsequently emitted by the same filters.
20 N02 can also be converted to HONO by reactions in indoor air. As noted above, HONO
21 occurs in the atmosphere mainly through multiphase processes involving N02. HONO has been
22 observed to form on surfaces containing partially oxidized aromatic structures (Stemmler et al.,
23 2006) and on soot particles (Ammann et al., 1998). Indoors, surface-to-volume ratios are much
24 larger than they are outdoors, and the surface-mediated hydrolysis of N02 is a major indoor
25 source of HONO (Brauer etal., 1990, 1993; Febo andPerrino, 1991; Spicer etal., 1993;
26 Spengler et al., 1993; Wainman et al., 2001; Lee et al., 2002). Lee et al. (2002) reported average
27 indoor HONO levels were ~6 times higher than outdoor levels (4.6 versus 0.8 ppb). Indoor
28 HONO concentrations averaged 17% of indoor N02 concentrations, and the two were strongly
29 correlated. Indoor HONO levels were higher in homes with humidifiers compared to homes
30 without humidifiers (5.9 versus 2.6 ppb). This last observation is consistent with the studies of
31 Brauer et al. (1993) and Wainman et al. (2001), indicating that the production rate of HONO
March 2008 2-39 DRAFT-DO NOT QUOTE OR CITE
-------
1 from N02 surface reactions increases with relative humidity. Spicer et al. (1993) reported that an
2 equilibrium between adsorption of HONO from the gas range (or other indoor combustion
3 sources) and HONO produced by surface reactions determines the relative importance of these
4 processes in producing HONO in indoor air.
5 A person's total exposure to NO2 cannot be estimated based on consideration of the
6 estimates of emissions given in emissions inventories. Indoor and other microenvironmental
7 sources and a person's activity pattern must be considered in determining the sources that exert
8 the largest influence on a person's total exposure to N02. As examples, exposures in vehicle
9 cabins while commuting to/from school or work, or exposures associated with operation of off-
10 road engines (e.g., lawn and garden or construction equipment), could be larger than integrated
11 24-h exposures due to infiltration of outdoor air into a home.
12
13 2.5.6 Relationships of Personal Exposures to Ambient Concentrations
14
15 2.5.6.1 Associations among Ambient and Outdoor Concentrations and Personal
16 Exposures
17 Results of studies reporting associations between ambient concentrations and personal
18 exposures are shown in Table 2.5-4A and results of studies reporting associations between
19 outdoor concentrations and personal exposures are shown in Table 2.5-4B. Study designs
20 (longitudinal, daily-averaged, and pooled) used in of each of these studies are also briefly
21 summarized in Tables 2.5-4A and B.
22 Figures 2.5-4a and b explicitly summarize the correlation coefficients between personal
23 exposures and ambient concentrations for different populations with a forest plot for U.S.
24 studies and European studies, respectively. Correlation coefficients shown in Figures 2.5-4a
25 and b were transformed from the coefficients in Table 2.5-4A. Fisher's Z transform was used,
26 (Z = 0.51n((l + r)/(l - r))), where r is the originally reported and Z is the transformed correlation
27 coefficient (Fisher, 1925). The variance of Z is expressed as l/(n-3), where n is the number of
28 observations defined by the one of the following three presentations. (1) When the correlation
29 coefficient was based on the average across subjects of personal exposures, n was the number of
30 sampling days. (2) When the partial correlation coefficient was used in the original study, n was
31 the total number of sampling by individual observations minus the sum of three and the number
32 of covariates.
March 2008 2-40 DRAFT-DO NOT QUOTE OR CITE
-------
location
Linn et al. (1996) Southern California All
Samatelal. (2001) Baltimore
Sarnat et al. (2001) Baltimore
Sarnat et al. (2005) Boston
Samat et al. (2005) Boston
Sarnal et al. (2006) Steubenville
Samat et al. (2006) Steubenville
Kim et al. (2006) Toronto
' Note: NR = Not reported
" Percent of data below detection limit
N - Number of observations
Season
Ail
Summer
Winter
Summer
Winter
Summer
Fall
All
Sampling.
Time
1day
1day
1day
1day
1day
1day
1day
1 day
H
107
217
484
298
341
183
228
15
Eisner's 2-TtamfonB
Day's avg of
children
Individual
Individual
Individual
Individual
Individual
Individual
Avg of individual
correlations
%
-------
1 (3) When the mean of individual correlations was used, the standard error was the standard
2 deviation of the correlations divided by the square root of the number of subjects minus one.
3 As shown in Table 2.5-4A and Figures 2.5-4a and b longitudinal and pooled correlations
4 between personal exposure and ambient N02 concentrations varied considerably among studies
5 and study subjects. Most studies report longitudinal correlation coefficients ranging from weak
6 to moderate but statistically significant, indicating that an individual's activities may have a
7 significant effect on personal exposure. Meanwhile, pooled studies usually report poor
8 correlation coefficients between personal exposures and ambient concentrations.
9 Two main aspects of these analyses are discussed below: (1) factors affecting the
10 strength of the association between personal N02 exposure and ambient N02 concentrations, and
11 (2) the meanings of the correlation coefficients in the context of exposure assessments in
12 epidemiologic studies.
13 The strength of the association between personal exposures and ambient and/or outdoor
14 concentrations for a population is determined by variations in indoor or other local sources, air
15 exchange rate, penetration, and decay rate of N02 in different microenvironments and the time
16 people spend in different microenvironments with different N02 concentrations.
17 Home ventilation is an important factor modifying the personal-ambient relationships;
18 one would expect to observe the strongest associations for subjects spending time indoors with
19 open windows. Aim et al. (1998) and Kodama et al. (2002) observed the association between
20 personal exposure and ambient concentration became stronger during the summer than the
21 winter. However, Sarnat et al. (2006) reported that R2 values decreased from 0.34 for a low-
22 ventilation population to 0.16 for a high-ventilation population in the summer, and from 0.47 for
23 a low-ventilation population to 0.34 for a high-ventilation population in the fall. The mixed
24 results remind us that the association between personal exposures and ambient concentrations is
25 complex and determined by many factors.
26 Local and indoor sources also affect the strength of the association between personal
27 exposures and ambient concentrations. Aim et al. (1998) found that the association between
28 personal exposure and outdoor concentration was stronger than the correlation between personal
29 exposure and central site concentration. However, Kim et al. (2006) found that the association
30 was not improved using the ambient sampler closest to a home. The lack of improvement in the
31 strength of the association by choosing the closest ambient monitor could be in part due to the
March 2008 2-42 DRAFT-DO NOT QUOTE OR CITE
-------
1 differences in the small-scale spatial heterogeneity of NOz in different urban areas, as shown in
2 Table 2.5-1. Higher personal to ambient correlations have been found for subjects living in rural
3 areas and lower correlations for subjects living in urban areas (Rojas-Bracho et al., 2002; Aim
4 et al., 1998). Spengler et al. (1994) also observed that the relationship between personal
5 exposure and outdoor concentration was highest in areas with lower ambient NOz levels
6 (R2 = 0.47) and lowest in areas with higher ambient NOz levels (R2 = 0.33). This might reflect
7 the highly heterogeneous distribution or the effect of local sources of N02 in an urban area.
8 Associations between ambient concentrations and personal exposures for the studies
9 examined for NOz were not stratified by the presence of indoor sources except in Aim et al.
10 (1998), Sarnat et al. (2006), Linaker et al. (2000) and Piechocki-Minguy et al. (2006). When
11 there is little or no contribution from indoor sources, ambient concentrations primarily determine
12 exposure; however, if there are indoor sources, the importance of outdoor levels in determining
13 personal exposures decreases. The association between ambient concentrations and personal
14 exposures strengthens after controlling for indoor sources. Raaschou-Nielsen et al. (1997),
15 Spengler et al. (1994), and Gauvin et al. (2001) reported that R2 values increased by 10 to 40%
16 after controlling for indoor sources, such as gas appliances and ETS (see Table 2.5-4A).
17 The strength of the associations between personal exposures and ambient concentrations
18 could also be affected by the quality of the data collected during the exposure studies. There are
19 at least five aspects associated with the quality of the data: method precision, method accuracy
20 (compared with FRM), percent of data above method detection limits (based on field blanks),
21 completeness of the data collection and sample size, and soundness of the quality
22 assurance/quality control procedures. Unfortunately, not all studies reported the five aspects of
23 the data quality issue. Although data imprecisions and inaccuracies are less than 10% in most
24 studies (Section 2.5.2), the fraction of data below the detection limit might be a concern for some
25 studies (see e.g., Sarnat et al., 2000, 2001, 2006). Correlation coefficients would be biased low if
26 data used in their calculation are below detection limits. Sampling interferences (caused by
27 some NCv compounds and other gas species) associated with both ambient (see Section 2.3) and
28 personal sampling (see Section 2.5.2) could also affect data quality. Therefore, caution must be
29 exercised when interpreting the results in Table 2.5-4A.
30 Another factor that can have a substantial effect on the value of the resultant correlation
31 coefficient is the exposure study design as presented in Table 2.5-4A. Not only does the
March 2008 2-43 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposure study design affect the strength of the association between personal exposures and
2 ambient concentrations, but it also determines the meaning of the correlation coefficients in the
3 context of exposure assessment in epidemiologic studies. The correlation coefficient between
4 personal exposures and ambient concentrations has different meanings for different study
5 designs.
6 There are three types of correlations generated from different study designs as listed in
7 Table 2.5-4A: longitudinal, "pooled," and daily-average correlations (U.S. Environmental
8 Protection Agency, 2004). Longitudinal correlations are calculated when data from a study
9 includes measurements over multiple days for each subject (longitudinal study design).
10 Longitudinal correlations describe the temporal relationship between daily personal N02
11 exposure or microenvironment concentration and daily ambient N02 concentration for the same
12 subject. The longitudinal correlation coefficient can differ between subjects. The distribution of
13 correlations across a population could be obtained with this type of data (e.g. Linn et al., 1996;
14 Aim et al., 1998; Linaker et al., 2000; Kim et al., 2006; Sarnat et al., 2000, 2001, 2005, 2006).
15 Pooled correlations are calculated when a study involves one or only a few measurements
16 per subject and when different subjects are studied on subsequent days. Pooled correlations
17 combine individual-subject/individual-day data for the calculation of correlations. Pooled
18 correlations describe the relationship between daily personal N02 exposure and daily ambient
19 N02 concentration across all subjects in the study (e.g., Piechocki-Minguy et al., 2006).
20 Daily-average correlations are calculated by averaging exposure across subjects for each
21 day. Daily-average correlations then describe the relationship between the daily average
22 exposure and daily ambient N02 concentration (e.g., Liard et al., 1999; Gauvin et al., 2001; U.S.
23 Environmental Protection Agency, 2004).
24 In the context of determining the effects of ambient pollutants on human health, the
25 association between the ambient component of personal exposures and ambient concentrations is
26 more relevant than the association between personal total exposures (ambient component +
27 nonambient component) and ambient concentrations. As described in Equations 2.5-2 and 2.5-4,
28 personal total exposure can be decomposed into two parts; an ambient and a nonambient
29 component. Usually, the ambient component of personal exposure is not directly measureable,
30 but it can be estimated by exposure models, or the personal total exposure can be regarded as the
31 personal exposure of ambient origin if there are no indoor or nonambient sources. Personal
March 2008 2-44 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposures were clearly stratified by indoor sources in only four studies among the studies
2 examined for N02 (Aim et al., 1998; Sarnat et al., 2006; Piechocki-Minguy et al., 2006; Linaker
3 et al., 2000) and only two studies (Aim et al., 1998; Piechocki-Minguy et al., 2006) compared the
4 association between personal total exposures and ambient concentrations and the association
5 between the ambient component of personal exposures and ambient concentrations. A stronger
6 association was observed between the ambient component of personal exposures and the ambient
7 concentrations (Aim et al., 1998; Piechocki-Minguy et al., 2006). It is expected that the
8 association between ambient concentrations and the ambient component of personal exposures
9 would be stronger than the association between ambient concentrations and personal total
10 exposures as long as the ambient and nonambient component of personal total exposure are
11 independent. The correlation coefficients between personal ambient N02 exposures and ambient
12 N02 concentrations in different types of exposure studies are relevant to different types of
13 epidemiologic studies.
14 A longitudinal correlation coefficient between the ambient component of personal
15 exposures and ambient concentrations is relevant to the panel epidemiologic study design. In
16 Table 2.5-4A, most longitudinal studies reported the association between personal total
17 exposures and ambient concentrations for each subject; for some subjects the associations were
18 strong and for some subjects the associations were weak. The weak personal and ambient
19 associations do not necessarily mean that ambient concentrations are not a good surrogate for
20 personal exposures, because the weak associations could have resulted from the day-to-day
21 variation in the nonambient component of total personal exposure. The type of correlation
22 analysis can have a substantial effect on the value of the resultant correlation coefficient. Mage
23 et al. (1999) showed that very low correlations between personal exposure and ambient
24 concentrations could be obtained when people with very different nonambient exposures are
25 pooled, even though their individual longitudinal correlations are high. Most studies (employing
26 either cross-sectional or longitudinal study designs) examined in the current review showed that
27 ambient N02 is associated with personal N02 exposure; however, the strength of the association
28 varied considerably.
29 The association between community average exposures (ambient component) and
30 ambient concentrations is more directly relevant to community time-series and long-term cohort
31 epidemiologic studies, in which ambient concentrations are used as a surrogate for community
March 2008 2-45 DRAFT-DO NOT QUOTE OR CITE
-------
1 average exposure to N02 of ambient origin. However, exposure of the population to N02 of
2 ambient origin has not been reported in all the studies examined. The following two European
3 studies reported the associations between population total exposures and ambient or outdoor
4 concentrations of N02. Liard et al. (1999) conducted an exposure study of 55 office workers and
5 39 children in Paris. Measurements were made during three 4-day-long measurement periods for
6 each group. Apart from occasional lapses, data from the same participants were collected during
7 each period. Liard et al. (1999) correlated the five-panel average personal exposures with
8 ambient monitoring data and derived a longitudinal Spearman correlation coefficient of 1
9 (p < 0.001). R2 between ambient monitors and individual personal exposures for adults was
10 0.41, and for children, R2 was 0.17. Four-day averaging periods were chosen in this study to
11 overcome limitations imposed by the levels of detection of the personal samplers. The results
12 show that passive samplers could be used to measure personal exposures in panel studies over
13 multiday periods and lend some credence to the use of stationary monitors as proxies for
14 personal exposures to ambient N02.
15 Monn et al. (1998) and Monn (2001) reported personal N02 exposures obtained in the
16 SAPALDIA study (eight study centers in Switzerland). In each study location, personal
17 exposures for N02 were measured simultaneously for all participants; in addition, residential
18 outdoor concentrations were measured for 1 year (Table 2.5-4B). Monn (2001) observed a
19 strong association between the average personal exposures in each study location and
20 corresponding average outdoor concentrations with an R2 of 0.965. As pointed out by the author,
21 in an analysis of individual single exposure and outdoor concentration data, personal versus
22 outdoor R2 was less than 0.3 (Monn et al., 1998). Because spatial heterogeneity in N02
23 concentrations likely produces stronger associations between average personal exposures and
24 residential monitors than with central site ambient monitors in urban areas, caution should be
25 exercised in using these data to infer that long-term averaged ambient concentrations are a good
26 surrogate for population exposures in long-term cohort epidemiologic studies.
27
28 2.5.6.2 Ambient Contribution to Personal NOi Exposure
29 Another aspect of the relationship of personal N02 exposure and ambient N02 is the
30 contribution of ambient N02 to personal exposures. The infiltration factor (Finf) and alpha (a)
31 are the keys to evaluate personal N02 exposure of ambient origin. As defined in Equations 2.5-2
32 through 2.5-5, the infiltration factor (Fin/) of N02, the physical meaning of which is the fraction
March 2008 2-46 DRAFT-DO NOT QUOTE OR CITE
-------
1 of ambient NOz found in the indoor environment, is determined by the NOz penetration
2 coefficient (P), air exchange rate (a), and the N02 decay rate (k). Alpha (a) is a function of Fmf
3 and the fraction of time people spend outdoors (y), and the physical meaning of a is the ratio of
4 personal ambient exposure concentration to ambient concentration, (i.e., in the absence of
5 exposures to nonambient sources (i.e., when Ena = 0).
6 The values for a and F^/can be calculated physically using Equations 2.5-2 through
7 2.5-5, if P, k, a, and_y are known. However, the values of P and k for NOz are rarely reported,
8 and in most mass balance modeling work, P is assumed to equal 1 and k is assumed to equal
9 0.99 h"1 (Yamanaka, 1984; Yang et al., 2004a; Dimitroulopoulou et al., 2001; Kulkarni et al.,
10 2002). Loupaetal. (2006) reported that k was 0.08 to 0.12 h"1 for NO and 0.04 to 0.11 h"1 for
11 N02 based on real-time measurements in two medieval churches in Cyprus. It is well known
12 that P and k are dependent on a large number of indoor parameters, such as temperature, relative
13 humidity, surface properties, surface-to-volume ratio, the turbulence of airflow, building type,
14 and coexisting pollutants (Lee et al., 1996; Cotterill et al., 1997; Monn et al., 1998; Garcia-Algar
15 et al., 2003; Sorensen et al., 2005; Zota et al., 2005). As a result, using a fixed value, as
16 mentioned above, would either over- or underestimate the true a or Fmf.
17 Although specific P, k, and a were not reported by most studies, a number of studies
18 investigated factors affecting P, k, and a (or indicators of P, k, and a), and their effects on indoor
19 and personal exposures (Lee et al., 1996; Cotterill et al., 1997; Monn et al., 1998; Garcia-Algar
20 et al., 2003; S0rensen et al., 2005; Zota et al., 2005). Garcia-Algar et al. (2003) observed that
21 double-glazed windows had a significant effect on indoor NOz concentrations. Homes with
22 double-glazed windows had lower indoor concentrations (6 ppb lower) than homes with single-
23 glazed windows. Cotterill et al. (1997) reported that having single- or double-glazed windows
24 was a significant factor affecting NOz concentrations in kitchens in homes with gas-cookers
25 (31.4 ppb and 39.8 ppb for homes with single- and double-glazed windows, respectively). The
26 reduction of ventilation resulting from the presence of double-glazed windows can block outdoor
27 N02 from coming into the indoor environment, and at the same time can also increase the
28 accumulation of indoor generated NOz.
29 A similar effect was found for homes using air conditioners. Lee et al. (2002) observed
30 that NOz was 9 ppb higher in homes with an air conditioner than in homes without. The authors
31 also observed that the use of a humidifier would reduce indoor N02 by 6 ppb.
March 2008 2-47 DRAFT-DO NOT QUOTE OR CITE
-------
1 House type was another factor reported affecting ventilation (Lee et al, 1996; Garcia-
2 Algar et al., 2003). Lee et al. (1996) reported that the building type was significantly associated
3 with air exchange rate: the air exchange rate ranged from 1.04 h'1 for single dwelling unit to
4 2.26 h'1 for large multiple dwelling unit. Zota et al. (2005) reported that the air exchange rates
5 were significantly lower in the heating season than the nonheating season (0.49 h'1 for the
6 heating season and 0.85 h'1 for the nonheating season.
7 Although models based on dynamic flow and mass transfer equations might help better
8 simulate indoor and outdoor concentration and personal exposure, in practice, people still rely
9 heavily on Equations 2.5-2 through 2.5-5 because of the lack of real-time measurement data.
10 The assumed equilibrium condition could result in missing the peak exposure and obscuring the
11 real short-term outdoor contribution to indoor and personal exposure. For example, the N02
12 concentrations at locations close to busy streets in urban environments may vary drastically with
13 time. If the measurement is carried out during a non-steady-state period, the indoor/outdoor
14 concentration ratio may indicate either a too low relative importance of indoor sources (if the
15 outdoor concentration is in an increasing phase) or a too high relative importance of indoor
16 resources (if the outdoor concentration is in a decreasing phase) (Ekberg, 1996). As a result, the
17 relationship between P, k, and a has not been thoroughly investigated, but factors mentioned
18 above can significantly affect P, k, and a, and thus affect the relationships between indoor and
19 outdoor N02 concentration and between personal exposure and outdoor N02 concentration. It
20 should also be pointed out that both P and k are functions of the complicated mass transfer
21 processes that occur on indoor surfaces and therefore are associated with air exchange rate,
22 which has an effect on the turbulence of indoor airflows. However, the relationship between P,
23 k, and a has not been thoroughly investigated.
24 Alternatively, the ratio of personal exposure to ambient concentration can be regarded as
25 a in the absence of indoor or nonambient sources. Only a few studies have reported the value
26 and distribution of the ratio of personal N02 exposure to ambient N02 concentration, and even
27 fewer studies have reported the value and distribution of a based on sophisticated study designs.
28 Rojas-Bracho et al. (2002) reported the median personal-outdoor ratio was 0.64 (with an IQR of
29 0.45), but the authors reported that a was overestimated by this ratio because of indoor sources.
30 The random component superposition (RCS) model is an alternative way to calculate Finf
31 or a using observed ambient and personal exposure concentrations (Ott et al., 2000). The RCS
March 2008 2-48 DRAFT-DO NOT QUOTE OR CITE
-------
1 statistical model (shown in Equations 2.5-2 through 2.5-5) uses the slope of the regression line of
2 personal concentration on the ambient N02 concentration to estimate the population averaged
3 attenuation factor and means and distributions of ambient and nonambient contributions to
4 personal N02 concentrations (the intercept of the regression is the averaged nonambient
5 contribution to personal exposure) (U.S. Environmental Protection Agency, 2004). As shown in
6 Table 2.5-5, a calculated by the RCS model ranges from 0.3 to 0.6. Similarly, as shown in Table
7 2.5-6 (see end of chapter), Finf ranges from 0.4 to 0.7.
8 The RCS model calculates ambient contributions to indoor concentrations and personal
9 exposures based on the statistical inferences of regression analysis. However, personal-outdoor
10 regressions could be affected by extreme values (outliers on either the x or the y axis). Another
11 limitation of the RCS model is that this model is not designed to estimate ambient and
12 nonambient contributions for individuals, in part because the use of a single value for a does not
13 account for the large home-to-home variations in actual air exchange rates and penetration and
14 decay rates of N02. In the RCS model, a is also determined by the selection of the predictor.
15 Using residential outdoor N02 concentrations as the model predictor might give a different
16 estimate of a than using ambient N02 because of the spatial variability of N02 mentioned early
17 in this section. As mentioned earlier, personal N02 exposure is affected not only by air
18 infiltrating from outdoors but also by indoor sources (see Section 2.5.5).
19 Nerriere et al. (2005) used data from the Genotox ER study in France (Grenoble, Paris,
20 Rouen, and Strasbourg) and reported that factors affecting the differences between personal
21 exposure to ambient N02 and corresponding ambient monitoring site concentrations were
22 season, city, and land use dependence. During the winter, city and land use categorization
23 account for 31% of the variation, and during the summer, 54% of the variation can be explained
24 by these factors. When data from the ambient monitoring site were used to represent personal
25 exposures, the largest difference between ambient and personal exposure was found at the
26 "proximity to traffic" site, while the smallest difference was found at the "background" site.
27 When using data from the urban background site, the largest difference was observed at the
28 "industry" site, and the smallest difference was observed at the background site, which reflected
29 the heterogeneous distribution of N02 in an urban area. During winter, differences between
30 ambient site and personal exposure concentrations were larger than those in the summer.
March 2008 2-49 DRAFT-DO NOT QUOTE OR CITE
-------
1 In summary, N02 is monitored at far fewer sites than either Os or PM. Significant spatial
2 variations in ambient N02 concentrations were observed in urban areas. Measurements of N02
3 are subject to artifacts both at the ambient level and at the personal level. Personal exposure to
4 ambient and outdoor N02 is determined by many factors as listed in Sections 2.5.1 and 2.5.2.
5 These factors all influence the contribution of ambient N02 to personal exposures. Personal
6 activities determine when, where, and how people are exposed to N02. The variations of these
7 physical and exposure factors determine the strength of the association between personal
8 exposure and ambient concentrations in both longitudinal and cross-sectional studies. In Section
9 2.5.6.1, three types of correlation coefficients were presented. The observed strength of the
10 association between personal exposures and ambient concentrations are not only affected by the
11 variation in physical parameters (e.g., P, k, a and indoor sources) but also affected by data
12 quality and study design. The association between the ambient component of personal exposures
13 and ambient concentrations is more relevant to the interpretation of epidemiologic evidence but
14 this type of correlation coefficient is not reported. Therefore, the weak association between
15 personal total exposures and ambient concentrations in some longitudinal studies might not
16 reflect the true association between the ambient component of personal exposures and ambient
17 concentrations. In the absence of indoor and local sources, personal exposures to N02 are
18 between the ambient level and the indoor level. However, personal exposures could be much
19 higher than either indoor or outdoor concentrations in the presence of these sources. A number
20 of studies found that personal N02 was associated with ambient N02, but the strength of the
21 association ranged from poor to good.
22 Some researchers concluded that ambient N02 may be a reasonable proxy for personal
23 exposures, while others noted that caution must be exercised if ambient N02 is used as a
24 surrogate for personal exposure. Reasons for the differences in study results are not clear, but
25 are related in large measure to differences in study design, to the spatial heterogeneity of N02 in
26 study areas, to control of indoor sources, to the seasonal and geographic variability in the
27 infiltration of ambient N02, and to differences in the time spent in different microenvironments.
28 Measurement artifacts at the ambient and personal levels and differences in analytical
29 measurement capabilities among different groups could also have contributed to the mixed
30 results. The collective variability in all of the above parameters, in general, contributes to
31 exposure misclassification errors in air pollution-health outcome studies.
March 2008 2-50 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.5.8 NO2 as a Component of Mixtures
2
3 2.5.8.1 Correlations between Ambient NOi and Ambient Copollutants
4 Relationships between ambient concentrations of N02 and other pollutants that are
5 emitted by the same sources, such as motor vehicles, should be evaluated in designing and
6 interpreting air pollution-health outcome studies, as ambient concentrations are generally used to
7 reflect exposures in epidemiologic studies. Thus, the majority of studies examining pollutant
8 associations in the ambient environment have focused on ambient N02, PM2.5 (and its
9 components), and CO, with fewer studies reporting the relationship between ambient N02 and
10 ambient Os or S02.
11 Data were compiled from EPA's AQS and a number of exposure studies. Correlations
12 between ambient concentrations of N02 and other pollutants, PM2.5 (and its components, where
13 available), CO, 03, and S02 are summarized in Table 2.5-7.
14 Mean values of correlations between monitoring sites are shown. As can be seen from
15 the table, N02 is moderately correlated with PM2.5 (range: 0.37 to 0.78) and with CO (0.41 to
16 0.76) in suburban and urban areas. At locations such as Riverside, CA, associations between
17 ambient N02 and ambient CO concentrations (both largely traffic-related pollutants) are much
18 lower, likely as the result of other sources of both CO and N02 increasing in importance in going
19 from urban environments to more rural and sparsely populated areas. These sources include
20 oxidation of methane (CH-i) and other biogenic compounds; residential wood burning and
21 prescribed and wild land fires for CO; and soil emissions, lightning, and residential wood
22 burning and wild land fires for N02. In urban areas, the ambient N02-C0 correlations vary
23 widely. The strongest correlations are seen between N02 and elemental carbon (EC). Note that
24 the results of Hochadel et al. (2006) for PM2.5 optical absorbance have been interpreted in terms
25 of EC. Correlations between ambient N02 and ambient Os are mainly negative, owing to the
26 chemical interaction between the two, with again considerable variability in the observed
27 correlations. Only one study (Sarnat et al., 2001) examined associations between ambient N02
28 and ambient S02 concentrations, and it showed a negative correlation during winter.
29 Figures 2.5-5a-d show seasonal plots of correlations between N02 and 03 versus
30 correlations between N02 and CO. As can be seen from the figures, N02 is positively correlated
31 with CO during all seasons at all sites. However, the sign of the correlation of N02 with 03
March 2008 2-51 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-7. PEARSON CORRELATION COEFFICIENT BETWEEN
AMBIENT NO2 AND AMBIENT COPOLLUTANTS
1
2
3
4
Study (Ambient)
This Assessment
This Assessment
This Assessment
This Assessment
Kim et al. (2006)
Sarnat et al. (2006)
Sarnat et al. (2006)
Connell et al. (2005)
Kim et al. (2005)
Sarnat et al. (200 1)1
Sarnat et al. (2001)
Hochadel et al. (2006)
Hazenkamp-von Arx et al.
(2004)
Cyiys et al. (2003)
Mosqueron et al. (2002)
Rojas-Bracho et al. (2002)
Location
Los Angeles, CA
Riverside, CA
Chicago, IL
New York, NY
Toronto, Canada
Steubenville, OH
(autumn)
Steubenville, OH
(summer)
Steubenville, OH
St. Louis, MO (RAPS)
Baltimore, MD (summer)
Baltimore, MD (winter)
Ruhr area, Germany
21 European cities
Ehrfurt, Germany
Paris, France
Santiago, Chile
PM25
0.49 (u2)
0.56 (s)
0.49 (s)
0.58 (u)
0.44
0.78
(0.70 for sulfate
0.82 for EC)
0.00
(0.1 for sulfate
0.24 for EC)
0.50
0.37
0.75
0.41
(0.93 for EC3)
0.75
0.50
0.69
0.77
CO O3 SO2
0.59 (u) -0.29 (u)
0.64 (s) -0.1 1 (s)
0.43 (u) 0.045 (u)
0.41 (s) 0.10 (s)
0.15 (r) -0.31 (r)
°'53 (U) -0 20 (u)
0.46 (s) l '
0.46 (u) -0.06 (u)
0.72
0.644
0.75 0.02
not significant
0.76 -0.71 -0.17
0.74
1 Spearman correlation coefficient was reported.
2 u: urban; s: suburban; and r: rural
3 Inferred based on EC as dominant contributor to PM2.5 absorbance.
4 Value with respect to NOX.
varies with season, ranging from negative during winter to slightly positive during summer.
There are at least two main factors contributing to the observed seasonal behavior. Ozone and
radicals correlated with it tend to be higher during the summer, thereby tending to increase the
ratio of N02 to NO. Nitrogen oxide compounds formed by further oxidation of NOX are also
March 2008
2-52
DRAFT-DO NOT QUOTE OR CITE
-------
o
*CS|
O
^
CO
o
CM
o
Winter
0.8-
0.6-
0.4-
0.2-
-0.8 -0.6 -0.4 -0.2 (
-0.2-
-0.4 •
-0.6-
-0.8-
0.2 0.4 0.6 0.8
• #*++}
* **
NO2: CO
Spring
0.8-
0.6-
0.4-
0.2-
1 -0.8 -0.6 -0.4 -0.2
% * ++1 +
I Q^ o5 0.6 Js
O
N
o
z
CO
o
o-
Summer
0.8-
0.6-
0.4-
0.2-
1 -0.8 -0.6 -0.4 -0.2 [
-0.2-
-0.4-
-0.6-
-0.8-
/ * t
) 0.2 0.4 0.6 0.8
NO2: CO
Fall
0.8-
0.6-
0.4-
0.2-
1 -0.8 -0.6 -0.4 -0.2 (
* ** \ *
) 0.2 0.4 ^0.6 *.8
NO2: CO
NO,: CO
Figure 2.5-5a-d. Correlations of NOi to Os versus correlations of NOi to CO for Los
Angeles, CA (2001-2005).
1 expected to be correlated with Os and increased summertime photochemical activity. Because
2 some of these additionally oxidized N compounds create a positive artifact in the FRM for NOz,
3 they may also tend to increase the correlation of NOz with Os during the warmer months.
4 A number of case studies show similar correlations between ambient N02 and other
5 pollutants presented above. Particulate and gaseous copollutant data were analyzed at 10 sites in
6 the St. Louis Regional Air Pollution Study (RAPS) dataset (1975, 1977) by Kim et al. (2005).
March 2008
2-53
DRAFT-DO NOT QUOTE OR CITE
-------
1 This study examined the spatial variability in source contributions to PM2.s. Table 2.5-8 shows
2 correlations between NOx and traffic pollutants measured in ambient air.
TABLE 2.5-8. PEARSON CORRELATION COEFFICIENT BETWEEN NOX
AND TRAFFIC-GENERATED POLLUTANTS
NOX:
NOX:
NOX:
NOX:
PM2.5 (MV component)
CO
Pb
Br
0,
0,
0,
0,
,48
-------
1 that between N02 and PM2.5 (r = 0.55) and that between N02 and PMio (r = 0.45). A time-series
2 mortality study (Wichmann et al, 2000; re-analysis by Stb'lzel et al., 2003) conducted in Erfurt,
3 Germany, measured, and analyzed UFP number and mass concentrations as well as N02. Unlike
4 Seaton and Dennekamp's data, in this data set, the correlation between N02 and various number
5 concentration indices were not much stronger than those between PM2.5 and number
6 concentration indices or those between PMio and number concentration indices. For example,
7 the correlation between NCo.oi-o.io (particle number concentration for particle diameter between
8 10 and 100 nm) and N02, PM2.5, and PMio were 0.66, 0.61, and 0.61, respectively.
9
10 2.5.8.2 Correlations of Personal and Ambient NOi and Personal and Ambient
11 Copollutants
12 Correlations between ambient concentrations of N02 and personal copollutants, PM2.s
13 (and its components where available), CO, Os, and S02 are summarized in Table 2.5-9.
TABLE 2.5-9. PEARSON CORRELATION COEFFICIENT BETWEEN
AMBIENT NO2 AND PERSONAL COPOLLUTANTS
Study
Sarnat et al.
(2006)
Sarnat et al.
(2006)
Vinzents et al.
(2005)
Location PM2.S
Steubenville, OH 0.71
Fall
Steubenville, OH 0.00
Summer
Copenhagen, —
Denmark
Sulfate EC
0.52 0.70
0.1 not 0.26
significant
— —
Ultrafine Particle
_
0.49 (R2) explained by ambient N02
and ambient temperature
14 Correlations between personal concentrations of N02 and ambient copollutants, PM2.5
15 (and its components where available), CO, Os, and S02 are summarized in Table 2.5-10, and
16 correlations between personal N02 concentrations and personal copollutant concentrations are
17 shown in Table 2.5-11.
18 Most studies examined above show that personal N02 concentrations are significantly
19 correlated with either ambient or personal level PM2.5 or other combustion-generated pollutants,
20 e.g., CO, EC.
March 2008
2-55
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-10. PEARSON CORRELATION COEFFICIENT BETWEEN
PERSONAL NO2 AND AMBIENT COPOLLUTANTS
Study
Sarnat et al.
(2006)
Sarnat et al.
(2006)
Kim et al.
(2006)
Rojas-Bracho et al. (2002)
Location
Steubenville, OH
Fall
Steubenville, OH
Summer
Toronto, Canada
Santiago, Chile
PM25
0.46
0.00
0.30
0.65
Sulfate EC PM10 CO
0.35 0.57 — —
0.1 0.17 — —
not significant
— — — 0.20
— — 0.39 —
TABLE 2.5-11. PEARSON CORRELATION COEFFICIENT BETWEEN
PERSONAL NO2 AND PERSONAL COPOLLUTANTS
Study
Kim et al.
(2006)
Modig et al.
(2004)
Mosqueron et al.
(2002)
Jarvis et al.
(2005)
Lee et al.
(2002)
Lai et al.
(2004)
Location PM2S CO VOCs HONO
Toronto, 0.41 0.12 —
Canada
Umea, — — 0.06 for —
Sweden 1,3-butadiene;
0.10 for benzene
Paris, France 0.1 2 but not — — —
significant
21 European — — — 0.77 for indoor N02 and
cities indoor HONO
— — — — 0.51 for indoor N02 and
indoor HONO
Oxford, -0.1 0.3 -0. 1 1 for TVOCs —
England
1 As might be expected from a pollutant having a major traffic source, the diurnal cycle of
2 N02 in typical urban areas is characterized by traffic emissions, with peaks in emissions
3 occurring during morning and evening rush hour traffic. Motor vehicle emissions consist mainly
4 of NO, with only -10% of primary emissions in the form of NOz. The diurnal pattern of NO and
5 NOz concentrations are also strongly influenced by the diurnal variation in the mixing layer
March 2008
2-56
DRAFT-DO NOT QUOTE OR CITE
-------
1 height. Thus, during the morning rush hour when mixing layer heights are still low, traffic
2 produces a peak in NO and NOz concentrations. As the mixing layer height increases during the
3 day, dilution of emissions occurs, and NO and N02 are converted to NOz. During the afternoon
4 rush hour, mixing layer heights are often still at or near their daily maximum values, resulting in
5 dilution of traffic emissions through a larger volume than in the morning. Starting near sunset,
6 the mixing layer height drops and conversion of NO to NOz occurs without subsequent
7 photolysis of NOz recreating NO.
8 The composite diurnal variability of N02 in selected urban areas with multiple sites
9 (New York, NY, Atlanta, GA, Baton Rouge, LA, Chicago, IL, Houston, TX, Riverside, CA, and
10 Los Angeles, CA) is shown in Figure 2.5-6. Figure 2.5-6 shows that lowest hourly median
11 concentrations are typically found at around midday and that highest hourly median
12 concentrations are found either in the early morning or in mid-evening. Median values range by
13 about a factor of two from -13 ppb to -25 ppb. However, individual hourly concentrations can
14 be considerably higher than these typical median values, and hourly NOz concentrations of >0.10
15 parts per million (ppm) can be found at any time of day. The diurnal pattern in median
16 concentrations shown in Figure 2.5-6 is consistent with that shown in Figures 2.4-5 and 2.4-6 for
17 Atlanta, indicating some commonality in sources across these cities. The pattern in the median
18 concentrations is consistent with traffic as the major source of variability. However, the patterns
19 in the upper end of the concentration distribution differ between cities and the composite,
20 indicating that other sources and meteorological processes affect N02 levels, causing them to
21 differ from city to city.
22 Information concerning the seasonal variability of ambient NOz concentrations is given
23 in the Annex in Section AX3.3. NOz levels are highest during the cooler months of the year and
24 still show positive correlations with CO. Mean NOz levels are lowest during the summer
25 months, though of course, there can be large positive excursions associated with the development
26 of high-pressure systems. In this regard, N02 behaves as a primary pollutant, although there is
27 no good reason to suspect strong seasonal variations in its emissions.
28
29 2.5.8.3 Associations among NOi and Other Pollutants in Indoor Environments
30 In addition to N02, indoor combustion sources such as gas ranges and unvented gas
31 heaters emit other pollutants that are present in the fuel or are formed during combustion. The
March 2008 2-57 DRAFT-DO NOT QUOTE OR CITE
-------
I
0>
o
o
o
0.20-
0.19-
0.18-
0.17-
0.16-
0.15-
0.14-
0.13-
0.12-
0.11-
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0,02-
0.01-
0.00-
x
X
012345
t
6
i • i
89
i ' i • i • i ' i ' i • i ' i ' i ' i • i ' i ' i • i ' i
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
Figure 2.5-6. Composite, diurnal variability in 1-h average NO2 in urban areas. Values
shown are averages from 2003 through 2005. Boxes define the interquartile
range, and the whiskers the 5th and 95th percentile values. X's denote
individual values above the 95th percentile.
1 major products from the combustion of natural gas are carbon dioxide (C02) and CO followed
2 by formaldehyde (HCHO) with smaller amounts of other oxidized organic compounds in the gas
3 phase. PM, especially in the ultrafine-size range and HONO are also emitted. The production of
4 pollutants by reactions of N02 in indoor air was covered in Section 2.5.5.
5
6 2.5.5.3.7 NO and HONO
7 Dennekamp et al. (2001) measured levels of NO, N02, and UFPs generated by gas and
8 electric cooking ranges in a test laboratory room. They found average levels of NO ranging from
9 -500 to -3,000 ppb, with peak (15-min average) levels ranging from -1,000 to -6,000 ppb
10 depending on how many burners (1 to 4) were turned on and for how long (15 min to 2 h).
11 Corresponding levels of N02 tracked those of NO but were typically factors of 2 to 5 lower.
12 Spicer et al. (1993) compared the measured increase in HONO in a test house resulting from
13 direct emissions of HONO from a gas range and from production by surface reactions of N02.
14 They found that emissions from the gas range could account for -84% of the measured increase
March 2008
2-58
DRAFT-DO NOT QUOTE OR CITE
-------
1 in HONO. In a study of homes in southern California, Lee et al. (2002) found that indoor levels
2 of N02 and HONO were positively associated with the presence of gas ranges.
3
4 2.5.8.3.2 Carbon-Containing Gaseous Pollutants
5 In a study of pollutants emitted by unvented gas heaters, Brown et al. (2004) found that
6 CO in a room test chamber ranged from 1 to 18 ppm and NOz, from 100 to 300 ppb.
7 Corresponding levels of HCHO were highly variable, ranging from <10 ppb to a few hundred
8 ppb (with an outlier at >2 ppm).
9
10 2.5.8.3.3 PM
11 PM in the sub-micrometer size range is also produced during natural gas combustion.
12 Dennenkamp et al. (2001) in the study mentioned above found enhancements in UFP
13 concentrations when gas burners were turned on. Peak (15-min average) concentrations for
14 different experiments ranged from -140,000 to ~ 400,000/cm3 corresponding to average levels of
15 -80,000 to 160,000/cm3. Concentrations before the experiments were begun were in the range
16 of a few thousand per cm3. However, Ristovski et al. (2000) measured emission rates for
17 individual particles, which are expected to be present mainly in the UFP size range but
18 concluded that these rates are low, and they could not detect an increase in particle number from
19 one of the two heater models tested.
20 Rogge et al. (1993) found that at least 22% of the fine particle mass emitted by natural
21 gas heaters consists of PAHs, oxy-PAHs, and aza-and thia-arenes. They also identified
22 emissions of speciated alkanes, w-alkanoic acids, polycyclic aromatic ketones, and quinones.
23 However, these accounted for only another -4% of the emitted fine PM. Although the PM
24 emissions rates were low and not likely to affect PM levels, the PAH content of natural gas
25 combustion emissions in this study indicates that natural gas combustion could be a significant
26 source of PAHs in indoor environments
27
28
29 2.6 DOSIMETRY OF INHALED NITROGEN OXIDES
30 This section provides a brief overview of N02 dosimetry and updates information
31 provided in the 1993 AQCD for Oxides of Nitrogen. A more extensive discussion of N02
32 dosimetry appears in Annex 4. NOz, classified as a reactive gas, interacts with surfactants,
March 2008 2-59 DRAFT-DO NOT QUOTE OR CITE
-------
1 antioxidants, and other compounds in the epithelial lining fluid (ELF). The compounds thought
2 to be responsible for adverse pulmonary effects of inhaled N02 are the reaction products
3 themselves or the metabolites of these products in the ELF.
4 Acute N02 uptake in the lower respiratory tract is thought to be rate-limited by chemical
5 reactions of N02 with ELF constituents rather than by gas solubility in the ELF (Postlethwait and
6 Bidani, 1990). Postlethwait and Bidani (1994) concluded that the reaction between N02 and
7 water does not significantly contribute to the absorption of inhaled N02. Rather, uptake is a
8 first-order process for N02 concentrations of < 10 ppm, is aqueous substrate-dependent, and is
9 saturable. Postlethwait et al. (1991) reported that inhaled N02 (<10 ppm) does not penetrate the
10 ELF to reach underlying sites and suggested that cytotoxicity may be due to N02 reactants
11 formed in the ELF. Related to the balance between reaction product formation and removal, it
12 was further suggested that cellular responses may be nonlinear with greater responses being
13 possible at low levels of NO2 uptake versus higher levels of uptake.
14 Glutathione (GSH) and ascorbate are the primary N02 absorption substrates in rat ELF
15 (Postlethwait et al., 1995). Velsor and Postlethwait (1997) investigated the mechanisms of acute
16 epithelial injury from N02 exposure. Membrane oxidation was not a simple monotonic function
17 of GSH and ascorbic acid levels. The maximal levels of membrane oxidation were observed at
18 low antioxidant levels versus null or high antioxidant levels. GSH- and ascorbic acid-related
19 membrane oxidation were superoxide- and hydrogen peroxide-dependent, respectively. The
20 authors suggested that increased absorption of N02 occurred at the higher antioxidant
21 concentrations, but little secondary oxidation of the membrane occurred because the reactive
22 species (e.g., superoxide and hydrogen peroxide) generated during absorption were quenched. A
23 lower rate of N02 absorption occurred at the low antioxidant concentrations, but oxidants were
24 not quenched and so were available to interact with the cell membrane. Illustrating the complex
25 interaction of antioxidants, some studies suggest that N02-oxidized GSH may be again reduced
26 by uric acid and/or ascorbic acid (Kelly et al., 1996; Kelly and Tetley, 1997).
27 Very limited work related to the quantification of N02 uptake has been reported since the
28 1993 AQCD for Oxides of Nitrogen. In both humans and animals, the uptake of N02 uptake by
29 the upper respiratory tract decreases with increasing ventilator rates. This causes a greater
30 proportion of inhaled N02 to be delivered to the lower respiratory tract. In humans, the
31 breathing pattern shifts from nasal to oronasal during exercise relative to rest. Since the nasal
March 2008 2-60 DRAFT-DO NOT QUOTE OR CITE
-------
1 passages absorb more inhaled N02 than the mouth, exercise (with respect to the resting state)
2 delivers a disproportionately greater quantity of the inhaled mass to the pulmonary region of the
3 lung, where the N02 is readily absorbed. Bauer et al. (1986) reported a statistically significant
4 increase in uptake from 72% during rest to 87% during exercise in a group of 15 asthmatic
5 adults. The minute ventilation also increased from 8.1 L/min during rest to 30.4 L/min during
6 exercise. Hence, exercise increased the dose rate of N02 by 5-fold in these subjects. Similar
7 results have been reported for beagle dogs where the dose rate of N02 was 3-fold greater for the
8 dogs during exercise than rest (Kleinman and Mautz, 1991).
9 Modeling studies also predict that the net N02 dose (N02 flux to air-liquid interface) is
10 relatively constant from the trachea to the terminal bronchioles and then rapidly decreases in the
11 pulmonary region. The pattern of net N02 dose rate or uptake rate is expected to be similar
12 between species and unaffected by age in humans. The predicted tissue dose and dose rate of
13 N02 (N02 flux to liquid-tissue interface) is low in the trachea, increases to a maximum in the
14 terminal bronchioles and the first generation of the pulmonary region, and then decreases rapidly
15 with distal progression. The site of maximal N02 tissue dose is predicted to be fairly similar
16 between species, ranging from the first generation of respiratory bronchioles in humans to the
17 alveolar ducts in rats. The production of toxic N02 reactants in the ELF and the movement of
18 these reactants to the tissues have not been modeled.
March 2008 2-61 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2.5-4A. ASSOCIATION BETWEEN PERSONAL EXPOSURE AND AMBIENT CONCENTRATION
»— t
o
tr
t-O
O
O
OO
t-O
1
fO
O
>
H
O
O
,0
c
O
m
0
0
H
m
Mean
Concentration Association
Study Study Design (ppb)
Linn et al. Type: Longitudinal; Location: Southern California Ambient:
(1996) Subjects: 269 school children 37
Time period: fall, winter, spring, 1992-1994
Method: 24-h avg, 1-wk consecutive measurement for Personal:
each season for each child. 22
Aim et al. Type: Longitudinal; Location: Helsinki, Finland Ambient:
(1998) Subjects: 246 children aged 3-6 yrs old 16.8-26.3
Time period: winter and spring, 1991
Method: 1 -wk averaged sample for each person, Personal:
6 consecutive wks in the winter and 7 consecutive 9-16.6
wks in the spring.
r o
Variable
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Personal vs. central
Location
Pooled
Downtown
Suburban
Downtown
Suburban
Downtown
(electric stove
home)
Downtown
(gas stove home)
Suburban
(electric stove
home)
Downtown
(non-smoking
home)
Downtown
(smoking home)
Suburban
(non-smoking
home)
Suburban
(smoking home)
Pooled
Season
Pooled
Spring
Spring
Winter
Winter
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
rp, r,, or R2
0.63 (rp) (n = 107)
0.64 (rp) p < 0.001
0.78 (rp) p < 0.001
- 0.06 (rp)p> 0.05
0.32 (rp) p > 0.05 (
0.42 (rp) p < 0.01 (
0.16 (rp)p>0.01 (
0.55 (rp) p < 0.001
0.47 (rp) p < 0.001
0.23 (rp) p > 0.01 (
0.53 (rp) p < 0.001
0.52 (rp) p < 0.001
0.37 (R2) (n = 24)
(n=NR**)
(n = NR)
(n = NR)
n = NR)
n = NR)
n = NR)
(n = NR)
(n = NR)
n = NR)
(n = NR)
(n = NR)
-------
TABLE 2.5-4A (cont'd). ASSOCIATION BETWEEN PERSONAL EXPOSURE AND AMBIENT CONCENTRATION
t-O
o
o
oo
t-o
I
CO
O
H
6
O
2
o
m
o
?d
o
K^
H
m
Study
Liard et al.
(1999)
Linaker et al.
(2000)
Study Design
Type: Daily avg/cross-sectional; Location: Paris, France
Subjects: 55 adults and 39 children
Time period: May -June 1996
Method: three 4-day avg measurements for each person,
during each measurement session, all subjects were
measured at the same time.
Type: Longitudinal; Location: Southampton,
Hampshire, UK
Subjects: 114 asthmatic children, aged 7-12
Time period: Oct 1994 to Dec 1995
Method: at least 16 consecutive samples (1-wk avgs) for
each child (mean duration of follow-up: 32 wks).
Mean
Concentration
(Ppb)
Ambient:
26.3-36.8
Personal:
15.8-26.3
Ambient:
6.5
Personal:
8.9
Association
Variable
Adults vs. central
Children vs. central
Personal vs. central
(overall
measurements across
children and time)
Location
Urban
Urban
Pooled, urban, no
major indoor
sources
Season rp, r^ or R2
Summer 0.41 (R2)
p < 0.0001 (n = NR)
Summer 0.17 (R2) p = 0.0004 (n = NR)
Pooled Not significant
(n = NR)
Personal vs. central
(subject-wise)
By person
Gauvin et al. Type: Daily avg/cross-sectional; Location: three French Ambient:
(2001) metropolitan areas 10.2-25.7
Subjects: 73 children
Time period: Apr-June 1998 in Grenoble Personal:
May-June 1998 in Toulouse; June-Oct 1998 in Paris 13.2-17
Method: one 48-h avg measurement for each child; all
children in the same city were measured on the same
day.
Piechocki- Type: Pooled; Location: Lille (northern France) Ambient:
Minguyetal. Subjects: 13 participants in the first campaign, and 31 15.8-57.9
(2006) participants in the second campaign
Time period: winter 2001 (first campaign); summer Personal:
2002 (second campaign) 8.9-20.0
Method: two 24-h sampling periods (one on workdays;
one on weekends) for each subject in each campaign;
during each sampling period, each subject received
four samplers to measure personal exposure in four
different microenvironments (home, other indoor
environment, transport, and outdoors).
Personal vs. central Urban
(Grenoble)
Personal vs. central Urban
(Toulouse)
Personal vs. central Urban
(Paris)
Personal (exposure at Urban
home) vs. central
Personal (exposure at Urban
home) vs. central (electric stove and
electric heater
home)
Pooled - 0.77 to 0.68 and median
-0.02 (rp)
(n = NR)
Pooled 0.01 (R2) (n = NR)
Pooled 0.04 (R2) (n = NR)
Pooled 0.02 (R2) (n = NR)
Pooled 0.09 (R2) p = 0.0101 (n = NR)
Summer 0.61 (R2)
p = 0.0001 (n = NR)
-------
TABLE 2.5-4A (cont'd). ASSOCIATION BETWEEN PERSONAL EXPOSURE AND AMBIENT CONCENTRATION
1— i
o
t-o
o
o
oo
t-o
O
J>
'Tl
H
b
0
Study
Kim et al.
(2006)
Sarnat et al.
(2000)
Sarnat et al.
(2001);
Koutrakis
et al. (2005)
Sarnat et al.
(2005);
Koutrakis
et al. (2005)
Study Design
Type: Longitudinal; Location: Toronto, Canada
Subjects: 28 adults with coronary artery disease
Time period: Aug 1999 to Nov 2001
Method: 1 day/wk, 24-h avg, for a max of 10 wks for
each person.
Type: Longitudinal; Location: Baltimore, MD
Subjects: 20 senior, healthy, non-smoking people
(average age 75)
Time period: summer of 1998; winter of 1999
Method: 1 day averaged sample, for 12 consecutive days
for each subject; four to six subjects were measured
concurrently during each 12-day monitoring period.
Type: Longitudinal; Location: Baltimore, MD
Subjects: 56 seniors, schoolchildren, and people with
COPD
Time period: summer of 1998 and winter of 1999
Method: 14 of 56 subjects participated in both sampling
seasons; all subjects were monitored for 12
consecutive days (24-h avg samples) in each of the
one or two seasons, except children, who were
measured for 8 consecutive days during the summer.
Type: Longitudinal; Location: Boston, MA
Subjects: 43 seniors and schoolchildren
Time period: summer of 1999; winter of 2000
Method: Similar study design as Sarnat et al. (2001).
Mean
Concentration
(Ppb)
Ambient:
24
Personal:
Ambient:
21.4-39.2
Personal:
7.9-42.7
Ambient:
20-25
Personal:
10-15
Ambient:
21.1-32.6
Personal:
10.6-29.6
Association
Variable Location
Personal vs. central Urban
(subject wise)
Personal vs. central Urban
(subject wise)
Personal vs. central Urban
(subject wise)
Personal vs. central Urban
(subject wise)
Season rp, r^ or R2
Pooled - 0.36 to 0.94 (rs) with a
median of 0.57 (15 subjects)
Summer -0.63 to 0.75 (rs) with a
median of
-0.01 (14 subjects)
Winter - 0.64 to 0.74 (rs) with a
median of
-0.01 (14 subjects)
Summer -0.45 to 0.85 (rs) with a
median of 0.05* (24 subjects)
Winter - 0.6 to 0.75 (rs) with a median
of 0.05* (45 subjects)
Summer -0.25 to 0.5 (rs) with a median
of 0.3* (n = NR) Slope = 0.19
0.08-0.30
Winter -0.5 to 0.9 (rs) with a median
of0.4*(n = NR)
O
o
o
H
m
o
o
H
m
Sarnat et al. Type: Longitudinal; Location: Steubenville, OH Ambient:
(2006) Subjects: 15 senior subjects 9.5-11.3
Time period: summer and fall of 2000
Method: two consecutive 24-h samples were collected Personal:
for each subject for each wk, 23 wks total 9.9-12.1
Personal vs. central Urban
Slope = -0.03
-0.21-0.15
Summer 0.14 (R2)
(n = 122) p< 0.05
Fall 0.43 (R2)
p < 0.05 (n = 138)
: Values were estimated from figures in the original paper.
:*NR: Not Reported.
-------
TABLE 2.5-4B. ASSOCIATION BETWEEN PERSONAL EXPOSURE AND OUTDOOR CONCENTRATION
»— t
O
tr
t-O
C~^ 1
O
oo
t-O
i
01
O
;>
H
1
O
0
2
O
H
O
m
o
o
m
Study Study Design
Kramer et al. Location: Germany; Subjects: 191 children
(2000) Time period: Mar and Sep 1996
Method: two 1-wk averaged measurements for each
child in each mo.
Rojas-Bracho Location: Santiago, Chile; Subjects: 20 children
et al. (2002) Time period: winters of 1998 and 1999
Method: five 24-h avg samples for 5 consecutive days for
each child.
Raaschou-Nielsen Location: Copenhagen, Denmark and rural areas; Subjects:
etal. (1997) 204 children
Time period: Oct 1994, Apr, May, and June 1995
Method: two 1-wk avg measurements for each child in each
mo.
Aim et al. (1998) Location: Helsinki, Finland; Subjects: 246 children aged 3-6
yrs old
Time period: winter and spring of 1991
Method: 1-wk averaged sample for each person for 6
consecutive wks in the winter and 7 consecutive wks in
the spring.
Association Variable
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Personal vs. outdoor
Location
Pooled
Urban
Urban
Urban
Rural
Downtown
Suburban
Downtown
Suburban
Downtown
(electric stove home)
Downtown
(gas stove home)
Suburban
(electric stove home)
Downtown (non-
smoking home)
Downtown (smoking
home)
Suburban (non-
smoking home)
Suburban (smoking
home)
Pooled
Season
Pooled
Pooled
Winter
Pooled
Pooled
Winter
Winter
Spring
Spring
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
Pooled
rp,
0.37 (rp)
0.06 (rp)
0.27 (R2)
0.15 (R2)
0.35 (R2)
0.46 (rp)
0.49 (rp)
0.80 (rp)
0.82 (rp)
0.55 (rp)
0.59 (rp)
0.63 (rp)
0.73 (rp)
(n = NR)
0.51 (rp)
(n = NR)
0.59 (rp)
(n = NR)
0.46 (rp)
(n = NR)
0.86 (R2)
(n = 23)
, rs, orR2
(n = 281)
(n=182)
(n = 87)
(n = 97)
(n = 99)
(n = NR)
(n = NR)
(n = NR)
(n = NR)
(n = NR)
(n = NR)
(n = NR)
-------
TABLE 2.5-4B (cont'd). ASSOCIATION BETWEEN PERSONAL EXPOSURE AND OUTDOOR CONCENTRATION
»— t
o
tr
t-o
O
O
OO
Study Study Design Association Variable Location Season rp, rs, or R2
Monn et al. Location: Geneva, Basel, Lugano, Aarau, Wald, Payerne, Personal vs. outdoor Pooled Pooled 0.33 (R2)
(1 998) Montana, and Davos (S APALDIA study, Switzerland) (n = 1 ,494)
Subjects: 140 subjects
Time period: Dec 1993 to Dec 1994
Method: each home was monitored for 3 periods of 1 mo; in
the 1st wk of each period, personal, indoor rand outdoor
levels were measured, and in the next 3 consecutive wks,
only outdoor levels were measured (1-wk averaged
measurement) .
O
H
6
O
2
o
H
m
o
?d
o
t—t
H
m
Levy et al.
(1998b)
Kodama et al.
(2002)
Spengler et al.
(1994)
Lai et al. (2004)
Location: 18 cities across 15 countries Personal vs. outdoor Urban
Subjects: 568 adults
Time period: Feb or Mar 1996
Method: one 2-day avg measurement for each person, all
people were measured on the same winter day.
Location: Tokyo, Japan Personal vs. outdoor Urban
Subjects: 150 junior-high school students and their family
member!> , , , Personal vs. outdoor Urban
lime period: Feb 24-26, June 2-4, July 13-15, and uct 14-
16 in 1998 and Jan 26-28 in 1999
Method: 3-day avg, personal exposures were monitored on
the same day.
Location: Los Angeles Basin, CA Personal vs. outdoor Pooled
Subjects: probability-based sample, 70 subjects
Time period: May 1987 to May 1988
Method: each participant was monitored during each of 8
cycles (48-h avg sampling period) throughout the yr in
the microenvironmental component of the study.
Location: Oxford, England Personal vs. outdoor Urban
Subjects: 50 adults
Time period: Dec 1998 to Feb 2000
Method: one 48-h avg measurement per person.
Winter 0.57
(n =
Summer 0.24
(n =
Winter 0.08
(n =
V
Pooled 0.48
(n =
Pooled 0.41
(n =
(r,)
546)
(rp)
NR)
(rp)
NR)
(R2)
NR)
(rp)
NR)
' Values were estimated from figures in the original paper.
'* NR: Not Reported.
-------
TABLE 2.5-5. SUMMARY OF REGRESSION MODELS OF PERSONAL EXPOSURE
TO AMBIENT/OUTDOOR NO2
t-O
o
o
oo
Study
Location
Season
Model Type
Slope
(SE) Intercept / ppb
R2
Rojas-Bracho et al.
(2002)
Aim et al.
(1998)
Monn et al.
(1998)
O
H
b
o
2
o
m
o
?a
o
t—t
H
m
Levy et al.
(1998b)
Spengler et al.
(1994)
Location: Santiago, Chile Winter
Subjects: 20 children
Time period: winters of 1998 and 1999
Method: five, 24-h avg samples on consecutive
days for each child.
Location: Helsinki, Finland Winter
Subjects: 246 children aged 3-6 yrs Spring
Time period: winter and spring of 1991
Method: 1-wk averaged sample for each person,
6 consecutive wks in the winter and 7 consecutive wks in the spring.
Location: Geneva, Basle, Lugano, Aarau, Wald, Payerne, Montana, and All
Davos (SAPALDIA study, Switzerland)
Subjects: 140 subjects
Time period: Dec 1993 to Dec 1994
Method: each home was monitored for 3 periods of 1 mo; in the 1st wk of
each period, personal, indoor rand outdoor levels were measured, and in the
next 3 consecutive wks, only outdoor levels were measured (1-wk averaged
measurement).
Location: 18 cities across 15 countries Winter
Subjects: 568 adults
Time period: Feb or Mar 1996
Method: One, 48-h avg measurement for each person, all people were
measured on the same day.
Location: Los Angeles Basin All
Subjects: probability-based sample, 70 subjects
Time period: May 1987 to May 1988
Method: in the microenvironmental component of the study, each participant
was monitored for 48 h during each of 8 sampling cycles throughout the yr.
Personal vs. outdoor 0.33 7.2
(n = 87) (0.05)
Population vs. outdoor 0.4 4.7
(n = 23)
Personal (all subjects) 0.45 7.2
vs. outdoor (n = 1,494)
Personal (no smokers 0.38 7.2
and gas cooking) vs.
outdoor (n = 943)
Personal vs. outdoor 0.49 14.5
(n = 546)
Personal vs. outdoor 0.56 15.8
0.27
0.86
0.33
0.27
0.51
S0rensen et al.
(2005)
Location: Copenhagen, Denmark
Subjects: 30 subjects (20-33 yrs old) in each measurement campaign
Time period: fall 1999, and winter, spring and summer of 2000
Method: four measurement campaigns in 1 yr; each campaign lasted 5 wks
with 6 subjects each wk; one 48-h avg NOR2R measurement for each
subject.
All
(>8 °C)
(<8 °C)
Personal vs. outdoor
(n = 73)
Personal vs. outdoor
(n = 35)
Personal vs. outdoor
(n = 38)
0.60 — —
(0.07)
0.68 — —
(0.09)
0.32 — —
(0.13)
-------
TABLE 2.5-5 (cont'd). SUMMARY OF REGRESSION MODELS OF PERSONAL EXPOSURE
TO AMBIENT/OUTDOOR NO2
t-O
o
o
oo
t-o
I
oo
O
H
6
O
2
o
o
H
m
o
?d
o
Study
Piechocki-Minguy
et al. (2006)
S0rensen et al.
(2005)
Aim et al.
(1998)
Sarnat et al.
(2001)
Sarnat et al.
(2005)
Location
Location: Pooled, Lille (northern France)
Subjects: 13 participants in the first campaign, and 31 participants in the
second campaign
Time period: winter 2001 (first campaign), and summer 2002 (second
campaign)
Method: two 24-h sampling periods (one during the workdays and the other
during the weekends) for each subject in each campaign; during each
sampling period, each subject received four samplers to measure personal
exposure in four different microenvironments (home, other indoor
environment, transport, and outdoors) .
Location: Copenhagen
Subjects: 30 subjects (20-33 yrs old) in each measurement campaign
Time period: fall 1999, and winter, spring and summer of 2000
Method: four measurement campaigns in 1 yr; each campaign lasted 5 wks
with 6 subjects each wk; one 48-h avg NOR2R measurement for each
subject.
Location: Helsinki, Finland
Subjects: 246 children aged 3-6 yrs
Time period: winter and spring of 1991
Method: 1-wk averaged sample for each person, 6 consecutive wks in the
winter and 7 consecutive wks in the spring.
Location: Baltimore, MD
Subjects: 56 seniors, Schoolchildren, and people with COPD
Time period: summer of 1998 and winter of 1999
Method: 14 of 56 subjects participated in both sampling seasons; all subjects
were monitored for 12 consecutive days (24-h avg sample) in each of the
one or two seasons, with the exception of children who were measured for 8
consecutive days during the summer.
Location: Boston, MA
Subjects: 43 seniors and schoolchildren
Time period: summer of 1999 and winter of 2000
Method: Similar study design as Sarnat et al., 2001.
Season
All
Summer
(homes with
no major
indoor NO 2
sources)
All
Winter +
Spring
Summer
Winter
Summer
Winter
Slope
Model Type (SE) Intercept / ppb
Personal vs. central 0.13 6.0
(Assuming people Q gg _ g 7
stayed indoors all the
time)
Personal vs. central 0.56 —
(n = 66) (0.09)
Population vs. central 0.3 5.0
(n = 24)
Personal vs. central 0.04* 9.5
(n = 225 for 24
subjects)
Personal vs. central -0.05* 18.2
(n = 487 for 45
subjects)
Personal vs. central 0.19 —
(n = 341)
Personal vs. central -0.03* —
(n = 298)
R2
0.09
0.61
—
0.37
—
—
—
m
-------
t-o
o
o
oo
t-o
I
CO
O
H
6
O
2
o
H
m
o
?d
o
t—t
H
m
TABLE 2.5-5 (cont'd). SUMMARY OF REGRESSION MODELS OF PERSONAL EXPOSURE
TO AMBIENT/OUTDOOR NO2
Study
Sarnat et al.
(2006)
Location
Location: Steubenville
Subjects: 15 senior subjects
Time period: summer and fall of 2000
Method: two consecutive 24-h samples were collected for each subject
for each wk, 23 wks total.
Season
Summer
Fall
Model Type
Personal vs. central
(n = 122)
Personal vs. central
(n 1 3R1
Slope (SE) Intercept / ppb
0.25 (0.06) —
0.49 (0.05) —
R2
0.14
0.43
*Not significant at the 5% level.
-------
TABLE 2.5-6. INDOOR/OUTDOOR RATIO AND THE INDOOR VS. OUTDOOR REGRESSION SLOPE
t-o
O
O
OO
Study
Description
Season Regression Format or Ratio
Indoor
Characteristics
Comments
t-o
I
o
O
H
a
o
2
o
o
H
m
o
?a
o
Baxter Location: Boston, MA Overall
et al. Subjects: 43 homes (a lower social-economic study
(2007a) status population) seasons
Time period: May-October (non-heating
season), and Dec-Mar (heating season), 2003-
2005
Method: indoor and outdoor 3- to 4-day samples
of N02 were collected simultaneously at each
home in both seasons; when possible, 2
consecutive measurements were collected.
Baxter Location: Boston, MA Overall
et al. Subjects: 43 homes (a lower social-economic study
(2007b) status population) seasons
Time period: May-Oct (non-heating season),
and Dec-Mar (heating season), 2003-2005
Method: indoor and outdoor 3- to 4-day samples
of N02 were collected simultaneously at each
home in both seasons; when possible, 2
consecutive measurements were collected.
Mosqueron Location: Paris, France Overall
et al. (2002) Subjects: 62 office workers study
Time period: Dec 1999 to Sept 2000 seasons
Method: 48-h residential indoor, workplace,
outdoor, and personal exposure were
measured.
Lee et al. Location: Hong Kong, China Overall
(1999) Subjects: 14 public places with mechanical study
ventilation systems, seasons
Time period: Oct 1996 to Mar 1997
Method: Teflon bags were used to collect indoor
and outdoor NO and NOR2R during peak
hours.
Residential indoor vs. ambient Gas stove usage
and indoor source and
proximity to traffic
0.66-0.79
The overall R' was 0.20-0.25.
Residential indoor vs.
residential outdoor
Overall homes 0.48
Homes with high 0.56
ventilation rate
Homes with low 0.47
ventilation rate
Residential indoor vs. Overall homes 0.53
residential outdoor and indoor
sources
Residential indoor vs. ambient Cooking 0.26 (n = 62)
and using gas cooking
Office indoor vs. ambient and None 0.56 (n = 62)
floor height
Indoor vs. outdoor — 0.59 (n = 14)
Home with an indoor/outdoor
sulfur ratio larger than 0.76 (the
median) was defined as a high
ventilation home; Home with an
indoor/outdoor sulfur ratio less
than 0.76 (the median) was
defined as a low ventilation
home.
The overall R2 was 0.16.
The overall R was 0.14, and
ambient NO 2 and indoor
cooking account accounted for
0.07 each.
The overall R2 was 0.24, partial
R2 for ambient and floor height
were 0.18 and 0.06,
respectively.
R2 was 0.59. The slopes for
NO and NOX were 1.11 and
1.04 respectively.
m
-------
TABLE 2.5-6 (cont'd). INDOOR/OUTDOOR RATIO AND THE INDOOR VS. OUTDOOR REGRESSION SLOPE
Regression Indoor
Description Season Format or Ratio Characteristics F;/!/
t-O
o
o
oo
Study
Comments
O
H
b
o
2
o
H
m
o
w
o
K^
H
m
Monn Location: Switzerland
et al. Subjects: 17 homes across Switzerland
(1997) Time period: winter 1994 to summer 1995
Method: 48- to 72-h indoor, outdoor, and personal
NOR2R were measured.
Lee Location: Boston area, MA
etal. Subjects: 517 residential homes
(1995) Time period: Nov 1984 to Oct 1986
Method: 2-wk averaged indoor (kitchen, living
room, and bedroom) and outdoor N02 were
measured.
Garrett Location: Latrobe Valley, Victoria, Australia
et al. Subjects: 80 homes
(1999) Time period: Mar-Apr 1994, and Jan-Feb 1995
Method: 4-day averaged indoor (bedroom, living
room, and kitchen) and outdoor NOR2R was
monitored.
Monn Location: Geneva, Basle, Lugano, Aarau, Wald,
et al. Payerne, Montana, and Davos (SAPALDIA
(1998) study, Switzerland)
Subjects: 140 subjects
Time period: Dec 1993 to Dec 1994
Method: each home was monitored for 3 periods
of 1 mo; in the 1st wk of each period, personal,
indoor and outdoor levels were measured; for
the next 3 wks, only outdoor levels were
measured (1-wk averaged measurement).
Spengler Location: Los Angeles Basin, CA
et al. Subjects: probability-based sample, 70 subjects
(1994) Time period: May 1987 to May 1988
Method: 48-h averaged, in the micro-
environmental component, each participant
was monitored during each of 8 sampling
cycles throughout the yr.
Overall
study
seasons
Indoor/outdoor
ratio
Summer Indoor/outdoor
ratio
Overall Indoor/outdoor
study ratio
seasons
Without gas cooking
Electric stove homes
0.4, -0.7
26)
n = —
No major indoor sources
(major sources were gas stove,
vented gas heater, and
smoking)
0.77 (bedroom)
(Sample size
was not
reported)
0.8(n =
Homes with gas stove and gas
stove with pilot light have an
I/O ratio > 1, but the values
were not reported.
The ratio increased to 1.3, to
1.8, and to 2.2 for homes with
one, two and three major indoor
sources.
Overall Residential indoor All homes
study vs. residential
seasons outdoor
0.47 (n = 1544) R2 was 0.37.
Homes without smokers and 0.40 (n = 968) R2 was 0.33.
gas-cooking
Overall Residential indoor Gas range with pilot light
study vs. residential
seasons outdoor
Electric stove
Gas range without pilot light
0.49 (n = 314)
0.4 (n = 148)
0.4 (n = 170)
R2 was 0.44.
R2 was 0.39.
R2 was 0.41.
-------
i 3. INTEGRATED HEALTH EFFECTS OF
2 NO2 EXPOSURE
o
4
5 In this chapter, we assess the health effects associated with human exposure to ambient
6 nitrogen dioxide (NO2) in the United States. The main goal of this chapter is to (1) integrate
7 newly available epidemiologic, human clinical, and animal toxicological evidence with
8 consideration of key findings from the 1993 Air Quality Criteria Document (AQCD) for Oxides
9 of Nitrogen (U.S. Environmental Protection Agency, 1993) and (2) draw conclusions about the
10 causal nature of NO2 relative to a variety of health effects. These causal determinations utilize
11 the framework outlined in Chapter 1.
12 This chapter is organized to present morbidity and mortality associated with short-term
13 exposures to NO2, followed by morbidity and mortality associated with long-term exposures.
14 Within these divisions, the chapter is organized by health outcome, such as respiratory symptoms
15 in asthmatics, emergency department (ED) visits and hospital admissions for respiratory and
16 cardiovascular diseases (CVDs), and premature mortality. The sections describe the findings of
17 epidemiologic studies that have characterized the association between ambient NO2 exposure
18 and heath outcomes and includes relevant human clinical and animal toxicologic data, when
19 available. This integrated discussion underlies judgments in causal inference.
20 The epidemiologic studies contain important information on potential associations
21 between health effects and exposures of human populations to ambient levels of NO2, and they
22 help to identify susceptible subgroups and associated risk factors. However, the associations
23 derived for specific air pollutants and health outcomes in epidemiologic studies may be
24 confounded or obscured by copollutants and/or meteorological conditions and can be influenced
25 by model specifications in the analytical methods. Extensive discussion of issues related to
26 confounding effects among air pollutants in epidemiologic studies is provided in the 2004
27 AQCD for Particulate Matter (PM) and so is not repeated in detail here. Briefly, though, the use
28 of multipollutant regression models has been the approach most commonly used to control for
29 these potential copollutant confounders. One specific concern has been that a given pollutant
30 may act as a surrogate for other unmeasured or poorly measured pollutants or pollutant mixtures.
31 Specifically, traffic is a nearly ubiquitous source of combustion pollutant mixtures that include
32 NO2 and can be an important contributor to NO2 levels in near-road locations. Although this
March 2008 3-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 complicates efforts to disentangle specific MVrelated health effects as distinct from those
2 effects of the whole traffic-generated combustion mix, multipollutant models with terms for
3 measured variables remain important tools for partitioning the variance structures in multisource
4 epidemiologic studies.
5 Model specification and model selection also are factors to consider in the interpretation
6 of the epidemiologic evidence. Epidemiologic studies investigated the association between
7 various measures of NC>2 (e.g., multiple lags, different exposure metrics) and various health
8 outcomes using different model specifications (for further discussion, see 2006 AQCD for Ozone
9 [Os] AQCD [U.S. Environmental Protection Agency, 2006]). The summary of health effects
10 evidence in this chapter is vulnerable to the errors of publication bias and multiple testing, and
11 efforts have been made to reduce the impact of multiple testing errors on the conclusions in this
12 evaluation. For example, although many studies examined multiple single-day lag models,
13 priority was given to effects observed at 0- or 1-day lags rather than at longer lags. Both single-
14 and multipollutant models that include NC>2 were considered and examined for robustness of
15 results.
16 Human clinical studies conducted in controlled exposure chambers use fixed
17 concentrations of air pollutants under carefully regulated environmental conditions and subject
18 activity levels to minimize possible confounding of the health associations by other factors.
19 Additionally, sensitive experimental techniques can be used to measure health effects that are not
20 evaluated in epidemiologic studies, e.g. airways hyperresponsiveness. These studies provide
21 important information on the biological plausibility of associations observed between NC>2
22 exposure and health outcomes in epidemiologic studies. While human clinical studies provide a
23 direct quantitative assessment of the NO2 exposure-health response relationship, such studies
24 have a number of limitations. First, it is requisite that subjects be either healthy individuals or
25 individuals whose level of illness does not preclude them from participating in the study.
26 Therefore, the results of human clinical studies may underestimate the health effects of exposure
27 to certain sensitive subpopulations. Second, studies of controlled exposure to NC>2 typically have
28 used concentrations that are higher than those normally present in ambient air. Third, human
29 clinical studies normally are conducted on a relatively small number of subjects, which reduces
30 the power of the study to detect significant differences in the health outcomes of interest between
31 exposure to varying concentrations of NC>2 and clean air.
March 2008 3-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 Similar to human clinical studies, animal toxicological studies have the advantage of
2 being conducted under controlled conditions, using fixed concentrations of air pollutants in
3 carefully regulated environmental conditions. These studies allow for evaluation of biological
4 responses with exposures to substances in doses that could be hazardous to human health and for
5 extended durations that are not possible in human clinical studies. However, limitations on study
6 population size require the use of higher doses to allow the identification of rare events. An
7 important caveat in interpretation of the toxicological data is that the high doses used in many of
8 the studies may produce different effects on the lung than inhalation exposures at lower ambient
9 concentrations. That is, "realistic" doses associated with ambient nitrogen oxides exposures may
10 activate cells and pathways entirely disparate from those activated at high experimental doses.
11 In addition, significant differences in biology can exist, depending on species and strain selected,
12 that can affect the response.
13 This chapter focuses on the important new scientific studies, with emphasis on those
14 conducted at or near current ambient concentrations. The attached annexes include a broad
15 survey of the relevant epidemiology, human clinical, and toxicology literature to supplement the
16 information presented here.
17
18
19 3.1 RESPIRATORY MORBIDITY RELATED TO NO2 SHORT-TERM
20 EXPOSURE
21 In the 1993 AQCD for Oxides of Nitrogen, human clinical evidence indicated that
22 caused decrements in lung function, particularly increased airways resistance in healthy subjects,
23 with exposures of >2.0 parts per million (ppm) for 2 h. Other studies showed increased airways
24 responsiveness in healthy subjects at concentrations of >1 ppm for 1 h. Asthmatics and chronic
25 obstructive pulmonary disease (COPD) patients demonstrated increased decrements in lung
26 function that were dependent on exposure conditions. However, concentration-response
27 relationships were not observed for changes in lung function, airways responsiveness, or
28 symptoms, and no association was apparent between lung function responses and respiratory
29 symptoms.
30 At the time of the 1993 AQCD for Oxides of Nitrogen, many of the available
3 1 epidemiologic studies consisted predominately of indoor NO2 exposure studies. Although indoor
32 sources in these studies include both gas-fueled cooking and heating appliances, in most of the
March 2008 3-3 DRAFT-DO NOT QUOTE OR CITE
-------
1 earlier studies the focus was primarily on cooking stoves. Although there was some evidence
2 suggesting that increased NO2 exposure was associated with increased respiratory symptoms in
3 children aged 5 to 12 years, the main conclusion was that there was insufficient epidemiologic
4 evidence for an association between short-term exposure and health effects. The 1993 AQCD
5 also presented an intervention study conducted in 1972 and 1973 in Chattanooga, TN (Shy and
6 Love, 1980; Love et al., 1982) that reported a reduction of the respiratory illness rate in 1973
7 associated with a strike at a primary source that resulted in lowered NC>2 pollution. The study
8 suggested that short-term (peak) exposure may be more important than long-term exposure to
9 NO2. A limitation of this study was that it offered only qualitative information evaluating the
10 question of removing exposures leading to reduced risk.
11 Animal toxicology studies evaluated in the 1993 AQCD identified biochemical and
12 cellular mechanisms whereby NC>2 induces effects at concentrations of as low as 0.04 ppm. The
13 biochemical effects observed in the respiratory tract after NC>2 exposure include chemical
14 alteration of lipids, amino acids, proteins, and enzymes and changes in oxidant/antioxidant
15 homeostasis. Membrane polyunsaturated fatty acids and thiol groups are the main biochemical
16 targets for NC>2 exposure. Data available in the 1993 AQCD indicated that NC>2 induces lipid
17 peroxidation and changes in lipid content of cell membranes. The biochemical pertubations
18 mentioned above could result in cellular damage either directly through the generation of
19 reactive oxygen species, or by rendering the cells more susceptible to injury by altering the
20 protective mechanisms (i.e. membrane integrity, antioxidant levels).
21 A large body of epidemiologic evidence has been published since the 1993 AQCD for
22 Oxides of Nitrogen on respiratory health outcomes associated with short-term exposure to NC>2.
23 The health outcomes studied included occurrence of respiratory symptoms, changes in lung
24 function, and ED visits and hospitalizations for respiratory diseases. Relatively few new clinical
25 and animal toxicologic studies have been published since 1993.
26
27 3.1.1 Lung Host Defenses and Immunity
28 Lung host defenses are sensitive to NC>2 exposure, with numerous measures of such
29 effects observed at concentrations of <1 ppm. Potential mechanisms, according to Chauhan et al.
30 (2003), include "direct effects on the upper and lower airways by ciliary dyskinesis (Carson
31 et al., 1993), epithelial damage (Devalia et al., 1993a), increases in pro-inflammatory mediators
March 2008 3-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 and cytokines (Devalia et al., 1993b), rises in IgE concentration (Siegel et al., 1997), and
2 interaction with allergens (Tunnicliffe et al., 1994), or indirectly through impairment of
3 bronchial immunity (Sandstrom et al., 1992a)." Table 3.1-1 summarizes a range of proposed
4 mechanisms by which exposure to NO2 in conjunction with viral infections may exacerbate
5 upper and lower airways symptoms (Chauhan et al., 1998). Another major concern has been the
6 potential for NO2 exposure to enhance susceptibility to or the severity of illness resulting from
7 respiratory infections and asthma, especially in children. The following discussion focuses on
8 studies published since the 1993 AQCD and conducted at near-ambient exposure concentrations
9 but, as needed, refers to studies in the 1993 AQCD for Oxides of Nitrogen.
10 Several epidemiologic studies investigated the host defenses interplay with prior NO2
11 exposure and viral infection. Personal exposure to NO2 and the severity of virus-induced asthma
12 (Chauhan et al., 2003), including risk of airflow obstruction (Linaker et al., 2000) was studied in
13 a group of 114 asthmatic children in England. Children were supplied with Palmes diffusion
14 tubes, which they attached to their clothing during the day and placed in their bedroom at night.
15 Tubes were changed every week for the duration of the 13-month study period. Nasal aspirates
16 were obtained and analyzed for a variety of respiratory illness-causing viruses. The authors
17 observed that exposure to NO2 levels of greater than 14 |ig/m3 (7.4 parts per billion [ppb]) in the
18 week preceding any viral infection was associated with increases in the four-point symptom
19 severity score (score increase of 0.6 [95% CI: 0.01, 1.18]) in the week immediately after the
20 infection. Associations also were observed for the respiratory syncytial virus (RSV) alone (score
21 increase of 2.1 [95% CI: 0.52, 3.81]). A significant reduction in peak expiratory flow (PEF) was
22 associated with exposure greater than 14 |ig/m3 (7.3 ppb) (by 12 L/min [95% CI: -23.6, -0.80])
23 (Chauhan et al., 2003). Exploration of the relationship between PEF and NO2 showed that the
24 risk of a PEF episode (as diagnosed by a clinician's review of each child's PEF data) beginning
25 within a week of an upper respiratory infection was significantly associated with exposure to
26 NO2 greater than 28 |ig/m3 (14.9 ppb) (relative risk [RR] = 1.9 [95% CI: 1.1, 3.4]) (Linaker
27 et al., 2000). Thus, high personal NO2 exposure in the week before an upper respiratory
28 infection was associated with either increased severity of lower respiratory tract symptoms or
29 reduction of PEF for all virus types together and for two of the common respiratory viruses, C
30 picornavirus and RSV, individually.
March 2008 3-5 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3.1-1. PROPOSED MECHANISMS WHEREBY NO2 AND RESPIRATORY
VIRUS INFECTIONS MAY EXACERBATE UPPER AND LOWER
AIRWAY SYMPTOMS
Proposed Mechanisms
Upper Airways
Epithelium
J, Ciliary beat frequency
t Epithelial permeability
Lower Airways
Epithelium
Cytokines
Inflammatory cells
Inflammatory mediators
Allergens
(as in upper airways)
| Epithelial-derived IL-8, GM-CSF, TNF-a
t Macrophage-derived IL-lb, IL-6, IL-8, TNF-a
t Mast cell tryptase
t Neutrophils
t Total lymphocytes
t NK lymphocytes
I T-helper/T-cytotoxic cell ratio
| Free radicals, proteases, TXA2, TXB2, LTB4
t Penetrance due to ciliostasis
| PD20-FEVi
t Antigen-specific IgE
t Epithelial permeability
Peripheral Blood
I Total macrophages
J, B and NK lymphocytes
| Total lymphocytes
Source: Adapted from Chauhan et al. (1998).
4
5
6
Several clinical studies have attempted to address the question of whether NO2 exposures
impair host defenses and/or increase susceptibility to infection (Rehn et al., 1982; Goings et al.,
1989; Rubenstein etal., 1991; Sandstrom etal. 1990, 1991, 1992a,b; Devlin 1992, 1999;
Frampton et al., 2002) (see the 1993 AQCD for details of older studies and Annex Table
AX5.2-1 for additional details on newer studies). These studies have reported inconsistent
results. One approach has been to examine the effects of in vivo NO2 exposure on the function
March 2008
3-6
DRAFT-DO NOT QUOTE OR CITE
-------
1 of alveolar macrophages (AMs) obtained by bronchoalveolar lavage (BAL), including the
2 susceptibility of these cells to viral infection in vitro. Two studies since 1993 involved 2.0-ppm
3 NC>2 exposures for 4 or 6 h with intermittent exercise and found no effect on AM inactivation of
4 influenza virus either immediately or 18 h after exposure (Azadniv et al., 1998; Devlin et al.,
5 1999). However, Devlin et al. (1999) found reduced AM phagocytic capacity after NO2
6 exposure, suggesting a reduced ability to clear inhaled bacteria or other infectious agents.
7 Frampton et al. (2002) examined NC>2 effects on viral infectivity of airways epithelial cells.
8 Subjects were exposed to air, or 0.6- or 1.5-ppm NO2, for 3 h, and bronchoscopy was performed
9 3.5 h after exposure. Epithelial cells were harvested from the airways by brushing and then
10 challenged in vitro with influenza virus and RSV. NC>2 exposure did not alter viral infectivity,
11 but appeared to enhance epithelial cell injury in response to infection with RSV (p = 0.024).
12 Similar results were reported with influenza virus. These findings suggest that prior exposure to
13 NC>2 may increase the susceptibility of the respiratory epithelium to injury by subsequent viral
14 challenge.
15 There is evidence from both animal and human studies indicating that exposure to NC>2
16 may alter lymphocyte subsets in the lung and possibly in the blood. Lymphocytes, particularly T
17 lymphocytes and NK cells, play a key role in the innate immune system and host defense against
18 respiratory viruses. Rubenstein et al. (1991) found that a series of four daily, 2-h exposures to
19 0.60-ppm NC>2 resulted in a small increase in NK cells recovered by BAL. Sandstrom et al.
20 (1990, 1991) observed a significant, dose-related increase in lymphocytes and mast cells
21 recovered by BAL 24-h after a 20-min exposure to NC>2 at 2.25 to 5.50 ppm. In contrast,
22 repeated exposures to 1.5- or 4-ppm NC>2 for 20 min every second day on six occasions resulted
23 in decreased CD16+56+ (NK cells) and CD19+ cells (B lymphocytes) in BAL fluid 24-h after the
24 final exposure (Sandstrom et al., 1992a,b). No effects were reported on polymorphonuclear
25 leukocytes (PMNs) or total lymphocyte numbers. Solomon et al. (2000) found a decrease in
26 CD4+ T lymphocytes in BAL fluid 18-h after three daily, 4-h exposures to 2.0-ppm NC>2.
27 Azadniv et al. (1998) observed a small but significant reduction in CD8+ T lymphocytes in
28 peripheral blood, but not BAL fluid, 18 h following single 6-h exposures to 2.0-ppm NO2.
29 Frampton et al. (2002) found small increases in BAL lymphocytes and decreases in blood
30 lymphocytes with exposures to 0.6 and 1.5 ppm NC>2 for 3 h.
March 2008 3-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 The observed effects on lymphocyte responses, as described above, have not been
2 consistent among studies. Differing exposure protocols and small numbers of subjects among
3 these studies may explain the varying and conflicting findings. Furthermore, the clinical
4 significance of transient, small changes in lymphocyte subsets is unclear. It is possible that the
5 inflammatory response to NC>2 exposure involves both lymphocytes and PMNs, with lymphocyte
6 responses occurring transiently and at lower concentrations, and PMN responses predominating
7 at higher concentrations or more prolonged exposures. The airways lymphocyte responses do
8 not provide convincing evidence of impairment in host defense.
9 One clinical study used fiber-optic bronchoscopy and found that 20-min exposures to
10 NO2 at 1.5 to 3.5 ppm transiently reduced airways mucociliary activity (Helleday et al., 1995).
11 Reduced mucus clearance is expected to increase susceptibility to infection by reducing the
12 removal rate of microorganisms from airways. However, the study was weakened by the lack of
13 a true air control exposure as well as by the absence of randomization and blinding. As a
14 clarification, Helleday et al. (1995) did not measure mucus clearance rates directly using
15 radiolabeled particles; rather they utilized an optical technique to characterize ciliary activity.
16 Rehn et al. (1982) examined the effect of NO2 exposure on mucociliary clearance of a
17 radiolabeled Teflon aerosol. After a 1-h exposure to either 0.27- or 1.06-ppm (500 or
18 2000 |ig/m3) NO2, there were no changes in airways clearance rates.
19 Animal studies provide clearer evidence that host defense system components such as
20 mucociliary transport and AMs (see Annex Table AX4.3) are targets for inhaled NC>2. Animal
21 studies further show that NC>2 can impair the respiratory host defense system sufficiently to
22 render the host more susceptible to respiratory infections (See Annex Table AX4.6). Exposure
23 of guinea pigs to 3- or 9-ppm NC>2 6 h/day, 6 days/week for 2 weeks resulted in concentration-
24 dependent decreases in ciliary activity of 12 and 30% of control values, respectively (Ohashi
25 etal., 1994). These concentration-dependent decreases were accompanied by a concentration-
26 dependent increase in eosinophil accumulation on the epithelium and submucosal connective
27 tissue layer of the nasal mucosa. For foreign agents such as some bacteria and viruses that
28 deposit below the mucociliary region in the gas-exchange region of the lung, AMs primarily
29 provide host defenses by acting to remove or kill viable particles, remove nonviable particles,
30 and process and present antigens to lymphocytes for antibody production. AMs are one of the
31 sensitive targets for NC>2, as evidenced by in vivo animal exposures and in vitro studies (see
March 2008 3-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 Annex Table AX4.3 for details of studies related to each of these morphological or functional
2 parameters in exposed animals).
3 Suppression of host defense mechanisms by NO2 as described in the studies above are
4 expected to result in an increased incidence and severity of pulmonary infections (Miller et al.,
5 1987, Gardner et al., 1979; Coffin and Gardner, 1972). Various experimental approaches have
6 been employed using animals in an effort to determine the overall functional efficiency of the
7 host's pulmonary defenses following NO2 exposure. In the most commonly used infectivity
8 model, animals are exposed to either NO2 or filtered air and the treatment groups are combined
9 and exposed briefly to an aerosol of a viable agent, such as Streptococcus spp., Klebsiella
10 pneumonias, Diplococcuspneumoniae, or influenza virus and mortality rates are determined
11 (Ehrlich, 1966; Henry et al., 1970; Coffin and Gardner, 1972; Ehrlich et al., 1979; Gardner,
12 1982). Although the endpoint is mortality, this experimental test is considered a sensitive
13 indicator of the depression of the defense mechanisms and is a commonly used assay for
14 assessing immunotoxicity. The susceptibility to bacterial and viral pulmonary infections in
15 animals also increases with NO2 exposures of as low as 0.5 ppm. No new studies published
16 since 1993 were identified that evaluated this endpoint. Annex Table AX4.6 summarizes the
17 effects of NO2 exposure and infectious agents in animal studies as compiled in the 1993 AQCD
18 for Oxides of Nitrogen, and provides evidence that the host's response to inhaled NO2 can be
19 influenced significantly by the duration and temporal patterns of exposure. This is important in
20 considering continuous versus intermittent exposures and attempting to understand observed
21 differences in reported results.
22
23 Summary of Evidence on the Effect of Short-Term Exposure to NO2 on Lung Host Defenses
24 and Immunity
25 Impaired host-defense systems and increased risk of susceptibility to both viral and
26 bacterial infections have been observed in epidemiologic, human clinical, and animal
27 toxicological studies. A recent epidemiologic study provided evidence that increased personal
28 exposures to NO2 worsened virus-associated lower respiratory tract symptoms in children with
29 asthma (Chauhan et al., 2003). The limited evidence from human clinical studies indicates that
30 NO2 may increase susceptibility to injury by subsequent viral challenge at exposures of as low as
31 0.6 ppm for 3 h (Frampton et al., 2002). Toxicological studies have shown that lung host
32 defenses are sensitive to NO2 exposure, with several measures of such effects observed at
March 2008 3-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations of less than 1 ppm. The epidemiologic and experimental evidence together show
2 coherence for effects of NO2 exposure on host defense or immune system effects. This group of
3 outcomes also provides plausibility and potential mechanistic support for other respiratory
4 effects described subsequently, such as respiratory symptoms or ED visits for respiratory
5 diseases.
6
7 3.1.2 Airways Inflammation
8 Epidemiologic studies have examined biological markers for inflammation (exhaled
9 nitric oxide [NO] and inflammatory nasal lavage [NAL] markers) and lung damage (urinary
10 Clara cell protein CC16). Several studies have been conducted in cohorts of children.
11 Steerenberg et al. (2001) studied 126 schoolchildren from urban and suburban communities in
12 the Netherlands. Sampling of exhaled air and NAL fluid was performed seven times, once per
13 week over the course of 2 months. On average, the ambient NO2 concentrations were 1.5 times
14 higher, and ambient NO concentrations were 7.8 times higher, in the urban compared to the
15 suburban community. Compared to children in the suburban community, urban children had
16 significantly greater levels of inflammatory NAL markers (interleukin [ILJ-8, urea, uric acid,
17 albumin) but not greater levels of exhaled NO. However, within the urban group, a statistically
18 significant concentration-response relationship for exhaled NO was observed. Exhaled NO
19 increased by 6.4 to 8.8 ppb per 20-ppb increase in NO2 lagged by 1 or 3 days. Another study by
20 Steerenberg et al. (2003) of 119 schoolchildren in the Netherlands found associations between
21 ambient NO2 and level of exhaled NO, but quantitative regression results were not given. The
22 authors concluded from their data that an established, ongoing inflammatory response to pollen
23 was not exacerbated by subsequent exposure to high levels of air pollution or pollen.
24 In one recent U.S. study, Delfino et al. (2006) evaluated the relationship between
25 personal and ambient levels of fine PM (PM^.s), elemental carbon (EC), organic carbon (OC),
26 and NO2 and fractional exhaled NO (FENo), a biomarker of airway inflammation, in a panel of
27 45 schoolchildren with persistent asthma living in two southern California communities
28 (Riverside and Whittier). FENo is higher in subjects with poorly controlled asthma. Positive
29 associations were found for FENo with several air pollutants, including NO2, with evidence from
30 multipollutant approaches suggesting that traffic-related sources of air pollutants underlie the
31 findings. The authors concluded that the "association of FENo with personal and ambient NO2
March 2008 3-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 was largely independent of personal and ambient EC and OC fractions of PM2.5 in two-pollutant
2 models, suggesting that both ambient and personal NO2 represent other causal pollutant
3 components not sufficiently captured by ambient EC or OC in the study regions." While the
4 effect was small (<2.5 ppb FENO), making it difficult to determine if it is clinically relevant, the
5 findings suggest that air pollutant exposure increases inflammation in children.
6 Several studies have evaluated effects in adult cohorts. Adamkiewicz et al. (2004)
7 studied 29 elderly adults in Steubenville, OH and found significant associations between
8 increased exhaled NO and increased daily levels of PM2.5, but no association was found with
9 ambient NO2. Timonen et al. (2004) collected biweekly urine samples for 6 months from 131
10 adults with coronary heart disease living in Amsterdam, Helsinki, and Erfurt, Germany.
11 Estimates using data from all three communities showed significant associations between urinary
12 levels of Clara cell protein CC16 (a marker for lung damage) with elevations in daily PM2 5
13 concentration, but not ambient NO2. In Helsinki, however, a statistically significant positive
14 association was observed between NO2 lagged by 3 days and CC16 levels. Interestingly, the
15 correlation between NO2 and PM2 5 was lower in Helsinki (r = 0.35) compared to this correlation
16 in Amsterdam (r = 0.49) or Erfurt (r = 0.82). Bernard et al. (1998) examined personal exposure
17 to NO2 and its effect on plasma antioxidants in a group of 107 healthy adults in Montpellier,
18 France. Subjects wore passive monitors for 14 days. When subjects were divided into two
19 exposure groups (above and below 40 |ig/m3 [21.3 ppb]), those in the high-exposure group had
20 significantly lower plasma P-carotene levels. This difference was even greater when the analysis
21 was stratified by dietary P-carotene intake: exposure to >40-|ig/m3 (21.3 ppb) NO2 had
22 the largest effect on plasma P-carotene level among subjects whose diet contained <4 mg/day
23 P-carotene (p < 0.005). No other pollutants were included in this study.
24 The 1993 AQCD for Oxides of Nitrogen cited preliminary findings from two clinical
25 studies showing modest airways inflammation, as indicated by increased PMN numbers in BAL
26 fluid after exposure to 2.0-ppm NO2 for 4 to 6 h with intermittent exercise. Both of those studies
27 now have been published in complete form (Azadniv et al, 1998; Devlin et al, 1999), and
28 additional studies summarized below provide a clearer picture of the airways inflammatory
29 response to NO2 exposure.
30 Annex Table AX5.1 summarizes the key clinical studies of NO2 exposure in healthy
31 subjects published since 1993, with a few key studies included prior to that date. Figure 3.1-1
March 2008 3-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 illustrates the dose-response relationship between NO2 exposure and inflammatory respoi
2 healthy
+
^m
— .
/*~*
,9
> — «.
0
subjects.
1
"vT
1
jmmmm,
*-\
1
]
&_
r
i.
Q
O
i
7
_
*
(s)(3h)
_ _ __ — _ — _
*
(7)(6h)
V^—s-sX x '
(11) (3h)
^_^*
D
O
10
Pulmonary Function
Airway Inflammation
Study ppm-min
1. Barck et a!. (2002) 7.8
2. Barck et a!, (2005a) 23.4
3, Barck et a!. (2005b) 7.8
4. Jenkins et al.
5. Jenkins etal.
(1999) 72 (6h)
(1999) 72 (3h)
6. Strand etal. (1997) 7.8
7. Strand e
tal. (
19981 31.2
8. Tunnicliffe etal. (1994) 24
9. Tunnicliffe etal. (1994) 6
10. Wangetal. (1995a,b) 144
11. Witten et al. (2005) 72
i i
50 100
I
150
NO2 ppm-minutes
Figure 3.1-1. Studies of airways inflammatory responses in relation to the total exposure to
NOi, expressed as ppm-minutes. All of the studies involved intermittent
exercise, and no attempt was made to adjust the exposure metric for varying
intensity and duration of exercise. Studies that did not include a proper
control air exposure and those that used multiple daily exposures were not
included in this figure.
3 Healthy volunteers exposed to 2.0-ppm NO2 for 6 h with intermittent exercise showed a
4 slight increase in the percentage of PMNs obtained in BAL fluid 18 h after exposure (air, 2.2 ±
5 0.3%; NO2, 3.1 ± 0.4%) (Azadniv et al., 1998). Gavras et al. (1994) studied a separate group of
6 subjects exposed using the same protocol but assessed immediately after exposure. In this case,
7 no effects were found in AM phenotype or expression of the cell adhesion molecule CD1 Ib or
8 receptors for IgG. Blomberg et al. (1997) reported that 4-h exposures to 2.0-ppm NO2 resulted
9 in an increase in IL-8 and PMNs in the proximal airways of healthy subjects, although no
10 changes were seen in bronchial biopsies. This group also studied the effects of repeated 4-h
March 2008
3-12
DRAFT-DO NOT QUOTE OR CITE
-------
1 exposures to 2-ppm NO2 on 4 consecutive days, with BAL, bronchial biopsies, and BAL fluid
2 antioxidant levels assessed 1.5-h after the last exposure (Blomberg et al., 1999). The bronchial
3 wash fraction of BAL fluid showed a 2-fold increase in PMNs and a 1.5-fold increase in
4 myeloperoxidase, indicating persistent mild airways inflammation with repeated NO2 exposure.
5 Devlin et al. (1999) exposed 8 healthy nonsmokers to 2.0-ppm NO2 for 4-h with intermittent
6 exercise. BAL performed the following morning showed a 3.1-fold increase in PMNs recovered
7 in the bronchial fraction, indicating small airways inflammation. These investigators also
8 observed a reduction in AM phagocytosis and superoxide production, indicating possible adverse
9 effects on host defense.
10 Pathmanathan et al. (2003) conducted four repeated daily exposures of healthy subjects to
11 4-ppm NO2 or air for 4 h, with intermittent exercise. Exposures were randomized and separated
12 by 3 weeks. Bronchoscopy and bronchial biopsies were performed 1-h after the last exposure.
13 Immunohistochemistry of the respiratory epithelium showed increased expression of IL-5, IL-10,
14 and IL-13, as well as intercellular adhesion molecule-1 (ICAM-1). These interleukins are
15 upregulated in Th2 inflammatory responses, which are characteristic of allergic inflammation.
16 The findings suggest repeated NO2 exposures may drive the airways inflammatory response
17 toward a Th2 or allergic-type response. Unfortunately, the report provided no data on
18 inflammatory cell responses in the epithelium or on the cells or cytokines in BAL fluid. Thus,
19 the findings cannot be considered conclusive regarding allergic inflammation. Furthermore, the
20 exposure concentrations of 4 ppm are considerably higher than ambient outdoor concentrations.
21 Recent studies provide evidence for airways inflammatory effects at concentrations of
22 <2.0 ppm. Frampton et al. (2002) examined NO2 concentration responses in 21 healthy
23 nonsmokers. Subjects were exposed to air or 0.6- or 1.5-ppm NO2 for 3 h, with intermittent
24 exercise, with exposures separated by at least 3 weeks. BAL was performed 3.5-h after
25 exposure. PMN numbers in the bronchial lavage fraction increased slightly (<3-fold) but
26 significantly (p = 0.0003) after exposure to 1.5-ppm NO2; no increase was evident at 0.6-ppm
27 NO2. Lymphocyte numbers increased in the bronchial lavage fraction after 0.6-ppm NO2, but
28 not 1.5 ppm. CD4+ T lymphocyte numbers increased in the alveolar lavage fraction, and
29 lymphocytes decreased in blood. These findings suggest a lymphocytic airways inflammatory
30 response to 0.6-ppm NO2, which changes to a mild neutrophilic response at 1.5-ppm NO2.
31 Solomon et al. (2000) also showed increased PMNs in the bronchial fraction of BAL 18 h after
March 2008 3-13 DRAFT-DO NOT QUOTE OR CITE
-------
1 the third consecutive day of exposure to 2.0 ppm NC>2 for 4 h with intermittent exercise. Torres
2 et al. (1995) found that 3-h exposures to 1-ppm NC>2 with intermittent exercise altered levels of
3 eicosanoids, but not inflammatory cells, in BAL fluid collected 1-h after exposure. Eicosanoids
4 are chemical mediators of the inflammatory response; their increase in BAL fluid reported in this
5 study suggests inflammation. The absence of an increase in PMN numbers may reflect the
6 timing of bronchoscopy (1 h after exposure). The peak influx of PMNs may occur several hours
7 after exposure, as it does following NC>2 exposure.
8 The clinical studies summarized above provide evidence for airways inflammation at
9 NC>2 concentrations of <2.0 ppm in healthy adults. Analyzing the bronchial fraction of BAL
10 separately appears to increase the sensitivity for detecting airways inflammatory effects of NC>2
11 exposure. The onset of inflammatory responses in healthy subjects appears to be between 100
12 and 200 ppm-min, i.e., 1 ppm for 2 to 3 h (see Figure 3.1-1).
13 Animal toxicological studies demonstrating changes in protein and enzyme levels in the
14 lung following inhalation of NC>2 are presented in Annex Table AX4.2. These studies are
15 reported in the 1993 AQCD and summarized below. Changes in protein and enzyme levels
16 reflect the ability of NC>2 to cause lung inflammation associated with concomitant infiltration of
17 serum protein, enzymes, and inflammatory cells. However, interpretation of the array of changes
18 observed may also reflect other factors. For example, NC>2 exposure may induce differentiation
19 of some cell populations in response to damage-induced tissue remodeling. Thus, some changes
20 in lung enzyme activity and protein content may reflect changes in cell types, rather than the
21 direct effects of NC>2 on protein infiltration. Furthermore, some direct effects of NC>2 on
22 enzymes are possible because NC>2 can oxidize certain reducible amino acids or side chains of
23 proteins in aqueous solution (Freeman and Mudd, 1981).
24 It has been reported that protein content changes in BAL fluid can be dependent on
25 dietary antioxidant status. NC>2 exposure increases the protein content of BAL fluid in vitamin
26 C-deficient guinea pigs at NC>2 levels of as low as 1880 |ig/m3 (1.0 ppm) after a 72-h exposure,
27 but a 1-week exposure to 752 |ig/m3 (0.4 ppm) did not increase protein levels (Belgrade et al.,
28 1981). However, Sherwin and Carlson (1973) found increased protein content of BAL fluid
29 from vitamin C-deficient guinea pigs exposed to 752-|ig/m3 (0.4 ppm) NC>2 for 1 week.
30 Differences in exposure techniques, protein measurement methods, and/or degree of vitamin C
31 deficiencies may explain the difference between the two studies. Hatch et al. (1986) found that
March 2008 3-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 the NO2-induced increase in BAL protein in vitamin C-deficient guinea pigs was accompanied
2 by an increase in lung content of nonprotein sulfhydryls and ascorbic acid and a decrease in
3 vitamin E content. The increased susceptibility to NO2 was observed when lung vitamin C was
4 reduced (by diet) to levels <50% of normal.
5 Studies in rats and mice published since the 1993 AQCD for Oxides of Nitrogen have
6 investigated the ability of NO2 to induce protein level changes consistent with inflammation.
7 Overall, these newer studies, such as Muller et al. (1994) and Pagani et al. (1994), suggest that
8 markers of inflammation measured in BAL fluid such as total protein content and content of
9 markers of cell membrane permeability (e.g., lactate dehydrogenase [LDH]) increase only at or
10 above 5-ppm exposure.
11
12 Summary of Evidence on the Effect of Short-Term Exposure to NO2 on Airways Inflammation
13 Overall, short-term exposure to NO2 has been found to increase airways inflammation in
14 human clinical and animal toxicological studies with exposure concentrations that are higher
15 than ambient levels. Human clinical studies provide evidence for increased airways
16 inflammation at NO2 concentrations of <2.0 ppm; the onset of inflammatory responses in healthy
17 subjects appears to be between 100 and 200 ppm-min, i.e., 1 ppm for 2 to 3 h. Increases in
18 biological markers of inflammation were not observed consistently in healthy animals at levels
19 of less than 5 ppm; however, increased susceptibility to NO2 concentrations of as low as 0.4ppm
20 was observed when lung vitamin C was reduced (by diet) to levels <50% of normal. The few
21 available epidemiologic studies are suggestive of an association between ambient NO2
22 concentrations and inflammatory response in the airways in children, though the associations
23 were inconsistent in the adult populations examined.
24
25 3.1.3 Airways Hyperresponsiveness
26 Inhaled pollutants such as NO2 may have direct effects on lung function, or they may
27 enhance the inherent responsiveness of the airways to challenge with a bronchoconstricting
28 agent. Asthmatics are generally more sensitive to nonspecific bronchoconstricting agents than
29 nonasthmatics, and airways challenge testing is used as a diagnostic test in asthma. There is a
30 wide range of airways responsiveness in healthy people, and responsiveness is influenced by
31 many factors, including medications, cigarette smoke, pollutants, respiratory infections,
32 occupational exposures, and respiratory irritants. Several drugs and other stimuli that cause
March 2008 3-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 bronchoconstriction have been used in challenge testing, including the cholinergic drugs
2 methacholine and carbachol, as well as histamine, hypertonic saline, cold air, and sulfur dioxide
3 (SO2). Challenge with "specific" allergens is considered in asthmatics. Standards for airways
4 challenge testing have been developed for the clinical laboratory (American Thoracic Society,
5 2000a). However, variations in methods for administering the bronchoconstricting agents may
6 substantially affect the results (Cockcroft et al., 2005).
7
8 3.1.3.1 Allergen Responsiveness
9
10 Clinical Studies of Allergen Responsiveness in Asthmatic Persons
11 In asthmatics, inhalation of an allergen to which a person is sensitized can cause
12 bronchoconstriction and increased airways inflammation, and this is an important cause of
13 asthma exacerbations. Aerosolized allergens can be used in controlled airways challenge testing
14 in the laboratory, either clinically to identify specific allergens to which the individual is
15 responsive or in research to investigate the pathogenesis of the airways allergic response or the
16 effectiveness of treatments. The degree of responsiveness is a function of the concentration of
17 inhaled allergen, the degree of sensitization as measured by the level of allergen-specific IgE,
18 and the degree of nonspecific airways responsiveness (Cockcroft and Davis, 2006).
19 It is difficult to predict the level of responsiveness to an allergen, and although rare,
20 severe bronchoconstriction can occur with inhalation of very low concentrations of allergen.
21 Allergen challenge testing, therefore, involves greater risk than nonspecific airways challenge
22 with drugs such as methacholine. Asthmatics may experience both an "early" response, with
23 declines in lung function within minutes after the challenge, and a "late" response, with a decline
24 in lung function hours after the exposure. The early response primarily reflects release of
25 histamine and other mediators by airways mast cells; the late response reflects enhanced airways
26 inflammation and mucous production. Responses to allergen challenge are typically measured as
27 changes in pulmonary function, such as declines in the forced expiratory volume in 1 s (FEVi).
28 However, the airways inflammatory response can also be assessed using BAL, induced sputum,
29 or exhaled breath condensate.
30 The potential for NO2 exposure to enhance responsiveness to allergen challenge in
31 asthmatics deserves special mention. Several recent studies, summarized in Annex Table
32 AX5.3-2, have addressed the question of whether low-level exposures to NO2, both at rest and
March 2008 3-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 with exercise, enhance the response to specific allergen challenge in mild asthmatics. These
2 recent studies involving allergen challenge suggest that NO2 may enhance the sensitivity to
3 allergen-induced decrements in lung function and increase the allergen-induced airways
4 inflammatory response. Figure 3.1-2 categorizes the allergen challenge studies as "positive,"
5 i.e., showing evidence for increased responses to allergen in association with NO2 exposure, or
6 "negative," with the exposure metric expressed as ppm-min. In comparing Figure 3.1-2 with
7 Figure 3.1-1, the enhancement of allergic responses in asthmatics occurs at exposure levels more
8 than an order of magnitude lower than those associated with airway inflammation in healthy
9 subjects. The dosimetry difference is even greater when considering that the allergen challenge
10 studies generally were performed at rest, while the airway inflammation studies in healthy
11 subjects were performed with intermittent exercise.
12 Tunnicliffe et al. (1994) exposed 8 subjects with mild asthma to 0.1- or 0.4-ppm NC>2 for
13 1 h at rest and reported that 0.4-ppm NO2 exposure slightly increased responsiveness to a fixed
14 dose of allergen during both the early and late phases of the response. In two U.K. studies
15 (Devalia et al., 1994; Rusznak et al., 1996), exposure to the combination of 0.4-ppm NC>2 and
16 0.2-ppm SC>2 increased responsiveness to subsequent allergen challenge in mild atopic
17 asthmatics, whereas neither pollutant alone altered allergen responsiveness.
18 A series of studies from the Karolinska Institute in Sweden have explored airways
19 responses to allergen challenge in asthmatics. Strand et al. (1997) demonstrated that single
20 30-min exposures to 0.26-ppm NO2 increased the late phase response to allergen challenge 4 h
21 after exposure. In a separate study (Strand et al., 1998), four daily repeated exposures to
22 0.26-ppm NC>2 for 30 min increased both the early and late phase responses to allergen. Barck
23 et al. (2002) used the same exposure and challenge protocol as used in the earlier Strand et al.
24 (1997) studies (0.26 ppm for 30 min, with allergen challenge 4-h after exposure) and performed
25 BAL 19-h after the allergen challenge to determine NC>2 effects on the allergen-induced
26 inflammatory response. NC>2 followed by allergen caused increases in the BAL recovery of
27 PMN and eosinophil cationic protein (ECP), with reduced volume of BAL fluid and reduced cell
28 viability, compared with air followed by allergen. ECP is released by degranulating eosinophils,
29 is toxic to respiratory epithelial cells, and is thought to play a role in the pathogenesis of airways
30 injury in asthma. These findings indicate that NC>2 exposure enhanced the airways inflammatory
31 response to allergen. Subsequently, Barck et al. (2005a) exposed 18 mild asthmatics to air or
March 2008 3-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 NO2 for 15 min on day 1, followed by two 15-min exposures separated by 1-h on day 2, with
2 allergen challenge after exposures on both days 1 and 2. Sputum was induced before exposure
(LS
1
(£)
~
-_
f9 )
r^
s^ V
M
3
0
©
_
I
50
(T) 10
_
x-v
k. 4 )
^ ~N
v-xn
Study
1. Barck et al. (2002)
2. Barck et al. (2005a)
3. Barck et al. (2005b)
4. Jenkins et al. (1999;
5. Jenkins etal. (1999)
6. Strand etal. (1997)
7. Strand etal. (1998)
ppm min
0.26 30
0.26 15
0.26 30
0.20 360
0.40 180
0.26 30
0.26 30
8. Tunnicliffe etal. (1994) 0.40 60
9. Tunnicliffe etal. (1994) 0.10 60
10. Wang etal. (1995a,b) 0.40 360
11. Witten etal. (2005)
I
0.40 180
I
100 150
NO2 ppm-minutes
Figure 3.1-2. Airways responsiveness to allergen challenge in asthmatic subjects following
a single exposure to NOi. Responsiveness was assessed using spirometric
(circles) and inflammatory (squares) endpoints. On the vertical axis, positive
and negative indicate studies finding statistically significant and non-
significant effects of NOi on group mean responsiveness to allergen,
respectively.
3 on day 1 and after exposures (morning of day 3). NC>2 + allergen, compared to air + allergen,
4 treatment resulted in increased levels of ECP in both sputum and blood and increased
5 myeloperoxidase levels in blood. A separate study examined NO2 effects on nasal responses to
6 nasal allergen challenge (Barck et al., 2005b). Single 30-min exposures to 0.26 ppm NC>2 did not
7 enhance nasal allergen responses. All exposures in the Karolinska Institute studies (Barck et al.,
8 2002, 2005a; Strand et al., 1997, 1998) used subjects at rest. These studies utilized an adequate
9 number of subjects, included air control exposures, randomized exposure order, and separated
March 2008
3-18
DRAFT-DO NOT QUOTE OR CITE
-------
1 exposures by at least 2 weeks. Together, they indicate that quite brief exposures to 0.26-ppm
2 NO2 can cause effects in allergen responsiveness in asthmatics.
3 The findings in these studies of allergen responsiveness may shed some light on the
4 variable results in earlier studies of NO2 effects on nonspecific airways responsiveness. It is
5 possible that some prior studies may have been variably confounded by environmental allergen
6 exposure, increasing the variability in subject responses to NC>2 and perhaps explaining some of
7 the inconsistent findings.
8 Several studies have been conducted using longer NO2 exposures. Wang et al. (1995a,b,
9 1999) found that more intense (0.4 ppm) and prolonged (6 h) NC>2 exposures enhanced allergen
10 responsiveness in the nasal mucosa in subjects with allergic rhinitis. Jenkins et al. (1999)
11 examined FEVi decrements and airways responsiveness to allergen in a group of mild, atopic
12 asthmatics. The subjects were exposed for 3-h to 0.4-ppm NC>2, 0.2-ppm Os, and 0.4-ppm
13 NC>2 + 0.2-ppm Os. The subjects were also exposed for 6-h to produce exposure concentrations
14 that would provide identical doses to the 3-h protocols (i.e., equivalent in concentration times
15 duration of exposure [C x T]). Significant increases in airways responsiveness to allergen
16 occurred following all the 3-h exposures, but not following the 6-h exposures. However, Witten
17 et al. (2005) did not find enhanced airways inflammation or a reduction in allergen provocative
18 dose that produces a 20% decrease in FEVi (PD2o-FEVi) with allergen challenge in 15 asthmatic
19 subjects allergic to house dust mite allergen who were exposed to air and 0.4 ppm NO2 for 3-h
20 with intermittent exercise. Allergen challenge was performed immediately after exposure, and
21 sputum induction was performed 6 and 26 h after the allergen challenge. There was no overall
22 effect of NC>2 on allergen responsiveness, although 3 subjects required a much smaller
23 concentration of allergen after NC>2 than after air exposure and were deemed to be NC>2
24 "responders." NC>2 exposure was surprisingly associated with a reduction in sputum eosinophils,
25 with no increase in allergen-induced neutrophilic inflammation.
26 The differing findings in these studies may relate in part to differences in timing of the
27 allergen challenge, the use of multiple- versus single-dose allergen challenge, the use of BAL
28 versus sputum induction, exercise versus rest during exposure, and differences in subject
29 susceptibility. Taken together, these studies suggest that NC>2 short-term exposures of less than
30 1 ppm enhance allergen responsiveness in some allergic asthmatics.
March 2008 3-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 Lastly, one study examined the effects on allergen responsiveness of exposure to traffic
2 exhaust in a tunnel (Svartengren et al., 2000). Twenty mild asthmatics sat in a stationary vehicle
3 within a busy tunnel for 30 min. Allergen challenge was performed 4 h later. The control
4 exposure was in a hotel room in a suburban area with low air pollution levels. Exposures were
5 separated by 4 weeks and the order was randomized. Median NO2 levels in the vehicle were
6 313 |ig/m3 (range, 203 to 462), or 0.166 ppm, (range, 0.106 to 0.242). PMio levels were
7 170 |ig/m3 (range, 103 to 613), and PM2.5 levels were 95 |ig/m3 (range, 61 to 128). Median NO2
8 levels outside the hotel were 11 |ig/m3 or 0.006 ppm. Subjects in the tunnel experienced
9 increased cough, and also reported awareness of noise and odors. More importantly, there was a
10 greater allergen-induced increase in specific airways resistance after the tunnel exposure than
11 after the control exposure (44% versus 31% respectively). Thoracic gas volume also was
12 increased to a greater degree after the tunnel exposure, suggesting increased gas trapping within
13 the lung. These findings were most pronounced in the subjects exposed to the highest levels of
14 NO2. This study suggests that exposure to traffic exhaust, and particularly the NO2 component,
15 increases allergen responsiveness in asthmatics, and the results fit well with the findings in
16 studies of clinical exposures of NO2 (Barck et al., 2002, 2005a). However, it was not possible to
17 blind the exposures, and the control exposure (hotel room, presumably quiet and relaxed) was
18 not well matched to the experimental exposure (vehicle, noisy, odorous). It remains possible that
19 factors other than NO2 contributed to, or were responsible for, the observed differences in
20 allergen responsiveness.
21 These recent studies involving allergen challenge suggest that NO2 may enhance the
22 sensitivity to allergen-induced decrements in lung function and increase the allergen-induced
23 airways inflammatory response. Enhancement of allergic responses in asthmatics occurs at
24 exposure levels of more than an order of magnitude lower than those associated with airways
25 inflammation in healthy subjects. The dosimetry difference is even greater when considering
26 that the allergen challenge studies generally were performed at rest, while the airways
27 inflammation studies in healthy subjects were performed with intermittent exercise.
28 Enhancement of allergen responses has been found at exposures of as low as 8 ppm-min, i.e.,
29 0.26 ppm for 30 min. Additional work is needed to understand more completely the exposure-
30 response characteristics of NO2 effects on allergen responses, as well as the effects of exercise,
March 2008 3-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 relationship to the severity of asthma, the role of asthma medications, and other clinical factors.
2 Additional animal and in vitro studies are needed to establish the precise mechanisms involved.
3
4 Toxicologic Studies of Allergen Responsiveness
5 Acute exposures of Brown Norway rats to NO2 at a concentration of 5 ppm for 3 h
6 resulted in increased specific immune response to house dust mite allergen and increased
7 immune-mediated pulmonary inflammation (Gilmour et al., 1996). Higher levels of antigen-
8 specific serum IgE, local IgA, IgG, and IgE were observed when rats were exposed to NO2 after
9 both the immunization and challenge phase but not after either the immunization or challenge
10 phase alone. Increases in the number of inflammatory cells in the lungs and lymphocyte
11 responsiveness to house dust mite allergen in the spleen and mediastinal lymph node were
12 observed. The authors concluded that this increased immune responsiveness to house dust mite
13 allergen may be the result of the increased lung permeability caused by NO2 exposure, enhancing
14 translocation of the antigen to local lymph nodes and circulation to other sites in the body.
15 A delayed bronchial response, seen as increased respiration rate, occurred in
16 NO2-exposed, Candida albicans-senshized guinea pigs 15 to 42 h after a challenge dose of
17 C. albicans (Kitabatake et al., 1995). Guinea pigs were given an intraperitoneal injection of
18 C. albicans, followed by a second injection 4 weeks later. Two weeks after the second injection,
19 the animals were given an inhalation exposure of killed C. albicans. Animals were also exposed
20 4 h/day to 4.76-ppm NO2 from the same day as the first injection of C. albicans., for a total of
21 30 exposures (5 days/week).
22 In a study with NO2-exposed rabbits, pulmonary function (lung resistance, dynamic
23 compliance) was not affected when immunized intraperitoneally within 24-h of birth until 3
24 months of age to either Alternaria tennis or house dust mite antigen. The rabbits were given
25 intraperitoneal injections once weekly for 1 month, and then every 2 weeks thereafter, and
26 exposed to 4-ppm NO2 for 2 h daily (Douglas et al., 1994).
27 To determine the effect of NO2 on allergenic airways responses in sensitized animals,
28 Hubbard et al. (2002) exposed ovalbumin (OVA)-sensitized mice to NO2 (0.7 or 5 ppm, 2 h/day
29 for 3 days) or air. While the air-exposed mice developed lower airways inflammation (increased
30 total BAL cellularity and increased eosinophil levels), the NO2-exposed mice had significantly
31 lower levels of eosinophils for both NO2 concentrations, with the greatest effect seen at the lower
32 NO2 concentration. These results were confirmed in a subsequent study (0.7-ppm NO2 for 3 or
March 2008 3-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 10 days) showing significant reductions in BAL cellularity and eosinophil levels for both time
2 points. In a similar study (Proust et al., 2002), mice were sensitized and challenged with OVA
3 and then exposed to NO2 (5 or 20 ppm, 3 h). The 20-ppm NO2 exposure resulted in a significant
4 increase in bronchopulmonary hyperreactivity 24 h after exposure, as compared to the OVA-air
5 and 5-ppm NO2 group. However, exposure to 5-ppm NO2 resulted in a marked reduction in
6 bronchopulmonary hyperreactivity as compared to both the 20-ppm NO2 and OVA-air groups.
7 By 72 h, bronchopulmonary hyperreactivity in all groups were comparable. The measurement of
8 fibronectin in the BAL fluid was used as a marker of epithelial permeability. At 24 h after
9 exposure, fibronectin levels were significantly higher in the 20-ppm NO2 group as compared to
10 both the 5-ppm NO2 and air groups. However, fibronectin levels in the 5-ppm NO2 group were
11 significantly lower than the OVA-air group. After 72 h, there was no difference in fibronectin
12 levels between the OVA-air and 5-ppm NO2 groups, while fibronectin levels of the 20-ppm NO2
13 group remained significantly higher than the 5-ppm NO2 group. The recruitment of PMNs as
14 measured in the BAL fluid at 24 h postexposure, revealed a dose-dependent increase reaching
15 significance only with the 20-ppm NO2 exposure. By 72 h, all groups were comparable. In
16 contrast, the recruitment of eosinophils, as measured in the BAL fluid, showed no significant
17 differences between groups at the 24 h time point, yet at the 72-h point, eosinophils were
18 significantly decreased in the 5 ppm NO2 group as compared to OVA-air group. Eosinophil
19 peroxidase (EPO) in the lung tissue showed a similar trend with NO2 exposure reducing the EPO
20 levels as compared to OVA-air controls. At 24 h, EPO was significantly lower in the 5- and
21 20-ppm NO2 groups as compared to the OVA-air group, while at 72 h, only the 5-ppm NO2
22 group was significantly lower. IL-5 was measured in the BAL fluid, and the 5-ppm NO2 group
23 was significantly lower in IL-5 than all other groups, and the 20-ppm NO2 was significantly
24 higher.
25
26 3.1.3.2 Nonspecific Responsiveness
27
28 Nonspecific Responsiveness in Healthy Individuals
29 Several observations indicate that NO2 exposures in the range of 1.5 to 2.0 ppm cause
30 small but significant increases in airways responsiveness in healthy subjects. Mohsenin (1988)
31 found that a 1-h exposure to 2-ppm NO2 increased responsiveness to methacholine, as measured
32 by changes in specific airways conductance, without directly affecting lung function.
March 2008 3-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 Furthermore, pretreatment with ascorbic acid prevented the NO2-induced increase in airways
2 responsiveness (Mohsenin, 1987a). A mild increase in responsiveness to carbachol was
3 observed following a 3-h exposure to 1.5-ppm NO2, but not to intermittent peaks of 2.0 ppm
4 (Frampton et al., 1991). Thus, the lower threshold concentration of NO2 for causing increases
5 in nonspecific airways responsiveness in healthy subjects appears to be in the 1- to 2-ppm range.
6
7 Nonspecific Responsiveness in Asthmatic Individuals
8 The 1993 AQCD for Oxides of Nitrogen reported results from some early studies that
9 suggested that NO2 might enhance subsequent responsiveness to challenge was observed in
10 some, but not all studies, at relatively low NO2 concentrations within the range of 0.2 to 0.3 ppm.
11 Appearing in Tables 15-9 and 15-10 of the 1993 AQCD, the meta-analysis by Folinsbee (1992)
12 also provided suggestive evidence of increased airways responsiveness in 63% of asthmatics
13 exposed to a NO2 concentration of only 0.1 ppm for 1 h during rest. However, numerous studies
14 had not reported independent effects of NO2 on lung function in asthmatic individuals.
15 Roger et al. (1990), in a comprehensive, concentration-response experiment, were unable
16 to confirm the results of a pilot study suggesting airways responses occur in asthmatic subjects.
17 Twenty-one male asthmatics exposed to NO2 at 0.15, 0.30, or 0.60 ppm for 75 min did not
18 experience significant effects on lung function or airways responsiveness compared with air
19 exposure. Bylin et al. (1985) found significantly increased bronchial responsiveness to histamine
20 challenge compared with sham exposure in 8 atopic asthmatics exposed to 0.30-ppm NO2 for
21 20 min. Five of 8 asthmatics demonstrated increased reactivity, while 3 subjects showed no
22 change, as assessed by specific airways resistance. Mohsenin (1987b) reported enhanced
23 responsiveness to methacholine in 8 asthmatic subjects exposed to 0.50-ppm NO2 at rest for 1 h;
24 airways responsiveness was measured by partial expiratory flow rates at 40% vital capacity,
25 which may have increased the sensitivity for detecting small changes in airways responsiveness.
26 Torres and Magnussen (1991) found no effects on lung function or methacholine responsiveness
27 in 11 patients with mild asthma after exposure to 0.25-ppm NO2 for 30 min with 10 min of
28 exercise. Strand et al. (1996) performed a series of studies in mild asthmatics exposed to
29 0.26 ppm for 30 min and found increased responsiveness to histamine as well as to allergen
30 challenge.
31 The effects of NO2 exposure on SO2-induced bronchoconstriction have been examined,
32 but with inconsistent results. Torres and Magnussen (1990) found an increase in airways
March 2008 3-23 DRAFT-DO NOT QUOTE OR CITE
-------
1 responsiveness to 862 in asthmatic subjects following exposure to 0.25-ppm NC>2 for 30 min at
2 rest; yet Rubenstein et al. (1990) found no change in responsiveness to 862 inhalation following
3 exposure of asthmatics to 0.30-ppm NC>2 for 30 min with 20 min of exercise.
4 The varied results of these studies have not been satisfactorily explained. It is evident
5 that a wide range of responses occurs among asthmatics exposed to NC>2. This variation may in
6 part reflect differences in subjects and exposure protocols: mouthpiece versus chamber,
7 obstructed versus non-obstructed asthmatics, rest versus exercise, and varying use of
8 medication(s) among subjects. Indeed, via meta-analysis, Folinsbee (1992) found that airways
9 responsiveness was greater in asthmatics exposed to NC>2 at rest than during exercise. Following
10 NC>2 exposures of between 0.2- and 0.3-ppm NC>2, only 52% of subjects exposed with exercise
11 had increased responsiveness, whereas 76% of subjects had increased responsiveness in
12 protocols using resting exposures. Identification of factors that predispose to NC>2
13 responsiveness also is needed. These studies have typically involved volunteers with mild
14 asthma; data are lacking from more severely affected asthmatics, who may be more susceptible.
15 Overall, there is suggestive evidence that short-term exposures to NC>2 at outdoor ambient
16 concentrations (<0.3 ppm) alters lung function or nonspecific airways responsiveness in people
17 with mild asthma. However, it remains possible that more severe asthmatics, or individuals with
18 particular sensitivity to NC>2 airways effects, would experience reductions in lung function or
19 increased airways responsiveness when exercising outdoors at NC>2 concentrations of <0.3 ppm.
20
21 Toxicological Studies of Airways Responsiveness
22 In the previous review, toxicological evidence supported a conclusion that airways
23 responsiveness was one of the key health responses to NC>2 exposure. A number of recent
24 animal studies have also reported airways responsiveness with NC>2 exposure. Overall, many
25 studies have demonstrated the ability of NO2 exposure to increase bronchial sensitivity to various
26 challenge agents, although the mechanisms for this response are not fully known.
27 Kobayashi and Miura (1995) studied the concentration- and time-dependency of airways
28 hyperresponsiveness to inhaled histamine aerosol in guinea pigs exposed subchronically to NO2.
29 In one experiment, guinea pigs were exposed by inhalation to 0-, 0.06-, 0.5-, or 4.0-ppm NO2,
30 24 h/day for 6 or 12 weeks. Immediately following the last exposure, airways responsiveness
31 was assessed by measurement of specific airways resistance as a function of increasing
32 concentrations of histamine aerosol. Animals exposed to 4-ppm NO2 for 6 weeks exhibited
March 2008 3-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 increased airways response to inhaled histamine aerosol; airways response at 12 weeks was not
2 determined. No effects were observed at the lower exposure levels. In another experiment
3 conducted in this study (Kobayashi and Miura, 1995), guinea pigs were exposed by inhalation to
4 0-, 1.0-, 2.0-, or 4.0-ppm NO2, 24 h/day for 6 or 12 weeks, and the airways hyperresponsiveness
5 was determined. Increased hyperresponsiveness to inhaled histamine was observed in animals
6 exposed to 4 ppm for 6 weeks, 2 ppm for 6 and 12 weeks, and 1 ppm for 12 weeks only. The
7 results also showed that at 1- or 2-ppm NC>2, airways hyperresponsiveness developed to a higher
8 degree with the passage of time. Higher concentrations of NO2 were found to induce airways
9 hyperresponsiveness faster compared to lower concentrations. When the specific airways
10 resistance was compared to values determined 1 week prior to initiation of the NO2 exposure,
11 values were increased in the 2.0- and 4.0-ppm animals at 12 weeks only. Specific airways
12 resistance was also increased to a higher degree with the passage of time.
13
14 3.1.3.3 Summary of Evidence on the Effect of Short-Term Exposure to NOi on Airways
15 Responsiveness
16 The evidence from human and animal experimental studies provides suggestive evidence
17 for increased airways responsiveness to specific allergen challenges following NC>2 exposure.
18 Recent human clinical studies involving allergen challenge suggest that NC>2 exposure may
19 enhance the sensitivity to allergen-induced decrements in lung function and increase the
20 allergen-induced airway inflammatory response at exposures of as low as 0.26-ppm NC>2 for 30
21 min (Figure 3.1-2). The inflammatory responses to the allergen challenge were not accompanied
22 by any changes in pulmonary function or subjective symptoms. Increased immune-mediated
23 pulmonary inflammation was also observed in rats exposed to house dust mite allergen following
24 exposure to 5-ppm NC>2 for 3 h.
25 Exposure to NO2 also has been found to enhance the inherent responsiveness of the
26 airways to subsequent nonspecific challenges in human clinical studies; however, the results are
27 less consistent than those of animal toxicologic studies. In general, small but significant
28 increases in nonspecific airways responsiveness were observed in the range of 1.5 to 2.0 ppm for
29 3 h in healthy adults and between 0.2- and 0.3-ppm NC>2 for 30 min for asthmatics, but a wide
30 range of responses were observed, particularly among the asthmatics. Subchronic exposures (6
31 to 12 weeks) of animals to NC>2 also increase responsiveness to nonspecific challenges at 1- to
32 4-ppm NO2.
March 2008 3-25 DRAFT-DO NOT QUOTE OR CITE
-------
1 There is inconsistency in the results of the human studies; with some, but not all studies,
2 finding increased responsiveness following exposure to NC>2. However, a variety of factors are
3 recognized that may lead to this apparent inconsistency. For instance, responsiveness has been
4 observed to be greater following resting than exercising exposures to NO2, despite the greater
5 dose of NC>2 to the respiratory tract during exercise. In addition, the methods for administering
6 the bronchoconstricting challenge agents and degree of sensitization to specific allergen also are
7 recognized to affect responsiveness (Cockcroft et al., 2005; Cockcroft and Davis, 2006).
8
9 3.1.4 Effects of Short-Term NOi Exposure on Respiratory Symptoms
10 Since the 1993 AQCD, additional studies have reported health effects associated with
11 NC>2 from indoor exposure, personal exposure, and ambient concentration studies. The
12 following section characterizes the results of these studies.
13
14 3.1.4.1 Indoor and Personal NOi Exposure and Respiratory Outcomes
15 Indoor NC>2 exposure studies may differ from ambient exposure in relation to pattern,
16 levels, and associated copollutants (see Annex Table AX6.3-1 for details). Samet and Bell
17 (2004) state that while "evidence from studies of outdoor air pollution cannot readily isolate an
18 effect of NO2 because of its contribution to the formation of secondary particles and Os,
19 observational studies of exposure indoors can test hypotheses related to NC>2 specifically
20 although confounding by combustion sources in the home is a concern."
21 Most of the studies conducted since 1993 have taken place in Australia and attempted to
22 capture indoor exposures (with passive diffusion badges) from both cooking and heating sources
23 in homes and schools (Pilotto et al., 1997a, 2004; Rodriguez et al., 2007; Garrett et al., 1998;
24 Smith et al., 2000). Several indoor exposure studies have also been conducted in the United
25 States (Kattan et al., 2007; Belanger et al., 2006; van Strien et al., 2004), Europe (Farrow et al.,
26 1997; Simoni et al., 2002, 2004), and Singapore (Ng et al., 2001). The results from these studies
27 are summarized in Annex Table AX6.3-1.
28 One intervention study provides strong evidence of a detrimental effect of exposure to
29 indoor levels of NO2. Pilotto et al. (2004) conducted a randomized intervention study of
30 respiratory symptoms of asthmatic children in Australia before and after selective replacement of
31 unflued gas heaters in schools. In the study, 18 schools using unflued gas heaters were randomly
32 allocated to have an electric heater (n = 4) or a flued gas heater (n = 4) installed or to retain their
March 2008 3-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 original heaters (n = 10). Changes to the heating systems were disguised as routine maintenance
2 to prevent bias in reporting of symptoms. Children were eligible for the study if they had
3 physician-diagnosed asthma and no unflued heater in their home. For the 114 children enrolled,
4 symptoms were recorded daily and reported in biweekly telephone interviews during 12 weeks
5 in the winter. Passive diffusion badges were used to measure NO2 exposure in classrooms
6 (6 h/day) and in the children's homes. Schools in the intervention group (with new heaters)
7 averaged overall means (SD) of 15.5 (6.6) ppb NC>2, while control schools (with unflued heaters)
8 averaged 47.0 (26.8) ppb. Exposure to NO2 in the children's homes was quite variable but with
9 similar mean levels. Levels at homes for the intervention group were 13.7 (19.3) ppb and 14.6
10 (21.5) ppb for the control group. Children attending intervention schools had significant
11 reductions in several symptoms (see Table 3.1-2): difficulty breathing during the day (RR = 0.41
12 [95% CI: 0.07, 0.98]) and at night (RR = 0.32 [95% CI: 0.14, 0.69]); chest tightness during the
13 day (RR = 0.45 [95% CI: 0.25, 0.81]) and at night (RR= 0.59 [95% CI: 0.28, 1.29]); and
14 asthma attacks during the day (RR = 0.39 [95% CI: 0.17, 0.93]).
15 Samet and Bell (2004) state that Pilotto et al. (2004) provide persuasive evidence of an
16 association between exposure to NO2 from classroom heaters and the respiratory health of
17 children with asthma and further that the intervention study design alleviates some potential
18 limitations of observational studies. The two groups of children studied had similar baseline
19 characteristics. In addition, the concentrations in the home environment were similar for the two
20 groups, implying that exposure at school was likely to be the primary determinant of a difference
21 in indoor NC>2 exposure between the two groups. It is, however, possible that confounding by
22 particle emissions, particularly ultrafine particles, may be present.
23 In an earlier study of the health effects of unflued gas heaters on wintertime respiratory
24 symptoms of 388 Australian schoolchildren, Pilotto et al. (1997a) measured NC>2 in 41
25 classrooms in 8 schools, with half using unflued gas heaters and half using electric heat.
26 Although similar methods were used to measure NC>2 levels (passive diffusion badge monitors
27 exposed for 6 h at a time), there were three major differences between this study and the Pilotto
28 et al. (2004) study: (1) the 1997 study was not a randomized trial, (2) enrollment was not
29 restricted to asthmatic children, and (3) enrollment was not restricted to children from homes
March 2008 3-27 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3.1-2. MEAN RATES (SD) PER 100 DAYS AT RISK AND UNADJUSTED
RATE RATIO (RR)* FOR SYMPTOMS/ACTIVITIES OVER 12 WEEKS DURING
THE WINTER HEATING PERIOD
Symptom/Activity
Wheeze during the day
Wheeze during the night
Difficulty breathing during the day
Difficulty breathing during the night
Chest tightness during the day
Chest tightness during the night
Cough during the day
Cough during the night
Difficulty breathing after exercise
Asthma attacks during the day
Asthma attacks during the night
Missed school due to asthma
Visit to health care facilities due to asthma
Taking any asthma medication
Taking any reliever
Taking any preventer
Mean Rate
Intervention
(n = 45)
4.9(15.2)
2.2 (5.6)
2.2 (3.7)
0.8 (2.2)
2.3 (4.3)
1.5(3.3)
17.5(21.5)
10.7 (16.6)
3.8 (7.4)
1.1 (2.3)
0.7(2.1)
1.6 (2.0)
0.5 (0.8)
26.9 (36.7)
13.8(23.2)
26.2(40.1)
Mean Rate
Control
(n = 69)
5.1(10.5)
2.3 (5.5)
5.4(12.1)
2.6 (6.9)
5.1 (9.9)
2.5 (6.2)
13.7(13.7)
11.6(12.4)
6.4(13.9)
2.7(5.3)
1.8(3.8)
1.2 (2.8)
0.8(1.2)
34.6(37.1)
22.4 (28.8)
29.9 (42.2)
RR
0.95
0.94
0.41
0.32
0.45
0.59
1.27
0.92
0.59
0.39
0.38
1.34
0.60
0.77
0.62
0.87
(95% CI)
(0.45, 2.01)
(0.36, 2.50)
(0.07, 0.98)
(0.14,0.69)
(0.25, 0.81)
(0.28, 1.29)
(0.81, 2.00)
(0.49, 1.73)
(0.31, 1.13)
(0.17,0.93)
(0.13, 1.07)
(0.68, 2.60)
(0.35, 1.03)
(0.49, 1.21)
(0.31, 1.25)
(0.53, 1.44)
Following adjustment for hay fever and parental education at baseline, results remained substantially unchanged except that difficulty
breathing during the day assumed borderline significance (RR = 0.46: 95% CI: 0.19, 1.08) while the reduction in asthma attacks during the
night reached statistical significance (RR = 0.33; 95% CI: 0.13, 0.84).
Source: Adapted from Pilotto et al. (2004).
1 without unflued gas heaters. In Pilotto et al. (1997a), only children from nonsmoking homes
2 were enrolled and a subset of children (n = 121) living in homes with unflued gas heaters were
3 given badges to be used at home. Each child's parents recorded symptoms daily. Children were
4 classified into low- and high-exposure groups based on their measured exposure at school, their
5 measured exposure at home (if they lived in homes with unflued gas heaters), or their reported
6 use of electric heat at home. Maximum hourly concentrations in these classrooms each day over
7 2 weeks of hourly monitoring were highly correlated with their corresponding 6-h concentrations
8 measured over the same 2 weeks (r = 0.85). Hourly peaks of NC>2 on the order of >80 ppb were
9 associated with 6-h average levels of approximately >40 ppb. They inferred that children in
10 classrooms with unflued gas heaters that had 6-h average levels of >40 ppb were experiencing
11 approximately 4-fold or higher 1-h peaks of exposure than the NO2 levels experienced by
12 children who had no gas exposure (6-h average levels of 20 ppb). The importance of this study
March 2008
3-28
DRAFT-DO NOT QUOTE OR CITE
-------
1 is that it examines the effect of repeated peaks over time as have been used in the toxicological
2 infectivity studies (e.g., Miller et al., 1987) that were noted earlier in Section 3.1.2.
3 Pilotto et al. (1997a) reported that during the winter heating season, children in the high-
4 exposure category (NO2 > 40 ppb) had higher rates of sore throat, colds, and absenteeism than all
5 other children. In models adjusted for personal risk factors including asthma, allergies, and
6 geographic area, classroom NO2 level and school absence were significantly associated (odds
7 ratio [OR] = 1.92 [95% CI: 1.13, 3.25]). Increased likelihood of individual respiratory
8 symptoms was not significantly associated with classroom NC>2 level (e.g., cough with phlegm
9 adjusted OR = 1.28 [95% CI: 0.76, 2.15]). Exposure-response relationships are illustrated in
10 Figure 3.1-3 for symptom rates for cough with phlegm and proportion of children absent from
11 school. Statistically significant positive exposure-response trends were found for mean rates for
12 cough with phlegm (p = 0.04, adjusted for confounders) and proportion of children absent from
13 school (p = 0.002) using mixed models allowing for correlation between children within
14 classrooms. Pilotto et al. (1997b) noted that this study "provides evidence that short-term
15 exposure to the peak levels of NO2 produced by unflued gas appliances affects respiratory health
16 and that the significant dose-response relationship seen with increasing NO2 exposure
17 strengthens the evidence for a cause-effect relationship."
18 In a cross-sectional survey of 344 children in Australia, Ponsonby et al. (2001) used
19 passive gas samplers to measure personal exposure to NO2. Personal badges were pinned to a
20 child's clothing at the end of each school day and removed when the child arrived at school the
21 next day. School exposures were measured with passive samplers placed in each child's
22 classroom. Sampling took place over two consecutive days. Mean (SD) personal exposure was
23 10.4(11.1) ppb and mean total NO2 exposure (personal plus schoolroom) was 10.1 (8.6) ppb. Of
24 the health outcomes measured (recent wheeze, asthma ever, lung function measured when NO2
25 sampling stopped), only the forced expiratory volume in 1 s/forced vital capacity (FEVi/FVC)
26 ratio following cold air challenge was significantly associated with NO2 levels measured with the
27 personal badges (-0.12 [95% CI: -0.23, -0.01]) per 1-ppb increase in personal exposure). In
28 Finland, Mukula et al. (1999, 2000) studied 162 preschool-age children. Mukula et al. (2000)
29 used passive monitors exposed for 1-week periods over the course of 13 weeks both indoors
March 2008 3-29 DRAFT-DO NOT QUOTE OR CITE
-------
2 0.12
I 0.10
Q.
1.0.08
If!
o> 0.06
o>
J-0.04
j= 0.02
o>
«5 o.oo
•
-
-
i
-
[A]
<
t
t
)
. • "
» ^^^ -1
__.---- *< P "*"""*
»"""""
<40* 'intermed*' 40-60 ' ' 60-80 ' ' 80-100 ' ' >100 '
n=105 n=39 n=46 n=12 n=94 n=92
Nitrogen dioxide, ppb
0.10r
o
o
M
O
^
<*••
C
0)
100
n=105 n=39 n=46 n=12 n=94 n=92
Nitrogen dioxide, ppb
Figure 3.1-3. Geometric mean symptom rates (95% confidence intervals) for cough with
phlegm (panel A) and proportions (95% confidence intervals) of children
absent from school for at least 1 day (panel B) during the winter heating
period grouped by estimated NO2 exposure at home and at school (n =
number of children at that NOi level). Group means estimated using mixed
models. * "<40 ppb" group (n = 105) includes children from electrically
heated schools while the "Intermed" group (n = 39) includes children from
unflued gas heater heaters where the exposures were consistently below
40 ppb. Both groups of children did not have exposure to gas combustion
at home.
Source: Adapted from Pilotto et al. (1997a).
and outdoors and on the clothing of preschool children attending eight day care centers in
Helsinki. The only significant association between personal NC>2 measurements and symptoms
was for cough during the winter (RR =1.86 [95% CI: 1.15, 3.02] for NO2 at level above
March 2008
3-30
DRAFT-DO NOT QUOTE OR CITE
-------
1 27.5 |ig/m3 [14.5 ppb]). Similar results were obtained when data were analyzed unstratified by
2 season, but including a factor for season (RR = 1.52 [95% CI: 1.00, 2.31] for NO2 at levels
3 above 27.5 |ig/m3 [14.5 ppb], Mukala et al., 1999).
4 One recent birth cohort study in the United States measured indoor exposure to NO2
5 (Belanger et al., 2006; van Strien et al., 2004). Families were eligible for this study if they had a
6 child with physician-diagnosed asthma (asthmatic sibling) and a newborn infant (birth cohort
7 subject). NO2 levels were measured using Palmes tubes left in the homes for 2 weeks. Higher
8 levels of NO2 were measured in homes with gas stoves (mean [SD], 26 [18] ppb) than in homes
9 with electric ranges (9 [9] ppb). Children living in multifamily homes were exposed to higher
10 NO2 (23 [17] ppb) than children in single-family homes (10 [12] ppb). The authors examined
11 associations between NO2 concentrations and respiratory symptoms experienced by the
12 asthmatic sibling in the month prior to sampling (Belanger et al., 2006). For children living in
13 multifamily homes, each 20-ppb increase in NO2 concentration increased the likelihood
14 of any wheeze or chest tightness (OR for wheeze = 1.52 [95% CI: 1.04, 2.21]; OR for chest
15 tightness = 1.61 [95% CI: 1.04, 2.49]) as well as increasing the risk of suffering additional days
16 of symptoms. No significant associations were found between level of NO2 and symptoms for
17 children living in single-family homes. The authors suggested that the low levels of exposure
18 may have been responsible for the lack of association observed in single-family homes. In these
19 same families, van Strien et al. (2004) compared the measured NO2 concentrations with
20 respiratory symptoms experienced by the birth cohort infants during the first year of life.
21 Although wheeze was not associated with NO2 concentration, persistent cough was associated
22 with increasing NO2 concentration in an exposure-response relationship (Figure 3.1-4)
23 (van Strien et al., 2004).
24 Results from a recent analysis of a subset of 469 asthmatic children enrolled in the
25 National Cooperative Inner City Asthma Study (NCICAS) (Kattan et al., 2007) where household
26 measurements of NO2 levels were also available are consistent with those described above for
27 Belanger et al. (2006). The median level of indoor NO2, measured with Palmes tubes left for
28 7 days, was 29.8 ppb, with median level in homes with gas stoves (31.4 ppb) significantly higher
29 than levels in homes with electric stoves (15.9 ppb). Associations between exposure to high
30 levels of NO2 and symptoms in the previous 2 weeks or peak flow of <80% predicted were
March 2008 3-31 DRAFT-DO NOT QUOTE OR CITE
-------
4.0
3.5
I 3-°
.1 2.5
1 2.0
1.5
1.0
0.5
. a. persistent cough
4.0
3.5
3,0
2.5
2.0
1.5
1 0
n *
b. shortness of breath
-
-
-
-
i
1
i
i
<5 5-10 10-17 > 17
NO, concentration quartile (ppb)
<5 5-10 10-17 >17
NOj concentration quartile (ppb)
Figure 3.1-4. Adjusted association of increasing indoor NOi concentrations with number
of days with persistent cough (panel a) or shortness of breath (panel b) for
762 infants during the first year of life. Relative risks from Poisson
regression analyses adjusted for confounders.
Source: Adapted from van Strien et al. (2004).
1 examined with models that adjusted for study site, gender, medication use, household smoking,
2 and SES variables and were stratified by season or by atopic status. Among the subset of 76
3 children without positive skin tests, the adjusted risk ratio (95% CI) for asthma symptoms was
4 1.75 (95% CI: 1.10, 2.78) for those with higher NO2 exposure. Among the 317 children with
5 NO2 measured in the cold season, the risk ratio for a peak flow measurement of <80% predicted
6 was 1.46 (95% CI: 1.07, 1.97). One limitation of the study is that the "high" NO2 level was
7 defined vaguely as approaching the U.S. Environmental Protection Agency (EPA) National
8 Ambient Air Quality Standards (NAAQS) level of 53 ppb.
9 Other studies have also collected personal exposure data for NO2. Nitschke et al. (2006)
10 used passive diffusion badges for measuring NO2 exposures in 6-h increments at home and
11 school for 174 asthmatic children in Australia. School and home measurements were based on
12 three consecutive days of sampling. The maximum of 9 days of sampling (for 6 h each day) NO2
13 value was selected as the representative daily exposure for exposure-response analyses. Children
14 kept a daily record of respiratory symptoms for the 12-week study period. Significant
15 associations were found between the maximum NO2 level at school or home and respiratory
16 symptom rates, though the exposure-response curve indicated that the major difference in
17 respiratory symptoms rates were between NO2 exposures of >80 ppb (see Annex Table
18 AX 6.3-1).
March 2008
3-32
DRAFT-DO NOT QUOTE OR CITE
-------
1 An important consideration in the evaluation of the indoor exposure studies is that NOx is
2 part of a complex mixture of chemicals emitted from unvented gas heaters. In addition to NO
3 and NO2, indoor combustion sources such as unvented gas heaters emit other pollutants that are
4 present in the fuel or are formed during combustion. These pollutants include carbon dioxide
5 (CC>2), carbon monoxide (CO), formaldehyde (HCHO) and other volatile organic compounds
6 (VOCs), polycyclic aromatic hydrocarbons (PAHs), and PM, particularly ultrafme particles, as
7 described in Section 2.5.8.3. The studies of unvented heaters or gas stoves did not measure
8 indoor concentrations of other combustion-related emissions. Unvented combustion is a
9 potential source of ultrafme particles. High numbers of ultrafme particles, along with NO2, are
10 generated during the operation of gas heaters, gas stoves, and during cooking (Dennekamp et al.,
11 2001; Wallace et al., 2004). It is possible that the improved respiratory symptoms observed in
12 the Pilotto et al. (2004) intervention study were related to reductions in ultrafme particle
13 exposure, other gaseous emissions, or the pollutant mix. The findings of these recent indoor and
14 personal exposure studies, combined with studies available in the previous AQCD, provide
15 evidence that NO2 exposure is associated with respiratory effects. These studies provide a
16 potential bridge between epidemiologic studies using ambient concentrations from centrally
17 located monitors and controlled human exposure studies, as discussed in the previous sections,
18 and provide some evidence of coherence for respiratory effects.
19
20 3.1.4.2 Ambient NOi Exposure and Respiratory Symptoms
21 Since the 1993 AQCD, results have been published from several single- and multicity
22 studies investigating ambient NO2 levels, including three large longitudinal studies in urban
23 areas covering the continental United States and southern Ontario: the Harvard Six Cities study
24 (Six Cities; Schwartz et al., 1994), the National Cooperative Inner-City Asthma Study (NCICAS;
25 Mortimer et al., 2002), and the Childhood Asthma Management Program (CAMP; Schildcrout
26 et al., 2006). Because of similar analytic techniques (i.e., multistaged modeling and generalized
27 estimating equations [GEE]), one strength of all three of these studies is that, as Schildcrout et al.
28 (2006) stated, they could each be considered as a meta-analysis of "large, within-city panel
29 studies" without some of the limitations associated with meta-analyses, e.g., "between-study
30 heterogeneity and obvious publication bias."
31 The report from the Six Cities study includes 1,844 schoolchildren, followed for 1 year
32 (Schwartz et al., 1994). Symptoms (in 13 categories, analyzed as cough, lower or upper
March 2008 3-33 DRAFT-DO NOT QUOTE OR CITE
-------
1 respiratory symptoms), were recorded daily. Cities included Watertown, MA, Baltimore, MD,
2 Kingston-Harriman, TN, Steubenville, OH, Topeka, KS, and Portage, WI. In Mortimer et al.
3 (2002), 864 asthmatic children from the eight NCICAS cities (New York City, NY, Baltimore,
4 MD, Washington, DC, Cleveland, OH, Detroit, MI, St Louis, MO, and Chicago, IL) were
5 followed daily for four 2-week periods over the course of 9 months. Morning and evening
6 asthma symptoms (analyzed as none versus any) and peak flow were recorded. Schildcrout et al.
7 (2006) reported on 990 asthmatic children living within 50 miles of one of 31 NO2 monitors
8 located in eight North American cities, seven of which included data for NO2 (Boston, MA,
9 Baltimore, MD, Toronto, ON, St. Louis, MO, Denver, CO, Albuquerque, NM, and San Diego,
10 CA). Symptoms (analyzed as none versus any per day) and rescue medication use (analyzed as
11 number of uses per day) were recorded daily such that each subject had an approximate average
12 of 2 months of data. All three studies found significant associations between ambient NO2
13 concentrations and risk of respiratory symptoms in children (Schwartz et al., 1994), and in
14 particular, asthmatic children (Mortimer et al., 2002; Schildcrout et al., 2006).
15 In Schwartz et al. (1994), a significant association was found between a 4-day mean of
16 NO2 exposure and incidence of cough among all children in single-pollutant models: the odds
17 ratio (OR) standardized to a 20-ppb increase in NO2 was OR = 1.61 (95% CI: 1.08, 2.43).
18 Cough incidence was not significantly associated with NO2 on the previous day. The local
19 nonparametric smooth of the 4-day mean concentration showed increased cough incidence up to
20 approximately the mean concentration (-13 ppb) (p = 0.01), after which no further increase was
21 observed. The significant association between cough and 4-day mean NO2 remained unchanged
22 in models that included O3, but was attenuated in two-pollutant models including PMio (OR for
23 20-ppb increase in NO2= 1.37 [95% CI: 0.88, 2.13]) or SO2 (OR = 1.42 [95% CI: 0.90,2.28]).
24 In Mortimer et al. (2002), the greatest effect of the pollutants studied for morning
25 symptoms was for a 6-day moving average. For increased NO2, the risk of any asthma
26 symptoms (cough, wheeze, shortness of breath) among the asthmatic children in the NCICAS
27 was somewhat higher than for the healthy children in the Six Cities study: OR = 1.48 (95% CI:
28 1.02, 2.16). Effects were generally robust in multipollutant models that included O3 (OR for
29 20-ppb increase in NO2 = 1.40 [95% CI: 0.93, 2.09]), O3 and SO2 (OR for NO2 = 1.31 [95% CI:
30 0.87, 2.09]), or O3, SO2, and PM with an aerodynamic diameter of < 10 |im (PMio) (OR for
31 NO2 = 1.45 [95% CI: 0.63, 3.34]).
March 2008 3-34 DRAFT-DO NOT QUOTE OR CITE
-------
1 In the CAMP study (Schildcrout et al., 2006), the strongest association between NO2 and
2 increased risk of cough was found for a 2-day lag: each 20-ppb increase in NO2 occurring 2 days
3 before measurement increased risk of cough (OR = 1.09 [95% CI: 1.03, 1.15]). Joint-pollutant
4 models including CO, PMio, or SO2 produced similar results (see Figure 3.1-5, panel A).
5 Further, increased NO2 exposure was associated with increased use of rescue medication in the
6 CAMP study, with the strongest association for a 2-day lag, both for single- and joint-pollutant
7 models (e.g., for an increase of 20-ppb NO2 in the single-pollutant model, the RR for increased
8 inhaler usage was 1.05 (95% CI: 1.01, 1.09) (See Figure 3.1-5, panel B).
9 Single-city studies also provide updated information to the 1993 AQCD, particularly with
10 regard to children. Two 3-month-long panel studies recruited asthmatic children from one
11 outpatient clinic in Paris: one study followed 84 children in the fall of 1992 (Segala et al., 1998),
12 and the other followed 82 children during the winter of 1996 (Just et al., 2002). Significant
13 associations were observed between respiratory symptoms and level of NO2 (See Annex Table
14 AX6.3-2). No multipollutant analyses were conducted. In metropolitan Sydney, 148 children
15 with a history of wheeze were followed for 11 months (Jalaludin et al., 2004). Daily symptoms,
16 medication use, and doctor visits were examined. Associations were found between increased
17 likelihood of wet cough and each 20-ppb increase in NO2 (OR =1.13 [95% CI: 1.00, 1.26]).
18 The authors reported that estimates did not change in multipollutant models including Os or
19 PMio. Ward et al. (2002) examined respiratory symptoms in a panel of 162 children in the
20 United Kingdom. No significant associations were reported for the winter period, but a
21 significant association was reported for the summer period for cough and NO2 (lag 0; OR = 1.09
22 [95% CI: 1.17, 1.01]).
23 Another Australian study includes a large number of children (n = 263) at risk for
24 developing allergy who were followed for 5 years (Rodriguez et al., 2007). Daily air pollutant
25 concentrations, including those for NO2, were averaged over 10 monitoring sites in the Perth
26 metropolitan region. Mean level of 24-h NO2 for the 8-year study period was 7 ppb (range
27 0-24 ppb). Significant associations were found between same-day level of NO2 (both 1- and
28 24-h avg) and cough (OR 1.0005 [95% CI: 1.0000, 1.0011]) per 20 ppb increase in 24-h avg
29 NO2). No multipollutant models were presented.
30 Boezen et al. (1999) reported associations between ambient NO2 exposure and lower
31 respiratory symptoms among children (n = 121) with bronchial hyperreactivity and elevated total
March 2008 3-35 DRAFT-DO NOT QUOTE OR CITE
-------
Asthma Symptoms
0.75 0.85 0.95 1.05
Odds Ratio
Rescue Inhaler Uses
1.15
0.75
0.85
0.95 1.05
Rate Ratio
1.15
Nitrogen dioxide
1 ~,n A A O7
1 an O 1
3-day moving sum 1 .01
Nitrogen dioxide and PM10
i an n 0 QQ —
| on 1 0 97
I an 9 1
Ldg Z
A
1-°6 1 r
1.04
A 1 1 n
1.09
aT A 11^
1.04
1.06
• 1 13
1.04
^ 1 11
1.08
r>2 * 11*
1.04
• •f n?
1.25
Nitrogen dioxide
LagO 1.00
Lag 1 0.99-
Lag2 1.0
3-day moving sum 1 .0'
Nitrogen dioxide and PM10
i ft A n 07
l_ElQ U \J.\jf i™"""™"™
1 an 1 0 Q7
Lag i u.a/ —
Lag 2 LOO
3-day moving sum 1 .00 -
1.04
1.04
1.05
•,
1.03
-•— 1
1.03
1.03
»,
1.04
1.02
— • — 1
I Bl
-1.08
-1.08
1 OQ
05
1 08
1.08
1 09
05
1.25
Figure 3.1-5. Odds ratios (95% confidence interval [CI]) for daily asthma symptoms
(panel A) and rate ratios (95% CI) for daily rescue inhaler use (panel B)
associated with shifts in within-subject concentrations of NOi for single- and
joint (with PMi0)-pollutant models from the Childhood Asthma Management
Program (November 1993-September 1995). The city-specific estimates from
Boston, Baltimore, Toronto, St. Louis, Denver, Albuquerque, and San Diego
were included in the calculations of study-wide effects.
Source: Schildcrout et al. (2006).
March 2008
3-36
DRAFT-DO NOT QUOTE OR CITE
-------
1 IgE in urban and rural areas of the Netherlands. These effects were seen for all lags examined
2 (lag 0-, 1-, 2-, and 5-day mean), with the strongest association for the 5-day mean (OR = 1.75
3 [95% CI: 1.37, 2.22]) for each 20 ppb increase). Significant associations between lower
4 respiratory symptoms and ambient exposures were seen in single-pollutant models with PMio,
5 black smoke, and SO2. No multipollutant models were reported.
6 For adults, most studies examining associations between ambient NO2 pollution and
7 respiratory symptoms have been conducted in Europe. Various studies have enrolled older
8 adults, (van der Zee et al., 2000; Harre et al., 1997; Silkoff et al., 2005), nonsmoking adults
9 (Segala et al., 2004), patients with COPD (Higgins et al., 1995; Desqueyroux et al., 2002), and
10 individuals with bronchial hyperresponsiveness (Boezen et al., 1998) or asthma (Hiltermann
11 et al., 1998; Forsberg et al., 1998; Von Klot et al., 2002). Associations were found between NO2
12 and either respiratory symptoms or inhaler use in a number of studies (van der Zee et al., 2000;
13 Harre et al., 1997; Silkoff et al., 2005; Segala et al., 2004; Hiltermann et al., 1998), but not in all
14 studies (Desqueyroux et al., 2002; Von Klot et al., 2002).
15 Among the studies discussed above, odds ratios and 95% CI for associations with asthma
16 symptoms in children are presented in Figure 3.1-6. The figure shows the several lag periods
17 presented in each study. In the figure, the area of the square denoting the odds ratio represents
18 the relative weight of that estimate based on the width of the 95% CI. When combined in a
19 random effect meta-analysis1, the combined OR for asthma symptoms from a meta-analysis was
20 1.14 (95% CI: 1.05, 1.24) and the test for heterogeneity had a p value of 0.055. The results of
21 multipollutant analyses for the three U.S. multicity studies are presented in Figure 3.1-7.
22 Associations with NO2 were generally robust to adjustment for copollutants, as stated previously.
23 Odds ratios were often unchanged with the addition of copollutants, though reductions in
24 magnitude are apparent in certain models, such as with adjustment for SO2 in the Six Cities study
25 results (Schwartz et al., 1994).
26
The effects used in the meta-analysis were selected using the following methodology. One lag period per study
was selected, with studies having 0 lag preferred to 1-day lags and moving averages; longer single-day lags were not
included in the meta-analysis. If a study had both incidence and prevalence, then the incidence effect was to be
used.
March 2008 3-37 DRAFT-DO NOT QUOTE OR CITE
-------
Study
Mortimer et al. (2002)*
Schildcrout et al. (2006)*
Schildcrout et al. (2006)
Schildcrout et al. (2006)
Delfinoetal.{2002)*
Just et al. (2002)
Just et al. (2002)
Just et al. (2002)*
Just etal. (2002)
Just etal. (2002)
Just etal. (2002)
Segala etal. (1998)'
Segala etal. (1998)
.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Odds ratio for asthma symptoms in std units
Figure 3.1-6. Odds ratios (95% CI) for associations between asthma symptoms and 24-h
average NOi concentrations (per 20 ppb). The size of the box of the central
estimate represents the relative weight of that estimate based on the width of
the 95% CI.
Asthma
D 1
Prevalence
Prevalence
Prevalence
Prevalence
Incidence
Incidence
Incidence
Lag
1 R MA
0
1
0-2
0-2
ft A
0_4
0
1
I
3
-B-
— i—
-i —
! I 1 I I I 1 I
1 3.1.4.3 Summary of Evidence on the Effects of Short-Term NOi Exposure on
2 Respiratory Symptoms
3 Consistent evidence has been observed for an association of respiratory effects with
4 indoor and personal NC>2 exposures in children at levels similar to ambient concentrations. In
5 particular, the Pilotto et al. (2004) intervention study provided evidence of improvement in
6 respiratory symptoms with reduced NC>2 exposure in asthmatic children.
7 The epidemiologic studies using community ambient monitors also find associations
8 between ambient NC>2 concentration and respiratory symptoms. The results of new U.S.
9 multicity studies (Schildcrout et al., 2006; Mortimer et al., 2002) provide further support for
March 2008
3-38
DRAFT-DO NOT QUOTE OR CITE
-------
Study Locations
Schwartz etal. (1994) 6 cities, US
Schildcrout et al. (2006) 8 North American cities 24-h
24-h
Mortimer et al. (2002) 8 cities, US
4-h
Odds Ratio
0.6 1.0 1.4 1.8 2.2 2.6
Pollutants
N0
N02 + 03
N02 + S02
N02
N02 + CO
N02 + S02
N02
N02 + CO
N02 + S02
N0
N02 + 03 + S02
N02+ 03 + S02 + PM
3.0 3.4
Cough Incidence
0 • 6-11 years
0 o 5-1 9 years
0 • 4-9 years
0 D 9-17 years
Asthma Symptoms
Rescue Inhaler Use
»
Morning Asthma Symptoms
Figure 3.1-7. Odds ratios and 95% confidence intervals for associations between asthma
symptoms and 24-h average NOi concentrations (per 20 ppb) from
multipollutant models.
1
2
3
4
5
6
1
8
9
10
11
12
associations with respiratory symptoms and medication use in asthmatic children. Associations
were observed in cities where the median range was 18 to 26 ppb for a 24-h avg (Schildcrout
et al., 2006) and the mean NO2 level was 32 ppb for a 4-h avg (Mortimer et al., 2002).
Multipollutant models in these multicity studies were generally robust to adjustment for
copollutants including 63, CO, and PMi0. Most human clinical studies did not report or observe
respiratory symptoms with NO2 exposure, and animal toxicologic studies do not measure effects
that would be considered symptoms. The experimental evidence on airways inflammation and
immune system effects discussed previously, however, provides some plausibility and coherence
for the observed respiratory symptoms in epidemiologic studies.
3.1.5 Effects of Short-Term NOi Exposure on Lung Function
March 2008
3-39
DRAFT-DO NOT QUOTE OR CITE
-------
1 3.1.5.1 Epidemiologic Studies of Lung Function
2
3 Spirometry in Children
4 Reliable measurement of lung function in children presents special challenges. The
5 method that produces the most accurate results is spirometry, which requires special equipment
6 and trained examiners. Of the short-term exposure studies reviewed here that did use spirometry
7 (Hoek and Brunekreef, 1994; Linn et al., 1996; Timonen et al., 2002; Moshammer et al., 2006),
8 all conducted repeated lung function measurements in schoolchildren. All found significant
9 associations between small decrements in lung function and increases in ambient NC>2 levels.
10 Hoek and Brunekreef (1994) enrolled 1,079 children in the Netherlands to examine the effects of
11 low-level winter air pollution on FVC, FEVi, maximal midexpiratory flow (MMEF), and PEF.
12 A significant effect was found only for the PEF measure: the mean (over all subjects) slope (SE)
13 was a reduction of 52 mL/s (95% CI: 21, 83) for a 20-ppb increase in the previous day's NC>2.
14 The authors do not present mean values for lung function measurements, so it is not possible to
15 calculate what percentage of PEF this decrement represents. Linn et al. (1996) examined 269
16 Los Angeles-area schoolchildren and short-term air pollution exposures. The authors found
17 statistically significant associations between previous-day 24-h avg NO2 concentrations and FVC
18 the next morning (mean decline of 8 mL [95% CI: 2, 14] per 20-ppb increase in NC^) and
19 current-day 24-h avg NC>2 concentrations and morning to evening changes in FEVi (mean
20 decline of 8 mL [95% CI: 2, 14] per 20-ppb increase in NO2). Timonen et al. (2002) enrolled 33
21 Finnish children with chronic respiratory symptoms to study the effects of exercise-induced lung
22 function changes and ambient air pollution. No significant effects were observed for lung
23 function changes due to exercise, but significant associations were observed for level of NC>2
24 lagged by 2 days and baseline FVC (mean decline of 21 mL [95% CI: -29, -12] for 20-ppb
25 NO2) and FEVi (mean decline of 20 mL [95% CI: -26, -13] for 20-ppb NO2). An Austrian
26 study enrolled 163 healthy children for repeated lung function testing (11 to 12 tests during the
27 school year) (Moshammer et al., 2006). A central site monitor adjacent to the school were used
28 to calculate 8-h avg (midnight, to 8 a.m.) PM and NO2 concentrations. The median 8-h avg NO2
29 concentration was 17.5 |ig/m3 (9.2 ppb). In both single pollutant and multipollutant models
30 including PM2.5, the authors found each 20-ppb increase in NC>2 level produced reductions in
31 lung function of around 4% for FEVi, FVC, forced expiratory volume in 0.5 s (FEVo.s), maximal
March 2008 3-40 DRAFT-DO NOT QUOTE OR CITE
-------
1 expiratory flow at 50% (MEF50), and maximal expiratory flow at 25% (MEF25). PM2.5 was not
2 significantly associated with lung function decrements in the multipollutant model.
3
4 Peak Flow Meter Measurements in Children
5 Studies involving supervised lung function measurements in schoolchildren using peak
6 flow devices do not show a consistent relationship between NO2 exposure and measurements of
7 peak flow (Scarlett et al., 1996; Peacock et al., 2003; Steerenberg et al., 2001) (Annex Table
8 AX6.3-2). Other studies using home-use peak flow meters with children did not report any
9 significant associations with ambient NO2 (Roemer et al., 1998 [2,010 children in the Pollution
10 Effects on Asthmatic Children in Europe (PEACE) study]; Roemer et al., 1999 [a subset of 1,621
11 children from the PEACE study with chronic respiratory symptoms]; Mortimer et al., 2002
12 [846 asthmatic children from the NCICAS]; Van der Zee et al., 1999 [633 children in the
13 Netherlands]; Timonen and Pekkanen, 1997 [169 children including asthmatics in Finland];
14 Ranzi et al., 2004 [118 children, some with asthma, in the Italian Asma Infantile Ricerca (AIRE)
15 study]; Segala et al., 1998 and Just et al., 2002 [over 80 asthmatic children in Paris]; Delfmo
16 et al., 2003a [22 asthmatic children in southern California]).
17 Ward et al. (2000) examined the effect of correcting peak flow for nonlinear errors on
18 NO2 effect estimates in a panel study of 147 children (9-year olds, 47% female). The correction
19 resulted in a small increase in the group mean PEF (1.1 L-min"1). For the entire panel, NO2
20 effect estimates were all corrected in the positive direction with a narrowing of the 95% CI, and
21 all but the result for 0-day lag were decreased in absolute size by up to 73% (e.g., effect estimate
22 for NO2 lagged 3 days corrected from -0.56 to -0.15% per 10 ppb). When only the
23 symptomatic/atopic children (i.e., reported wheezing and positive skin test) were considered, the
24 estimates for associations with 5-d avg NO2 decreased in size from -5.0 to -1.8% per 20 ppb. In
25 the case of lag 0, the effect estimate became significant with an increase in magnitude from -1.1
26 to -2.3% per 20 ppb. The authors concluded that correction for PEF meter measurements
27 resulted in small but important shifts in the direction and size of effect estimates and probable
28 interpretation of results. The effects of correction were, however, not consistent across
29 pollutants or lags and could not be easily predicted.
30
March 2008 3-41 DRAFT-DO NOT QUOTE OR CITE
-------
1 Lung Function in Adults
2 Spirometry was used in a large cross-sectional study in Switzerland (Schindler et al.,
3 2001). A subset of 3,912 lifetime nonsmoking adults participated in the spirometric lung
4 function measurements in the SAPALDIA study (Study of Air Pollution and Lung Diseases in
5 Adults). Significant inverse relationships were found between increases in NO2 and decreases in
6 FVC (by 2.74% [95% CI: 0.83, 4.62]) and FEVi (by 2.52% [95% CI: 0.49, 4.55]) for a 20-ppb
7 increase in NO2 on the same day as the examination. Forced expiratory flow at 25 to 75% of
8 FVC (FEF25.75) was found to decrease by 6.73% (95% CI: 0.038, 13.31) for each 20-ppb
9 increase in average NO2 concentration over the previous 4 days. One study (Lagorio et al.,
10 2006) of COPD patients found significant inverse relationships for FEVi in both COPD and
11 asthmatic patients. Another study of COPD subjects (Silkoff et al., 2005) observed no adverse
12 effects of ambient air pollution on lung function for the first winter; however, in the second
13 winter, a significant decrease in morning PEF associated with same day and previous day NO2
14 level was seen (quantitative results not provided). In a study of 60 asthmatic adults in London,
15 decreases in two lung function measures, FEVi and FEF2s-75, and increased FENo were reported
16 with increased NO2 exposure while walking along a roadway with heavy traffic; associations
17 were also reported with PM2.5, ultrafme particles, and EC (McCreanor et al., 2007).
18 Of the adult studies reviewed that employed portable peak flow meters for
19 subject-measured lung function, none reported significant associations with NO2 levels (van der
20 Zee et al., 2000 [489 adults in the Netherlands]; Higgins et al., 1995 [153 adults in the United
21 Kingdom, including COPD and asthma patients]; Park et al., 2005a [64 asthmatic adults in
22 Korea]; Hiltermann et al., 1998 [60 asthmatic adults in the Netherlands]; Harre et al., 1997 [40
23 adults with COPD in New Zealand]; Forsberg et al., 1998 [38 adult asthmatics in Sweden];
24 Higgins et al., 2000 [35 adults with COPD or asthma in the United Kingdom]).
25
26 3.1.5.2 Clinical Studies of Lung Function
27
28 Healthy Adults
29 Studies examining responses of healthy volunteers to acute exposure to NO2 have
30 generally failed to show alterations in lung mechanics such as airways resistance (Hackney et al.,
31 1978; Kerr et al., 1979; Linn et al., 1985a; Mohsenin, 1987a, 1988; Frampton et al., 1991; Kim
32 et al., 1991; Morrow et al., 1992; Rasmussen et al., 1992; Vagaggini et al., 1996; Azadniv et al.,
March 2008 3-42 DRAFT-DO NOT QUOTE OR CITE
-------
1 1998; Devlin et al., 1999). Exposures ranging from 75 min to 5 h at concentrations of up to
2 4.0-ppm NO2 did not alter pulmonary function. Bylin et al. (1985) found increased airways
3 resistance after a 20-min exposure to 0.25-ppm NO2 and decreased airways resistance after a
4 20-min exposure to 0.5-ppm NO2, but no change in airways responsiveness to aerosolized
5 histamine challenge in the same subjects. These effects have not been confirmed in other
6 laboratories.
7 Few human clinical studies of NC>2 have included elderly subjects. Morrow et al. (1992)
8 studied the responses of 20 healthy volunteers (13 smokers, 7 nonsmokers) of mean age
9 61 years, following exposure to 0.3-ppm NC>2 for 4 h with light exercise. There was no
10 significant change in lung function related to NC>2 exposure for the group as a whole. However,
11 the 13 smokers experienced a slight decrease in FEVi during exposure, and their responses were
12 significantly different from the 7 nonsmokers (percent change in FEVi at end of exposure:
13 -2.25 versus + 1.25%, p = 0.01). The post-hoc analysis and small numbers of subjects,
14 especially in the nonsmoking group, limits the interpretation of these findings.
15 The controlled human exposure studies reviewed in the Os AQCD (U.S. Environmental
16 Protection Agency, 2006) generally reported only small pulmonary function changes after
17 combined exposures of NC>2 or nitric acid (HNOs) with 63, regardless of whether the interactive
18 effects were potentiating or additive. Hazucha et al. (1994) found that preexposure of healthy
19 women to 0.6-ppm NC>2 for 2 h enhanced spirometric responses and methacholine airways
20 responsiveness induced by a subsequent 2-h exposure to 0.3-ppm Os, with intermittent exercise.
21 Following a 1-h exposure with heavy exercise, Adams et al. (1987) found no differences between
22 spirometric responses to 0.3-ppm Os and the combination of 0.6-ppm NC>2 + 0.3-ppm Os.
23 However, the increase in airways resistance was significantly less for adults exposed to 0.6-ppm
24 NC>2 + 0.3-ppm 63 compared to 0.3-ppm 63 alone.
25 Gong et al. (2005) studied 6 healthy elderly subjects (mean age 68 years) and 18 patients
26 with COPD (mean age 71 years), all exposed to: (a) air, (b) 0.4-ppm NC>2, (c) -200 |ig/m3
27 concentrated ambient fine particles (CAPs), and (d) CAPs + NC>2. Exposures were for 2-h with
28 exercise for 15 min of each half hour. CAPs exposure was associated with small reductions in
29 midexpiratory flow rates on spirometry, and reductions in oxygen saturation, but there were no
30 effects of NC>2 on lung function, oxygen saturation, or sputum inflammatory cells. However, the
31 exposures were not fully randomized or blinded, and most of the NC>2 exposures took place
March 2008 3-43 DRAFT-DO NOT QUOTE OR CITE
-------
1 months after completion of the CAPs and air exposures. In addition, the small number of healthy
2 subjects severely limits the statistical power for this group.
3
4 Patients with COPD
5 Few studies have examined responses to NO2 in subjects with COPD. Hackney et al.
6 (1978) found no lung function effects of exposure to 0.3-ppm NO2 for 4-h with intermittent
7 exercise in smokers with symptoms and reduced FEVi. In a group of 22 subjects with moderate
8 COPD, Linn et al. (1985b) found no pulmonary effects of 1-h exposures to 0.5-, 1.0-, or 2.0-ppm
9 NO2 with 30 min of exercise.
10 In a study by Morrow et al. (1992), 20 subjects with COPD were exposed for 4-h to
11 0.3-ppm NO2 in an environmental chamber, with intermittent exercise. Progressive decrements
12 in FVC occurred during the exposure, becoming statistically significant only at the end of the
13 exposure. The decrements in FVC occurred without changes in flow rates. These changes in
14 lung function were typical of the "restrictive" pattern seen with NO2 rather than the obstructive
15 changes described by some studies of NO2 exposure in asthmatics.
16 Gong et al. (2005) exposed 6 elderly healthy adults and 10 COPD patients to four
17 separate atmospheres: (a) air, (b) 0.4-ppm NO2, (c) ~200-|ig/m3 CAPs, or (d) CAPs + NO2. As
18 noted above, there were no significant effects of NO2 in either the healthy or the COPD subjects.
19
20 Patients with Asthma
21 Kleinman et al. (1983) evaluated the response of lightly exercising asthmatic subjects to
22 inhalation of 0.2-ppm NO2 for 2 h, during which resting minute ventilation doubled. Forced
23 expiratory flows and airways resistance were not altered by the NO2 exposure. Bauer et al.
24 (1986) studied the effects of mouthpiece exposure to 0.3-ppm NO2 for 30 min (20 min at rest
25 followed by 10 min of exercise at -40 L/min) in 15 asthmatics. At this level, NO2 inhalation
26 produced significant decrements in forced expiratory flow rates after exercise, but not at rest.
27 Torres and Magnussen (1991) found no effects on lung function in 11 patients with mild asthma
28 exposed to 0.25-ppm NO2 for 30-min, including 10-min of exercise. However, small reductions
29 in FEVi were observed following 1-ppm NO2 exposure for 3-h with intermittent exercise in
30 12 mild asthmatics. Koenig et al. (1994) found no pulmonary function effects of exposure to
31 0.3-ppm NO2 in combination with 0.12-ppm Os, with or without sulfuric acid (H2SO4)
March 2008 3-44 DRAFT-DO NOT QUOTE OR CITE
-------
1 (70 |ig/m3) or HNO3 (0.05 ppm), in 22 adolescents with mild asthma. However, 6 additional
2 subjects dropped out of the study citing uncomfortable respiratory symptoms.
3 Jenkins et al. (1999) examined FEVi decrements and airways responsiveness to allergen
4 in a group of mild, atopic asthmatics. The subjects were exposed during rest for 6 h to filtered
5 air, 200-ppb NO2, 100-ppb O3, or 200-ppb NO2 + 100-ppb O3. The subjects were also exposed
6 for 3 h to 400-ppb NO2, 200-ppb O3, or 400-ppb NO2 + 200-ppb O3 to provide doses identical to
7 those in the 6-h protocols (i.e., equal C x T). Immediately following the 3-h exposure, but not
8 after the 6-h exposure, there were significant decrements in FEVi following O3 and NO2 + O3
9 exposures.
10
11 3.1.5.3 Summary of Evidence of the Effect of Short-Term NOi Exposure on Lung
12 Function
13 In summary, epidemiologic studies using data from supervised lung function
14 measurements (spirometry or peak flow meters) report small decrements in lung function (Hoek
15 and Brunekreef, 1994; Linn et al., 1996; Moshammer et al., 2006; Schindler et al., 2001; Peacock
16 et al., 2003). No significant associations were reported in any studies using unsupervised, self-
17 administered peak flow measurements with portable devices. Correcting peak flow
18 measurements for nonlinear errors resulted in small but important shifts in the direction and size
19 of effect estimates; however, these effects were not consistent across pollutants or lags.
20 Overall, clinical studies have not provided compelling evidence of NO2 effects on
21 pulmonary function. Acute exposures of young, healthy volunteers to NO2 at levels of as high as
22 4.0 ppm do not alter lung function as measured by spirometry or airways resistance. The small
23 number of studies of COPD patients prevents any conclusions about effects on pulmonary
24 function. The Morrow et al. (1992) study, performed in Rochester, NY, suggested restrictive
25 type effects of 0.3-ppm NO2 exposure for 4 h. However, three other studies, performed in
26 southern California at similar exposure concentrations, found no effects. The contrasting
27 findings in these studies may, in part, reflect the difference in duration of exposure or the
28 differing levels of background ambient air pollution to which the subjects were exposed
29 chronically, as there were much lower background levels in Rochester, NY than in southern
30 California. For asthmatics, the effects of NO2 on pulmonary function have also been inconsistent
31 at exposure concentrations of less than 1-ppm NO2. Overall, clinical studies have failed to show
32 effects of NO2 on pulmonary function at exposure concentrations relevant to ambient exposures.
March 2008 3-45 DRAFT-DO NOT QUOTE OR CITE
-------
1 However, the range of findings in COPD and asthmatic patients may reflect that some
2 individuals within such groups may be particularly more susceptible to NC>2 effects than others.
3
4 3.1.6 Hospital Admissions and ED Visits for Respiratory Outcomes
5 Total respiratory causes for ED visits and hospitalizations typically include asthma,
6 bronchitis and emphysema (collectively referred to as COPD), pneumonia, upper and lower
7 respiratory infections, and other minor categories. Temporal associations between ED visits or
8 hospital admissions for respiratory diseases and the ambient concentrations of NC>2 have been the
9 subject of more than 50 well-conducted research publications since 1993. These studies form a
10 new body of literature that was unavailable in 1993, when the previous criteria document was
11 published. In addition to considerable statistical and analytical refinements, the more recent
12 studies have examined responses of morbidity in different age groups and multipollutant models
13 to evaluate potential confounding effects of copollutants.
14
15 3.1.6.1 All Respiratory Outcomes (ICD9 460-519)
16 Overall, the majority of studies that have examined all respiratory outcomes as a single
17 group have focused on hospital admission data. The results from the hospitalization and ED visit
18 studies, for all ages and stratified by age group are presented in Figures 3.1-8 and 3.1-9. More
19 details are provided in Annex Tables AX6.3-1, AX6.3-2, and AX6.3-3. Collectively, studies of
20 hospitalizations and ED visits provide suggestive evidence of an association between ambient
21 NO2 levels and ED visits and hospitalizations for all respiratory causes when participants of all
22 ages are considered in the analyses. Stronger and more consistent associations were observed
23 among children and older adults (65+ years) compared to adults (<65 years), with an
24 interquartile range (IQR) of 1 to 13% excess risk estimated per 20 ppb incremental change in
25 24-h avg NO2 or 30 ppb incremental change in 1-h max NO2.
26 Peel et al. (2005) examined ED visits for all respiratory causes among all ages in relation
27 to ambient NO2 concentrations in Atlanta, GA during the period of 1993 to 2000. They found a
28 2.4% (95% CI: 0.9, 4.1) increase in respiratory ED visits associated with a 30-ppb increase in
29 1-h max NO2 concentrations. Tolbert et al. (2007) recently reanalyzed these data with
30 4 additional years of data and found similar results (2.0% increase, 95% CI: 0.5, 3.3).
31 Two multicity studies combined the effects of ambient air pollution (including NO2) in
32 several cities and describe similar response rates and respiratory health outcomes as measured by
March 2008 3-46 DRAFT-DO NOT QUOTE OR CITE
-------
Location
Atlanta, GA
Atlanta, GA
Windsor, ON
Windsor, ON
West Midlands, UK
London, UK
London, UK
London, UK
Torrelavega, Spain
Dram men, Norway
Drammen, Norway
Reggio Emilia, Italy
Perth, Australia
Brisbane, Australia
Vancouver, BC
Windsor, ON
Windsor, ON
West Midlands, UK
London, UK
London, UK
London, UK
Pisa, Italy
Brisbane, Australia
Brisbane, Australia
Multeity-Australia
Multicity-Australia
Multi city-Australia
Hong Kong, China
Sao Paulo, Brazil
Sao Paulo, Brazil
Lag Other
0-2
0-2
0-3 Female
0-3 Male
0-1
1
NR
2
NR
3
0-3
4
1
1
1
0-3 Girls
0-3 Boys
0-1
2
1
2
0-2
3 0-4 yrs
0 5-14 yrs
0-1 0 yrs
0-1 1-4 yrs
0-1 5-14 yrs
0-3
0-4
0
In | All ages
f
1
1
— i-
—
•
— i —
Children
-1-
f
fr
I
— 1
_+_
.+.
I
l I I I I I
75 1 1.25 1.5 1.75 2 2,25
Relative risk
Figure 3.1-8. Relative Risks (95% CI) for hospital admissions or ED visits for all
respiratory disease stratified by all ages or children. Results from studies
using 24-h average standardized to a 20-ppb increase, results from studies
using 1-h max standardized to a 30-ppb increase (* indicates ED visits, all
others are hospital admissions; A indicates 1-h max averaging times, all
others are 24-h mean averaging times).
March 2008
3-47
DRAFT-DO NOT QUOTE OR CITE
-------
Reference
Luginaahetal,,(2005)A
Luginaahetal.,(2005)A
Spixetal.(1998)
Anderson etal. (2001 )"
Atkinson etal. (1999a)A
Atkinson et al. (1 999b)*A
Ponce de Leon etal. (1996)
Schouten etal. (1996)
Schouten etal. (1996)
Petroeschevsky et al. (2001)
Wong etal. (1999)
Luginaah et al. (2005}*
Luginaah et al. (2005)*
Fung etal. (2006)
Yang etal. (2003)
Spix etal. (1998)
Anderson etal. (2001 )A
Atkinson etal. (1999a)A
Atkinson etal.(1999b)*A
Ponce de Leon etal. (1996)
Andersen et al. (2007b)
Andersen et al. (2007a)
Schouten etal. (1996)
Schouten etal. (1996)
Simpson et al. (2005a}A
Hinwood et al. (2006)
Petroeschevsky etal. (2001)
Wong etal. (1999)
Location
Windsor, ON
Windsor, ON
Multicity, Europe
West Midlands, UK
London, UK
London, UK
London, UK
Amsterdam, Netherlands
Rotterdam, Netherlands
Brisbane, Australia
Hong Kong, China
Windsor, ON
Windsor, ON
Vancouver, BC
Vancouver, BC
Multicity, Europe
West Midlands, UK
London, UK
London, UK
London, UK
Copenhagen, Denmark
Copenhagen, Denmark
Amsterdam, Netherlands
Rotterdam, Netherlands
Multicity-Australia
Perth, Australia
Brisbane, Australia
Hong Kong, China
Lag Other
0-3 Female
0-3 Male
1-3
0-1
1
2
0
1
1
0
0-3
0-3 Female
0-3 Male
0-3
1
1-3
0-1
3
0
2
0-4
0-4
2
0
0-1
1
5
0-3
•
,
— i —
^^^
[Adults]
i-
i-
+—
h
[65+1
,
.
i—
i —
•1-
^~
i-
i
]
,
i i i
.75 1 1.25 1.5
Relative risk
Figure 3.1-9. Relative Risks (95% CI) for hospital admissions or ED visits for all
respiratory disease stratified by adults and older adults (^65 years). Results
from studies using 24-h average standardized to a 20-ppb increase, results
from studies using 1-h max standardized to a 30-ppb increase (* indicates ED
visits, all others are hospital admissions; A indicates 1-h max averaging times,
all others are 24-h mean averaging times).
March 2008
3-48
DRAFT-DO NOT QUOTE OR CITE
-------
1 increased hospital admissions (Barnett et al., 2005; Simpson et al., 2005a). Barnett et al. (2005)
2 used a case-crossover method to study ambient air pollution effects on respiratory hospital
3 admissions of children (age groups 0, 1 to 4, and 5 to 14 years) in multiple cities in both
4 Australia and New Zealand during the study period 1998 to 2001. No significant associations
5 were observed between NO2 and increased hospital admissions for infants. For all respiratory
6 admissions among children 1 to 4 years, a 9.6% (95% CI: 2.3, 17.3) increase was found for a
7 30-ppb increase in the daily 1-h max concentration of NO2, and for children aged 5 to 14 years
8 the same increase in NO2 resulted in a 16.5% increase in admission for all respiratory disease
9 (95% CI: 5.4, 28.8) both lagged 0 to 1 day (Barnett et al., 2005).
10 In a multicity study of all hospitalizations for respiratory disease for adults ages >65
11 years, Simpson et al. (2005a) examined the response to a change in the daily 1-h max level of
12 NO2. The standardized percent increase was 8.4% (95% CI: 4.6%, 12.4%; lag 0 to 1 day per
13 30-ppb increase). The authors presented results from three statistical models that produced
14 similar results overall for the four cities.
15 Two Canadian studies compared multiple statistical methods for data analysis in studies
16 of hospitalizations for all respiratory outcomes. In Vancouver, Fung et al. (2006) used time-
17 series analysis, the method of Dewanji and Moolgavkar (2000), and case-crossover analyses to
18 examine the association of ambient NO2 concentrations with all respiratory hospitalizations for
19 adults aged 65 years and older. All three methods showed similar results, with positive
20 associations between incremental changes in NO2 of 5.43 ppb (IQR) from a mean concentration
21 of 16.83 ppb. Using a time-series analysis, Fung et al. (2006) reported a percent increase
22 (standardized to 20 ppb) of 6.8% ([95% CI: 1.1%, 13.1%] lag 0), while the case-crossover
23 analysis showed a significant change in the percent increase of 10.7% ([95% CI: 3.7%, 15.5%]
24 lag 0). The Dewanji and Moolgavkar (2000) model did not produce a statistically significant
25 association between NO2 and hospitalization for an increase of 20 ppb, though the central
26 estimate remained positive (percent increase = 4.5% [95% CI: -1.1%, 10.3%] lag 0)]. In the
27 second of these two studies, Luginaah et al. (2005) used two approaches that included both time-
28 series and case-crossover analyses segregated by sex. They noted a positive trend between an
29 incremental change in 24-h avg NO2 of 20 ppb and respiratory admissions. Though associations
30 for females in each of the age groups examined were positive, the authors found only one
31 statistically significant association in females aged 0 to 14 years that identified an increased
March 2008 3-49 DRAFT-DO NOT QUOTE OR CITE
-------
1 percent of hospitalization of 24.1% using the case-crossover analysis (24.1% [95% CI: 0.3%,
2 53.8%] lag 2). The results of the time-series analyses from the Fung et al. (2006) and Luginaah
3 et al. (2005) studies are presented in Figures 3.1-8 and 3.1-9, respectively.
4 European studies on associations with respiratory hospitalizations were conducted in
5 London, Paris, and in Drammen, Norway (Ponce de Leon et al., 1996; Dab et al., 1996; Oftedal
6 et al., 2003). Ponce de Leon et al. (1996) found significant positive relative risks for all ages and
7 for children (0 to 14 year olds), but not for adults (15 to 64 years). Dab et al. (1996) determined
8 that there was no statistically significant association between admissions for all respiratory
9 causes combined based on an incremental change of 52.35 ppb, though the estimates were
10 positive. Oftedal et al. (2003) reported that the relative rate of hospitalizations for all respiratory
11 disease increased based on an increment of 20 ppb NO2 (RR = 1.111 [95% CI: 1.031, 1.19.9] lag
12 3 days). Other studies also found positive outcomes (Andersen et al., 2007a,b; Atkinson et al.,
13 1999a,b; Bedeschi et al., 2007; Burnett et al., 2001; Farchi et al., 2006; Hinwood et al., 2006; Lin
14 et al., 1999; Llorca et al., 2005; Pantazopoulou et al., 1995; Vigotti et al., 2007; Wong et al.,
15 1999; Yang et al., 2003). Several studies presented null results (Anderson et al., 2001; Gouveia
16 and Fletcher, 2000; Hagen et al., 2001; Schouten et al., 1996). Finally, a number of studies were
17 considered that could not inform the association of NO2 concentration on all respiratory disease
18 hospital admissions or ED visits. These studies are included in Annex Tables AX6.3-1, AX6.3-
19 2, and AX6.3-3 (Atkinson et al., 2001; Buchdahl et al., 1996; Burnett et al., 1997a; Chen et al.,
20 2005; Fung et al., 2007; Linares et al., 2006; Pantazopoulou et al., 1995; Prescott et al., 1998;
21 Villeneuve et al., 2006).
22 To assess potential confounding by copollutants, results from multipollutant models were
23 evaluated. As noted in Annex 3B, multipollutant models may have limited utility to distinguish
24 the independent effects of specific pollutants if model assumptions are not met. Despite this
25 limitation, these models are widely used in air pollution research. Figures 3.1-10 and 3.1-11
26 present NO2 risk estimates for all respiratory causes with and without adjustment for various
27 particulate and gaseous copollutants, respectively, in two-pollutant models. Collectively,
28 copollutant regression analyses indicated that NO2 risk estimates for respiratory ED visits and
29 hospitalizations, in general, were not sensitive to the inclusion of additional gaseous or
30 particulate pollutants.
March 2008 3-50 DRAFT-DO NOT QUOTE OR CITE
-------
Reference Location Age Lag
Wong et al. (1999) Hong Kong All 0-3
Hagen et al. (2000) Drammen, Norway All 0-3
Oftedaletal. (2003) Drammen, Norway All 3
Yang etal, (2007) Taipei, Taiwan All 0-2
Yang et al. (2007) Taipei, Taiwan All 0-2
Burnett etal.(1997b)* Toronto, ON All 0
Gouveia and Fletcher Sao Paulo, Brazil <5 0
(2000)*
Andersen et al. (2007b) Copenhagen, 65+ 0-4
Denmark
Andersen et al. (2007a) Copenhagen,
Denmark
65+ 0-4
Other Pollutants
NO,
N02+PM10 .
N02
N02+PM10 .
N02
N02+PM10 .
>25 C N02
N02+PM10 .
<25 C N02
N02+PM10 .
N02
N02+PM1(J .
N02+PM25 .
N02+PM1M5 .
N02
N02+PM10 .
N02
N
»-
°
—
— • — ° Single pollutant model
• Copollutant model
Simpson el al. (2005a)* Mullicity- Australia 65+ 0-1
\^ I \^
.9 1.1 1.3 1.5 1.7 1.9 2.1
Relative risk
Figure 3.1-10. Relative Risks (95% CI) for hospital admissions or emergency department
visits for all respiratory causes, standardized from two-pollutant models
adjusted for particle concentration. (* indicates 1-h peak avg times, all
others are 24-h avg; effect estimates from studies using 1-h peak
measurements are standardized to a 30-ppb increase; effect estimates from
studies using 24-h average measurements are standardized to a 20-ppb
increase).
March 2008
3-51
DRAFT-DO NOT QUOTE OR CITE
-------
Reference Location Age Lag
Wong etal. (1999) Hong Kong All 0-3
Other
Pollutant!
NO,
N02+03
Yang et al. (2007) Taipei, Taiwan All 0-2 >25C N02
N02+03
Yang et al. (2007) Taipei, Taiwan All 0-2 <25C N02
Yang et al, (2003) Vancouver, BC <3 1
Gouveia and Fletcher Sao Paulo, Brazil <5 0
(2000)
Simpson et al. (2005a)* Multicity - Australia 65+ 0-1
N(V°3
N02
N02+03
N02
N02+03
NO,
N02+03
Yang et al. (2003) Taipei, Taiwan 65+ 1 >25 C N02
Yang et al. (2007) Taipei, Taiwan
0-2 <25 C
Yang et al. (2007) Taipei, Taiwan All 0-2
Burnett etal.(1997b)' Toronto, ON All 0
N02+03
N02
N02+S02
N02
N02+S02
NO,
N02+S02
Yang et al. (2007) Taipei. Taiwan All 0-2 >25C NO,
N02+C0
Yang etal. (2007) Taipei, Taiwan All 0-2 <25C N02
Oftedal et al. (2003) Drammen, Norway All 3
N02+C0
N02
NO, + Benzene
> Single pollutant model
1 Copollutant model
I \ \ I
.98 1.18 1.38 1.58
Relative risk
Figure 3.1-11. Relative Risks (95% CI) for hospital admissions or emergency department
visits for all respiratory causes, standardized from two-pollutant models
adjusted for gaseous pollutant concentration. (* indicates 1-h peak
averaging times, all others are 24-h average; effect estimates from studies
using 1-h peak measurements are standardized to a 30-ppb increase; effect
estimates from studies using 24-h average measurements are standardized
to a 20-ppb increase).
March 2008
3-52
DRAFT-DO NOT QUOTE OR CITE
-------
1 3.1.6.2 Asthma (ICD9 493)
2 Studies of ED visits and hospitalizations provide suggestive evidence of an association
3 between ambient NO2 levels and ED visits and hospitalizations for asthma among children and
4 adults. Figures 3.1-12 and 3.1-13 show the relative risks (and 95% confidence limits) of
5 hospitalizations and visits to the ED for asthma associated with daily NO2 concentrations, for
6 allages and stratified by age. Larger effect estimates were generally observed for children
7 compared to adults and older adults (65+ years), with an IQR of 1 to 25% excess risk estimated
8 per 20 ppb incremental change in 24-h avg NC>2 or 30 ppb incremental change in 1-h max NC>2.
9 The few studies that examined the association of asthma and NC>2 levels among older adults (65+
10 years) generally reported positive central estimates, though none of these was statistically
11 significant. When subjects of all ages were examined, the results of ED visits and
12 hospitalizations were overwhelmingly positive, especially when the 24-h averaging time was
13 used. The epidemiologic studies of ED visits and hospital admissions for asthma are
14 summarized in Annex Tables AX6.3-1, AX6.3-2, and AX6.3-3.
15 In Atlanta, GA, Peel et al. (2005) examined various respiratory ED visits in relation to
16 pollutant levels from 1993 to 2000. Results for the a priori single-pollutant models examining a
17 3-day moving average (lag 0, 1, and 2) of NC>2 showed a small positive, but not statistically
18 significant, association with asthma visits (percent increase = 2.1% [95% CI: -0.4%, 4.5%) for
19 all age groups. In a secondary analysis of patients ages 2 to 18 years, a 30-ppb increase in the
20 day 5 lag of the NO2 concentration yielded a percent increase of 4.1% (95% CI: 0.8%, 7.6%).
21 In New York City, NY, Ito et al. (2007) examined numbers of ED visits for asthma in
22 relation to pollution levels from 1999 to 2002. NC>2 was generally the most significant (and
23 largest in effect size per the same distributional increment) predictor of asthma ED visits among
24 PM2.5, O3, SO2, and CO (percent increase = 12% (95% CI: 7%, 15%) per 20 ppb increase).
25 Further, MV s risk estimates were most robust to the addition of other pollutants in the model,
26 and the addition of NO2 reduced other pollutant's risk estimates most consistently.
27 Jaffe et al. (2003) examined the effects of ambient pollutants during the summer months
28 (June through August) on the daily number of ED visits for asthma among Medicaid recipients
29 aged 5 to 34 years from 1991 to 1996 in Cincinnati and Cleveland. The percent change in ED
30 visits for asthma as the primary diagnosis per 20-ppb increase in 24-h avg NC>2 concentration
March 2008 3-53 DRAFT-DO NOT QUOTE OR CITE
-------
Reference Location
Peel et al. (2005)*A Atlanta, GA
lto(2007)* New York, NY
Burnett et al. (1999) Toronto, ON
Anderson et al. (1998) London, UK
Atkinson et al. (1999a)A London, UK
Atkinson et al. (1 999b)*A London, UK
Galan et al. (2003)* Madrid, Spain
Chardon et al. (2007)* Paris, France
Schouten et al. (1 996) Amsterdam, Netherlands
Migliaretti et al. (2005)* Turin, Italy
Migliaretti and Cavallo (2004) Turin, Italy
Hinwood et al. (2006) Perth, Australia
Petroeschevsky et al. (2001)" Brisbane, Australia
Wong etal. (1999) Hong Kong, China
Tsai et al. (2006) Kaohsiung, Taiwan
Tsai et al. (2006) Kaohsiung, Taiwan
Yang et al. (2007) Taipei, Taiwan
Yang et al. (2007) Taipei, Taiwan
Peeletal.(2005)*A Atlanta, GA
Tolbert et al. (2000)*A Atlanta, GA
Lin et al. (2003) Toronto, ON
Lin et al. (2003) Toronto, ON
Sunyer et al. (1 997)* Multicity-Europe
Anderson et al. (1 998) London , UK
Atkinson et al. (1999a)A London, UK
Atkinson et al. (1999b)*A London, UK
Thompson et al. (2001)* Belfast, Ireland
Andersen et al. (2007b) Copenhagen, Denmark
Andersen et al. (2007a) Copenhagen, Denmark
Migliaretti et al. (2005)* Turin, Italy
Migliaretti and Cavallo (2004) Turin, Italy
Migliaretti and Cavallo (2004) Turin, Italy
Barnett et al. (2005) Multicity-Australia
Barnett et al. (2005) Multicity-Australia
Hinwood etal. (2006) Perth, Australia
Petroeschevsky et al. (2001)" Brisbane, Australia
Petroeschevsky et al. (2001)" Brisbane, Australia
Morgan et al. (1998a) Sydney, Australia
Ko et al. ( 2007a) Hong Kong, China
Lee et al. (2006) Hong Kong, China
Gouveia and Fletcher (2000)A Sao Paulo, Brazil
Lag Other
0-2
0-1
0
0-3
0
0
3
0-3
2
0-3
1-3
0
0-2
0-3
0-2 Warm
0-2 Cool
0-2 >25 C
0-2 <25 C
2-18
1
0-5 Boys
0-5 Girls
0-3
0-3
1
1
0-3
0-5
0-5
0-3
1-3 4-15 yrs
1-3 <4yrs
0-1 1-4 yrs
0-1 4-15 yrs
0
0-2 0-4 yrs
1 5-14 yrs
0
0-4
3
2
([AJFages
*
3
b*~ | Children
f
I
^—
— -
•1-
1 1 1 1 1 1 1 1
.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75
Relative risk
Figure 3.1-12.
Relative Risks (95% CI) for hospital admissions or emergency department
visits for asthma stratified by all ages or children. Results from studies
using 24-h average standardized to a 20-ppb increase, results from studies
using 1-h max standardized to a 30-ppb increase (* indicates ED visits, all
others are hospital admissions; A indicates 1-h max averaging times, all
others are 24-h mean averaging times).
March 2008
3-54
DRAFT-DO NOT QUOTE OR CITE
-------
Reference
Jaffeetal. (2003)*
Jaffeetal. (2003)*
Linn etal. (2000)
Sunyer etal. (1997)*
Anderson etal. (1998)
Atkinson etal. (1999a)A
Atkinson etal.(1999b)*A
Boutin-Forzano et al. (2004)*
Tenias etal. (1998)*
Castellsague etal. (1995)*
Migliaretti etal. (2005)*
Morgan etal. (1998a)
Ko etal. (20073)
Anderson etal. (1998)
Atkinson et al. (1999a)A
Migliaretti etal. (2005)*
Hinwood et al. (2006)
Koetal.(2007a)
Location
Cleveland, OH
Cleveland, OH
Los Angeles, LA
Multicity-Europe
London, UK
London, UK
London, UK
Marseille, France
Valencia, Spain
Barcelona, Spain
Turin, Italy
Sydney, Australia
Hong Kong, China
London, UK
London, UK
Turin, Italy
Perth, Australia
Hong Kong, China
Lag Other
1
1
0-1
0-3
0-1
1
1
0
0
0-2
0-3
0
0-4
0-3
3
0-3
0
0-4
T~ I
75 1
• Adults
!
f
I
+
HI-
•h-
-fl-
— 1 |65+
-1—
+
i i n
1.25 1.5 1.75
Relative risk
Figure 3.1-13. Relative Risks (95% CI) for hospital admissions or emergency department
(ED) visits for asthma stratified by adults and older adults (^65 years).
Results from studies using 24-h average standardized to a 20-ppb increase,
results from studies using 1-h max standardized to a 30-ppb increase (*
indicates ED visits, all others are hospital admissions; A indicates 1-h max
averaging times, all others are 24-h mean averaging times).
4
5
was 12% (95% CI: -2, 28) in Cincinnati and 8% (95% CI: -2, 16.6) in Cleveland, with an
overall percent increase in ED visits of 6% (95% CI: -2, 14).
Barnett et al. (2005) examined specific respiratory disease outcomes and did not find
associations between incremental changes in NO2 concentration and respiratory admissions for
asthma among children 1 to 4 years old. The largest association found in this study was a 25.7%
March 2008
3-55
DRAFT-DO NOT QUOTE OR CITE
-------
1 increase in asthma admissions in the 5- to 14-year age group related to a 20-ppb increase in 24-h
2 NO2, with evidence of a seasonal impact that resulted in larger increases in admissions during the
3 warm season. When the same groups were examined for the effect of a 30-ppb change in the 1-h
4 max concentration of NO2, there were no significant associations between NC>2 and
5 hospitalizations for asthma.
6 Lin et al. (2004) studied gaseous air pollutants and 3,822 asthma hospitalizations (2,368
7 boys, and 1,454 girls) among children 6 to 12 years of age with low household income in
8 Vancouver, Canada, between 1987 and 1998. NO2 levels were derived from 30 monitoring
9 stations, and daily levels were found to be significantly and positively associated with asthma
10 hospitalizations for males in the low socioeconomic group but not in the high socioeconomic
11 group. This effect did not persist among females. Lin et al. (2003) conducted a case-crossover
12 analysis of the effect of short-term exposure to gaseous pollution on 7,319 asthma
13 hospitalizations (4,629 boys, 2,690 girls), in children in Toronto between 1980 and 1994. NO2
14 concentrations measured from four monitoring stations were positively associated with asthma
15 admissions in both sexes. Differences in the results of these two studies might be attributed to
16 differences in the study designs or differences in subject population sizes.
17 A time-series analysis in Sydney examined respiratory outcomes in children and adults,
18 but reported no association between changes in NC>2 (24-h avg) for asthma admissions (Morgan
19 et al., 1998a). For children aged 1 to 14, a 10.9% increase in hospital admissions for asthma
20 ([95% CI: 2.2, 20.3] lag 0) was associated with the daily 1-h maximum value based on 30-ppb
21 incremental change. The association with adults was positive, but not statistically significant.
22 Studies of ED visits and hospitalizations for asthma have been reported in London, U.K.
23 (Atkinson et al., 1999a,b; Hajat et al., 1999); Belfast, Ireland (Thompson et al., 2001); Valencia,
24 Barcelona, and Madrid, Spain (Tenias et al., 1998; Galan et al., 2003; Castellsague et al., 1995);
25 Turin, Italy (Migliaretti and Cavallo, 2004; Migliaretti et al., 2005); Marseille and Paris, France
26 (Boutin-Forzano et al., 2004; Dab et al., 1996); Amsterdam and Rotterdam, the Netherlands
27 (Schouten et al., 1996), and Melbourne, Brisbane and Perth, Australia (Erbas et al., 2005;
28 Hinwood et al., 2006). Sunyer et al. (1997) have described a meta-analysis of several cities
29 under the umbrella of the Air Pollution on Health: a European Approach (APHEA) protocol
30 (Katsouyanni et al., 1996). Additional studies report a positive association between NC>2
31 concentration and hospital admissions or ED visits (Andersen et al., 2007a; Anderson et al.,
March 2008 3-56 DRAFT-DO NOT QUOTE OR CITE
-------
1 1998; Arbex et al., 2007; Burnett et al., 1999; Kim et al., 2007; Ko et al., 2007; Lee et al., 2006;
2 Linn et al., 2000; Tsai et al 2006; Wong et al., 2001; Yang et al., 2007). Several studies have
3 reported null or negative associations (Andersen et al., 2007b; Anderson et al., 1998; Chardon
4 et al., 2007; Gouveia and Fletcher 2000; Petroeschevsky et al., 2001; Spix et al., 1998; Tanaka
5 et al., 1998; Tolbert et al., 2000).
6 Copollutant and multipollutant regression analyses were performed in several of these
7 studies. Results generally indicated that NC>2 risk estimates for respiratory ED visits and
8 hospitalizations were not sensitive to the inclusion of additional gaseous or particulate pollutants.
9 Finally, there were a number of studies that were considered but did not inform the
10 association of NO2 concentration on all respiratory disease hospital admissions or ED visits.
11 These studies are included in Annex Tables AX6.3-1, AX6.3-2, and AX6.3-3 (Atkinson et al.,
12 2001; Bates et al., 1990; Chew et al., 1999; Garty et al., 1998; Kesten et al., 1995; Lipsett et al.,
13 1997; Magas et al., 2007; Neidell, 2004; Ponka, 1991; Ponka and Vitanen 1996; Rossi et al.,
14 1993; Stieb et al., 1996; Sun et al., 2006; Tobias et al., 1999).
15
16 3.1.6.3 COPD (ICD9 490-496)
17 Relatively few studies have examined the association of ED visits and hospitalizations for
18 COPD and ambient NO2 levels. The epidemiologic studies of ED visits and hospital admissions
19 for COPD are summarized in Annex Tables AX6.3-1, AX6.3-2, and AX6.3-3. Studies
20 examining COPD outcomes have focused on hospital admission data, including multicity studies
21 in the United States (Moolgavkar, 2000, 2003), Europe (Anderson et al., 1997) and Australia
22 (Simpson et al., 2005a), and single-city studies in the United States (Peel et al., 2005), Canada
23 (Yang et al., 2005), Europe (Anderson et al., 2001; Atkinson et al., 1999a; Dab et al., 1996;
24 Tenias et al., 2002), Australia (Morgan et al., 1998a; Hinwood et al., 2006), and Asia (Lee et al.,
25 2007; Yang and Chen, 2007).
26 In a time-series study in Vancouver, an area with low pollution concentrations (24-h
27 mean NO2 of 17.03 ppb), Yang et al. (2005) reported associations between NO2 and hospital
28 admissions for COPD in patients >65 years for both the lag 1 day (RR =1.19; 95% CI: 1.04,
29 1.37) and 7-day extended lag period (RR = 1.46 [95% CI: 1.15, 1.94]). Additional studies found
30 weaker, though statistically significant positive associations with ambient levels of NO2 and
31 COPD (Moolgavkar, 2003; Anderson et al., 1997; Simpson et al., 2005a). A time-series analysis
March 2008 3-57 DRAFT-DO NOT QUOTE OR CITE
-------
1 in Sydney, Australia, examined respiratory outcomes in children and adults but did not show an
2 association between changes in NO2 (24-h average) for increased hospital admissions among
3 COPD patients >65 years (Morgan et al., 1998a). Similarly, a study in Paris, France, of COPD
4 and related obstructive respiratory disease found that NO2 was not statistically significantly
5 associated with increased hospital admissions (Dab et al., 1996).
6
7 3.1.6.4 Respiratory Diseases Other than Asthma or COPD
8 ED visits or hospital admissions for respiratory diseases include upper respiratory
9 infections (URIs), pneumonia, bronchitis, allergic rhinitis, and lower respiratory disease (LRD).
10 The reviewed epidemiologic studies of ED visits and hospital admissions for these respiratory
11 diseases are summarized in Annex Tables AX6.3-1, AX6.3-2, and AX6.3-3. Though some of
12 these studies reported positive and statistically significant results (Atkinson et al., 1999a; Burnett
13 et al., 1997b, 1999; Farchi et al., 2006; Gouveia and Fletcher, 2000; Hwang and Chan, 2002;
14 Ilabaca et al., 1999; Lin et al., 2005; Peel et al., 2005; Simpson et al., 2005a), others reported null
15 or negative associations (Barnett et al., 2005; Chardon et al., 2007; Hinwood et al., 2006; Karr
16 et al., 2006; Lin et al., 1999; Ponka and Virtanen, 1994; Zanobetti and Schwartz, 2006). Finally,
17 there are two studies that were considered but could not inform the association of NC>2
18 concentration on all respiratory disease hospital admissions or ED visits (Bates et al., 1990;
19 Linares et al., 2006). These studies are included in Annex Tables AX6.3-1, AX6.3-2, and
20 AX6.3-3.
21
22 3.1.6.5 Summary of the Evidence on the Effect of Short-Term Exposure to NOi on
23 Respiratory ED Visits and Hospitalizations
24 In summary, many studies have observed positive associations between ambient NC>2
25 concentrations and ED visits and hospitalizations for all respiratory diseases and asthma. These
26 associations are particularly consistent among children and older adults (65+ years) for hospital
27 admissions for all respiratory diseases. For asthma hospitalization, the effect estimates were
28 largest when children and subjects of all ages were included in the analysis. Results from
29 copollutant models suggested that the effect of NC>2 on ED visits and hospitalizations for all
30 respiratory causes and asthma were generally robust and independent of the effects of ambient
31 particles or gaseous copollutants. In preceding sections, exposure to NC>2 has been found to
32 result in host defense and immune system changes, airways inflammation, and airways
March 2008 3-58 DRAFT-DO NOT QUOTE OR CITE
-------
1 responsiveness. While not providing specific mechanistic data linking exposure to ambient NC>2
2 and respiratory hospitalization or ED visits, these findings provide plausibility and coherence for
3 such a relationship.
4 However, the limited evidence does not support a relationship between ED visits and
5 hospitalizations for COPD and ambient NC>2 levels, and there were limited studies providing
6 inconsistent results for many of the health outcomes other than asthma, making it difficult to
7 draw conclusions about the effects of NC>2 on these diseases.
8
9 3.1.7 Summary and Integration—Respiratory Health Effects with
10 Short-Term NOi Exposure
11 Taken together, the findings of epidemiologic, human clinical, and animal toxicological
12 studies provide evidence that is sufficient to infer a likely causal relationship for respiratory
13 effects with short-term NC>2 exposure. The body of evidence from epidemiologic studies has
14 grown substantially since the 1993 AQCD and provides scientific evidence that short-term
15 exposure to NC>2 is associated with a broad range of respiratory morbidity effects, including
16 altered lung host defense, inflammation, airways hyperresponsiveness, respiratory symptoms,
17 lung function decrements, and ED visits and hospital admissions for respiratory diseases. New
18 evidence comes from large longitudinal studies, panel studies, and time-series studies. NC>2
19 exposure is associated with aggravation of asthma effects that include symptoms, medication
20 use, and lung function. Effects of NO2 on asthma were most evident with cumulative lag of 2 to
21 6 days, rather than same-day levels of NC>2. Time-series studies also demonstrated a relationship
22 in children between hospital admissions or ED visits for asthma and NC>2 exposure. In many of
23 these studies, there were high correlations between ambient measures of NC>2 and CO and PM;
24 however, the effect estimates for NC>2 were robust after the inclusion of CO and PM in
25 multipollutant models. Recent epidemiologic studies provide somewhat inconsistent evidence
26 on short-term exposure to NO2 and inflammatory responses in the airways, as well as for
27 associations with lung function decrements. The epidemiologic evidence for these effects can be
28 characterized as consistent, in that associations are reported in studies conducted in numerous
29 locations with a variety of methodological approaches. While the individual risk estimates are
30 small in magnitude, and thus not considered strong individually, the body of epidemiologic
31 evidence has strength in that fairly precise and robust risk estimates have been reported from
32 multicity studies.
March 2008 3-59 DRAFT-DO NOT QUOTE OR CITE
-------
1 Important evidence also is available from epidemiologic studies of indoor NO2
2 exposures. A number of recent studies show associations with wheeze, chest tightness, and
3 length of symptoms (Belanger et al., 2006); respiratory symptom rates (Nitschke et al., 2006);
4 school absences (Pilotto et al., 1997a); respiratory symptoms, likelihood of chest tightness, and
5 asthma attacks (Smith et al., 2000); and severity of virus-induced asthma (Chauhan et al., 2003).
6 A particular intervention study (Pilotto et al., 2004) provides strong evidence of a detrimental
7 effect of exposure to NC>2. Considering this large body of epidemiologic studies alone, the
8 findings are coherent in the sense that the studies report associations with respiratory health
9 outcomes that are logically linked together.
10 Experimental evidence offers some coherence and plausibility for the observed
11 epidemiologic associations. Toxicologic studies have also shown that lung host defenses,
12 including mucociliary clearance and AM and other immune cell functions, are sensitive to NC>2
13 exposure, with effects observed at concentrations of less than 1 ppm (see Annex Table AX4.3
14 and AX4.5). The limited evidence from human studies indicates that NC>2 may increase
15 susceptibility to injury by subsequent viral challenge. Devlin et al. (1999) found reduced AM
16 phagocytic capacity after NC>2 exposure, which suggest a reduced ability to clear inhaled bacteria
17 or other infectious agents. Frampton et al. (2002) found enhanced epithelial cell injury in
18 response to RS V infection after NC>2 exposure. Taken together with the epidemiologic evidence
19 described above linking NC>2 exposure with viral illnesses, there is coherent and consistent
20 evidence that NC>2 exposure can result in lung host defense or immune system effects. This
21 group of outcomes provides some plausibility for other respiratory system effects as well. For
22 example, effects on ciliary action (clearance) or on macrophage function (i.e. phagocytosis,
23 cytokine production) can lead to the type of outcomes assessed in epidemiologic studies, such as
24 respiratory illness or symptoms.
25 Controlled human exposure studies provide evidence for airways hyperresponsiveness
26 i.e., a heightened bronchoconstrictive response to a challenge agent, following short-term
27 exposure to NC>2. In acute exacerbations of asthma, bronchial smooth muscle contraction
28 (bronchoconstriction) occurs quickly to narrow the airways in response to exposure to various
29 stimuli including allergens or irritants. Bronchoconstriction is the dominant physiological event
30 leading to clinical symptoms and interference with airflow (National Heart, Lung, and Blood
31 Institute, 2007). Recent studies involving allergen challenge in asthmatics suggest that NC>2 may
March 2008 3-60 DRAFT-DO NOT QUOTE OR CITE
-------
1 enhance the sensitivity to allergen-induced decrements in lung function and affect allergen-
2 induced inflammatory responses following exposures as low as 0.26 ppm NC>2 for 30 min during
3 rest. Nonspecific responsiveness is also increased following 30-min exposures of resting
4 asthmatic subjects to 0.2- to 0.3-ppm NO2 and following 1-h exposures to 0.1-ppm NC>2.
5 The few recent epidemiologic studies have reported associations between ambient NCh
6 exposure and airways inflammation. These studies are suggestive of effects in children, but offer
7 more limited evidence for effects in adults. Controlled human exposure studies provide
8 consistent evidence for airways inflammation at NC>2 concentrations of <2.0 ppm; the onset of
9 inflammatory responses in healthy subjects appears to be between 100 and 200 ppm-min, i.e.,
10 1 ppm for 2 to 3 h. Biological markers of inflammation are reported in antioxidant-deficient
11 laboratory animals with exposures to 0.4-ppm NO2, though healthy animals do not respond until
12 exposed to much higher levels, i.e., 5-ppm NC>2. The biochemical effects observed in the
13 respiratory tract following exposure to NC>2 include chemical alteration of lipids, amino acids,
14 proteins, enzymes, and changes in oxidant/antioxidant homeostasis, with membrane
15 polyunsaturated fatty acids and thiol groups as the main biochemical targets for NC>2 exposure.
16 However, the biological implications of such alterations are unclear. Potential mechanisms for
17 effects on the respiratory system include membrane damage from increases in reactive oxygen
18 species, lipid and protein pertubations, and recruitment of inflammatory cells from epithelial cell
19 injury by reactive oxygen species.
20 In evaluating the potential relationships between short-term exposure to NO2 and
21 respiratory effects, it is important to note the interrelationships between NC>2 and other
22 pollutants, and the potential for NC>2 to serve as a marker for a pollutant mixture, particularly
23 traffic-related pollution. As outlined in the preface to this draft Integrated Science Assessment
24 (ISA), this includes consideration of potential pathways, such as the direct causal pathway for
25 effects, mediation of effects, the pollutant acting as a surrogate for a pollutant mixture, or
26 confounding between pollutants. As observed above, associations with NC>2 were often robust to
27 adjustment for traffic-related pollutants (e.g., PM and CO), even in locations where the
28 correlations between pollutants were substantial. The epidemiologic evidence has thus been
29 found to be consistent and coherent for respiratory symptoms and respiratory hospitalization and
30 ED visits. In addition, toxicologic and clinical studies report effects of exposure to gaseous NC>2,
31 as discussed previously, for outcomes related to lung host defense and immune system changes.
March 2008 3-61 DRAFT-DO NOT QUOTE OR CITE
-------
1 The experimental studies indicate that NO2 is solely responsible for the effects reported. The
2 findings of direct effects of NO2 in toxicologic or human clinical studies, in combination with
3 robust associations reported in epidemiologic studies, support a conclusion that NO2 is
4 independently responsible for some respiratory effects. There is little available evidence to
5 evaluate the potential for NO2 effects to be mediated by other pollutants or exposures; further,
6 clinical and epidemiologic study findings do not appear to suggest that coexposure with another
7 pollutant is required to observe NO2-related effects.
8 The evidence summarized here supports the conclusion that there is a likely causal
9 relationship between short-term exposure to NO2 and effects on the respiratory system.
10 However, the challenge remains in considering the potential for NO2 to serve as a surrogate for a
11 mixture of combustion-related pollutants. Most studies examined show that personal NO2
12 exposures are significantly correlated either with ambient or personal level PM2 5, or other
13 combustion-generated products (e.g., CO and EC). As discussed in Chapter 2, ambient NO2
14 measurements can provide a valid estimate of personal exposure to ambient NO2 as used in most
15 epidemiology studies. Although the evidence indicates that NO2 exposure is independently
16 associated with some respiratory health effects, there remains the possibility that NO2 also serves
17 as a marker for combustion-related emissions, particularly from traffic, for some health
18 outcomes.
19
20
21 3.2 CARDIOVASCULAR EFFECTS ASSOCIATED WITH
22 SHORT-TERM NO2 EXPOSURE
23 The current review includes approximately 40 studies published since 1993
24 characterizing the effect of short-term NOx exposure on hospitalizations or ED visits for CVD.
25 These studies form a new body of literature that was unavailable in 1993, when the previous
26 AQCD was published.
27
28 3.2.1 Heart Rate Variability, Repolarization Changes, Arrhythmia, and
29 Markers of Cardiovascular Function in Humans and Animals
30
31 3.2.1.1 Heart Rate Variability
32 Heart rate variability (HRV), a measure of the beat-to-beat change in heart rate, is a
33 reflection of the overall autonomic control of the heart. It is hypothesized that increased air
March 2008 3-62 DRAFT-DO NOT QUOTE OR CITE
-------
1 pollution levels may stimulate the autonomic nervous system and lead to an imbalance of cardiac
2 autonomic control characterized by sympathetic activation unopposed by parasympathetic
3 control (Liao et al., 2004; Brook et al., 2004). Such an imbalance of cardiac autonomic control
4 may predispose susceptible people to greater risk of ventricular arrhythmias and consequent
5 cardiac deaths (Liao et al., 2004; Brook et al., 2004). HRV has been studied most frequently in
6 coronary artery disease populations, particularly in the post-myocardial infarction (MI)
7 population. Lower time domain as well as frequency domain variables (i.e., measures of reduced
8 HRV) are associated with an increase in cardiac and all-cause mortality among this susceptible
9 population. Those variables most closely correlated with parasympathetic tone appear to have
10 the strongest predictive value in heart disease populations. Specifically, acute changes in RR-
11 variability may temporally precede and are predictive of increased long-term risk for the
12 occurrence of ischemic sudden death and/or precipitating ventricular arrhythmias in individuals
13 with established heart disease (for example, see La Rovere et al., 2003). Findings from studies
14 of ambient NO2 and HRV were mixed with some studies reporting an adverse effect (reduction
15 in variability) (Liao et al., 2004; Chan et al., 2005; Wheeler et al., 2006), while other studies
16 reported no significant change (Luttman-Gibson et al., 2006; Holguin et al., 2003; Schwartz et al.
17 2005). In some studies reporting reductions in HRV, reductions for PM were similar to those
18 observed for NO2 (Liao et al., 2004; Wheeler et al. 2006). See Annex AX6.3-10 for a detailed
19 discussion of HRV studies.
20
21 3.2.1.2 Arrhythmias Recorded on Implanted Defibrillators
22 Results from studies directly measuring ventricular arrhythmias were inconsistent and
23 potentially confounded by PM (Peters et al., 2000; Dockery et al., 2005; Rich et al., 2005, 2006a;
24 Metzger et al,. 2007). Among the ambient air pollutants, the strongest association with
25 arrhythmias was observed for PM, which was highly correlated to NO2 concentrations in these
26 studies (Dockery et al., 2005; Rich et al., 2005; Metzger et al., 2007). Rich et al. (2006b) did not
27 observe an association between NO2 level and paroxysmal atrial fibrillation (PAF). See Annex
28 AX6.3-11 for detailed discussion of defibrillator studies.
29
30 3.2.1.3 Repolarization Changes
31 In addition to the role played by the autonomic nervous system in arrhythmogenic
32 conditions, myocardial vulnerability and repolarization abnormalities are believed to be key
March 2008 3-63 DRAFT-DO NOT QUOTE OR CITE
-------
1 factors contributing to the mechanism of such diseases. Measures of repolarization include QT
2 duration, T-wave complexity, variability of T-wave complexity, and T-wave amplitude.
3 Henneberger et al. (2005) reported that NO2 and NO were not associated with repolarization
4 abnormalities.
5
6 3.2.1.4 Markers of Cardiovascular Disease Risk
7 Several investigators have explored potential mechanisms by which air pollution could
8 cause CVD. In particular, markers of inflammation, cell adhesion, coagulation, and thrombosis
9 have been evaluated in epidemiologic studies. Pekkanen et al. (2000) reported a significant
10 increase in fibrinogen associated with short-term NO2 exposure while Steinvil et al. (2007)
11 reported significant decreases in fibrinogen associated with NO2. Schwartz (2001) reported
12 increases in fibrinogen and platelet count associated with NO2 level in single-pollutant models,
13 which changed direction in multipollutant models also containing PMi0. Liao et al. (2005) did
14 not observe differences in white blood cell (WBC) count, Factor VIII-C, fibrinogen, von
15 Willibrand Factor (VWF), or albumin associated with 24-h avg NO2 levels. However, PMio was
16 associated with factor VIII-C in the cohort examined. Ruckerl et al. (2006) observed a
17 significant association of NO2 (lagged 2-6 days) with C-reactive protein (CRP) greater than the
18 90th percentile but the strongest effect on CRP was observed for ultrafme particles. Baccarelli
19 et al. (2007) reported a shorter prothrombin time (PT) with increasing NO2 levels but, a similar
20 decrease in PT was observed for PMi0.
21 Collectively, associations reported for NO2 and markers of cardiovascular risk in
22 epidemiologic studies appear to be potentially confounded by PM and other traffic-related
23 pollutants. Several authors suggest that these biomarker studies provide evidence for biologic
24 plausibility of the effect of PM on cardiovascular health rather than NO2 (Schwartz 2001; Seaton
25 and Dennekamp, 2003).
26 A limited number of controlled human exposure studies suggest effects of NO2 exposure
27 on cardiac output, blood pressure, and circulating red blood cells at concentrations of less than
28 2.0 ppm (Drechsler-Parks, 1995; Linn et al., 1985a; Posin et al., 1978; Frampton et al., 2002)
29 require confirmation. Drechsler-Parks (1995) observed a lower mean stroke volume for NO2 +
30 O3 than for air and speculated that chemical interactions between O3 and NO2 at the level of the
31 epithelial lining fluid led to the production of nitrite, leading to vasodilatation, with reduced
32 cardiac preload and cardiac output. Linn et al. (1985a) reported small but statistically significant
March 2008 3-64 DRAFT-DO NOT QUOTE OR CITE
-------
1 reductions in blood pressure after exposure to 4-ppm NC>2 for 75 min, a finding consistent with
2 systemic vasodilatation in response to the exposure; this finding has not been repeated.
3 Frampton et al. (2002) reported a concentration-related reduction in hematocrit and hemoglobin
4 in both males and females, among health subjects exposed to NO2, confirming the findings of an
5 earlier study conducted by Posin et al. (1978). See Annex AX6 for a detailed discussion of these
6 studies.
7 The results on the effect of NC>2 on various hematological parameters in animals are
8 inconsistent and, thus, provide little biological plausibility for the epidemiology findings. There
9 have also been reported changes in the red blood cell membranes of experimental animals
10 following NO2 exposure. Red blood cell D-2,3-diphosphoglycerate was reportedly increased in
11 guinea pigs following exposure to 0.36-ppm NC>2 for 1 week (Mersch et al., 1973). An increase
12 in red blood cell sialic acid, indicative of a younger population of red blood cells, was reported in
13 rats exposed to 4.0-ppm NC>2 continuously for 1 to 10 days (Kunimoto et al., 1984). However, in
14 another study, exposure to the same concentration of NC>2 resulted in a decrease in red blood cell
15 number (Mochitate and Miura, 1984). A more recent study (Takano et al., 2004) using an obese
16 rat strain found changes in blood triglycerides, high-density lipoprotein cholesterol (HDL), and
17 HDL/total cholesterol ratios with a 24-week exposure to 0.16-ppm NC>2. In the only study
18 conducted with an exposure of less than 5-ppm NC>2 that evaluated methemoglobin formation,
19 Nakajima and Kusumoto (1968) reported that, in mice exposed to 0.8-ppm NC>2 for 5 days, the
20 amount of methemoglobin was not increased. This is in contrast to some (but not all) in vitro
21 and high-concentration NC>2 in vivo studies, which have found methemoglobin effects
22 (U.S. Environmental Protection Agency, 1993).
23
24 3.2.1.5 Toxicology of Inhaled Nitric Oxide
25 Nitric oxide is used in humans therapeutically as a pulmonary vasodilator, and has shown
26 little evidence for adverse respiratory effects. The literature on therapeutic uses of nitric oxide
27 provides the strongest evidence for its lack of toxicity. Infants and adults with acute respiratory
28 failure and refractory hypoxemia, as well as pulmonary hypertension, are sometimes considered
29 candidates for inhaled NO. Inhaled NO acts as a selective pulmonary vasodilator, causing
30 vascular smooth muscle relaxation and increased perfusion in ventilated lung regions. Beneficial
31 effects in patients with respiratory failure include reduced pulmonary artery pressures and
March 2008 3-65 DRAFT-DO NOT QUOTE OR CITE
-------
1 improved ventilation-perfusion matching. Nitric oxide is used clinically at concentrations
2 ranging from five ppm to as high as 80 ppm. There has been little or no toxicity reported, even
3 when used in premature infants with respiratory failure. In a recently published multicenter
4 study (Kinsella et al., 2006), 793 premature infants with respiratory failure were randomized to
5 therapy with inhaled NO or air. NO therapy was associated with a reduced risk of brain injury,
6 and in a reduced risk of bronchopulmonary dysplasia, a chronic lung condition resulting from
7 lung injury in infancy, in infants weighing at least 1000 gm. NO can cause methemoglobinemia,
8 and this was seen transiently in only 2 infants. NO can inhibit activation of blood leukocytes and
9 platelets (Gianetti et al. 2002); however there was no evidence for increased susceptibility to
10 infection or bleeding. One of the concerns about NO therapy is the potential for NO to be
11 oxidized to NO2, so administration systems are designed to avoid this.
12
13 3.2.2 Studies of Hospital Admissions and ED Visits for CVD
14 Cases of CVD are typically identified using ICD codes, which are recorded on hospital
15 discharge records in these studies. However, counts of hospital or ED admissions are used in
16 some studies. Studies of ED visits may include cases that are less severe than those included in
17 hospital admission studies. Hospital admission studies are distinguished from ED visit studies in
18 the annex tables (Annex AX6.3-6 through AX6.3-9). Many studies group all CVD diagnoses
19 (ICD9 codes 390-459), evaluating cardiac diseases (ICD9 codes 390-429), and cerebrovascular
20 disease (ICD9 430-448) together. Other studies evaluate cardiac and cerebralvascular diseases
21 separately or further distinguish ischemic heart disease (IHD: ICD9 410-414), myocardial
22 infarction (MI: ICD9 410), congestive heart failure (CHF: ICD9 428), cardiac arrhythmia
23 (ICD9 427), angina pectoris (ICD9 413), or stroke (ICD9 430-438).
24 Numerous studies have shown a positive association between both 24-h avg and 1-h max
25 NO2 levels and hospital admissions or ED visits for all CVD, in single-pollutant models (Linn
26 et al., 2000; Metzger et al., 2004; Tolbert et al., 2007; Ballester et al., 2001, 2006; Anderson
27 et al., 2007a; Atkinson et al., 1999a,b; Poloniecki et al., 1997; Barnett et al., 2006; Hinwood et
28 al., 2006; Jalaludin et al., 2006; Chang et al., 2005; Wong et al., 1999; Yang et al., 2004b). A
29 discussion of results from studies reporting associations between NO2 and all CVD are found in
30 Annex AX6.2.1.
31
March 2008 3-66 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.2.2.1 Cardiac Disease (ICD9 390-429)
2 Findings from studies examining the association of NO2 with cardiac disease are found in
3 Figure 3.2-1. Most investigators who distinguished cardiac disease from all CVD report
4 significant positive associations in single-pollutant models. Increased risks were observed in
5 Canadian populations (Burnett et al., 1997b; Fung et al., 2005). The average daily 1-h max NO2
6 level was approximately 39 ppb in metropolitan Toronto, ON, where these studies were
7 conducted. Estimates from two Australian multicity studies (Barnett et al., 2006; Simpson et al.,
8 2005a) were also significantly increased. The 24-h NO2 level in the Australian cities studied by
9 Barnett et al. (2006) was 7 to 11.5 ppb. The range of 1-h max NO2 level in cities studied by
10 Simpson et al. (2005a) was 16 to 24 ppb. Von Klot et al. (2005) observed a statistically
11 significant association between readmission for cardiac disease among MI survivors, a
12 potentially susceptible subpopulation and NO2 concentrations in five European cities. The range
13 in 24-h NO2 level was 15.8 to 26 ppb in the five cities studied. Two single-city Australian
14 studies and one single-city Taiwanese study also reported positive single-pollutant model results
15 (Jalaludin et al., 2006; Morgan et al., 1998a; Chang et al., 2005). Studies of the association of
16 24-h avg and 1-h max NO2 level with IHD, MI, CHF and arrhythmia are less consistently
17 positive and significant. Results from these studies are described in Annex AX6.2-1.
18 Most investigators reporting results from multipollutant models observed diminished
19 effect estimates for NO2 and hospital admissions or ED visits for CVDs. In two U.S. studies
20 conducted in Los Angeles, investigators indicated that their analyses were unable to distinguish
21 the effects of NO2 from PM, CO, and other traffic pollutants (Linn et al., 2000; Mann et al.,
22 2002). In both studies, CO was more highly correlated with NO2 than PM. In an Atlanta study,
23 Metzger et al. (2004) and Tolbert et al. (2007) also observed a diminished effect of NO2 on visits
24 for CVD when CO was modeled with NO2, while the effect of CO remained robust. Tolbert
25 et al. (2007) discussed the limitations of multipollutant models and concluded that these models
26 might help researchers identify the strongest predictor of disease, but might not isolate the
27 independent effect of each pollutant. NO2 was not robust to adjustment for other pollutants in
28 several non-U.S. studies (Jalaludin et al., 2006; Ballester et al., 2006; Simpson et al., 2005a;
29 Poloniecki et al., 1997; Barnett et al., 2006; Llorca et al., 2005). However, in other studies,
30 investigators reported that the effect of NO2 was robust in multipollutant models (Von Klot et al.,
March 2008 3-67 DRAFT-DO NOT QUOTE OR CITE
-------
Reference
Fung etal. (2005)*
Fung etal, (2005)*
Fung et al. (2005)*
von Klot et al. (2005)
Ballesteretal. (2001)*
Bailester et al. (2006)
Barnett et al. (2006)
Barnett et al. (2006)
Simpson et al. (2005a)*
Simpson et al. (2005a)*
Simpson et al. (2005a)*
Jaludin et al. (2006)*
Jaludin et al. (2006)*
Jaludin et al. (2006)*
Morgan etal. (1998a)*
Morgan etal. (1998a)*
Morgan etal. (1998a)*
Chang et al. (2005)
Chang et al. (2005)
Relative risk
Figure 3.2-1. Relative risks (95% CI) for associations of 24-h NOi (per 20 ppb) and daily
1 hour maximum* NOi (per 30 ppb) with hospitalizations or emergency
department visits for cardiac diseases. Primary author and year of
publication, city, stratification variable(s), and lag are listed. Results for lags
0 or 1 are presented as available.
Location Season
Ontario
Ontario
Ontario
Europe
Valencia
Spain, Multicity
Australia, NZ
Australia, NZ
Australia, Multicity
Australia, Multicity
Australia, Multicity
Sydney
Sydney
Sydney
Sydney
Sydney
Sydney
/ /
Taipei Warm
Taipei Cool
Age
65+
65+
65+
Ml Survivors 35+
All ages
All ages
65+
15-64
All
15-64
65+
65+
65+
65+
All
65+
0-65
All ages
All ages
Lag
0
0-1 .
0-2 .
0
2
0-1 .
0-1 .
0-1 .
0-1 .
0
0-1
0
1
0-1 .
0
0
0
0-2 .
0-2 .
-i—
JL
1
-f-
-1-
-1-
1
_|_
1
I III
.9 1.1 1.3 1.5
1
2
3
4
5
6
7
2005; Yang et al., 2004b; Chang et al., 2005; Morgan et al., 1998a; Burnett et al., 1997a, 1999).
See Annex AX6.2.1.6 for a detailed description of results from multipollutant models.
3.2.2.2 Hospital Admissions for Stroke and Cerebrovascular Disease (ICD9 430-448)
Studies of the association between all cerebrovascular disease and ambient NC>2
concentration are summarized in Figure 3.2-2. Results from these studies are generally
inconsistent. Metzger et al. (2004) reported a significant increase in cerebrovascular disease
March 2008
3-68
DRAFT-DO NOT QUOTE OR CITE
-------
Reference Location
Linn et al. (2000) Metro LA
Metzger et al. (2004)* Atlanta
Peel et al. (2007)* Atlanta
Peel et al. (2007)* Atlanta
Ballesteretal. (2001)* Valencia
Poloneiki et al. (1 997) London
Chan etal. (2006) Taipei
Wong etal. (1999) Hong Kong
Age Lag
All Ages 0
Ail Ages 3 d moving
Ail Ages 0-2
Ail Ages 0-2
Ail Ages 4
Ail Ages 1
50+ 0
Ail Ages 0-1
mmm
L^u
H|
-I-
1
1
I III
.9 1.1 1.3 1.5
Relative risk
Figure 3.2-2. Relative risks (95% CI) for associations of 24-h NOi exposure (per 20 ppb)
and daily 1 h maximum NOi* (per 30 ppb) with hospitalizations for all
cerebrovascular disease. Primary author and year of publication, city,
stratification variable(s), and lag are listed. Results for lags 0 or 1 are
presented as available.
1 emergency visits in Atlanta. However, Peel et al. (2007) did not find associations between
2 cerebrovascular disease visits and NC>2 concentrations among those with hypertension and
3 diabetes in the same city. The daily 1-h max NO2 level in Atlanta during the study period ranged
4 from 26 to 45.9 ppb (Metzger et al., 2004; Peel et al., 2007). Ballester et al. (2001) reported a
5 relatively large increased risk in cerebrovascular admissions in the Spanish city of Valencia at
6 lag 4, while Poloniecki et al. (1997) and Ponka and Virtanen (1996) did not observe associations
7 in London and Helsinki. Two Asian studies report positive but nonsignificant associations of
8 cerebrovascular disease with 24-h avg NO2 (Chan et al., 2006; Wong et al., 1999). The 24-h avg
March 2008
3-69
DRAFT-DO NOT QUOTE OR CITE
-------
1 NO2 levels reported for Taipei and Hong Kong were approximately 30 ppb and 27 ppb,
2 respectively (Chan et al., 2006; Wong et al., 1999).
3 Studies of hospital admissions or ED visits for specific cerebrovascular diseases provide
4 little evidence for a NO2 effect. In a large study, conducted in metropolitan Los Angeles where
5 the mean 24-h NO2 level ranges from 28 to 41 ppb depending on the season, no association was
6 observed for all cerebrovascular disease (Linn et al., 2000). However, authors reported an
7 increase in hospitalizations of 4.0% (95% CI: 2.0, 6.0) for occlusive stroke per 20 ppb increase
8 in NO2.
9 Wellenius et al. (2005) found a 5% increase in ischemic stroke (IS) admissions per
10 20-ppb increase in 24-h avg NO2 level. A study of all-stroke in Ontario reported null findings
11 for 24-h avg NO2 at lags 0 and 1 (Ito et al. 2004). Villeneuve et al. (2006) reported an
12 association between NO2 exposure and IS during the winter months among the elderly (OR =
13 1.41 [95% CI: 1.13, 1.75], per 20 ppb, lag 3 day average). Villeneuve et al. (2006) also reported
14 positive but nonsignificant associations for hemorrhagic stroke (HS) (OR = 1.25 95% CI: 0.91,
15 1.71 per 20-ppb increase in NO2). No associations between air pollutants and stroke were
16 reported in a multicity study conducted in Australia and New Zealand (Barnett et al., 2006). An
17 increase in 24-h avg NO2 resulted in increased risk of hospitalization for primary intracerebral
18 hemorrhage (PIH) (OR: 1.68 [95% CI: 1.39, 2.04] lag 0 to 2 per 20 ppb increase), and ischemic
19 stroke (IS) (OR: 1.67 95% CI: 1.49 1.88, lag 0-2) during the warm season in Taiwan (Tsai
20 et al., 2003).
21 Several investigators presented estimates for the association of NO2 with cerebrovascular
22 outcomes from multipollutant models. The association of NO2 with stroke was not robust to
23 adjustment for CO in a Canadian study (Villeneuve et al., 2006). Although results from a
24 Taiwanese study indicated the effect of NO2 on stroke admissions was robust in two-pollutant
25 models, the authors noted that the association of NO2 with stroke might not be causal if NO2 is a
26 surrogate for other components of the air pollution mixture (Tsai et al., 2003).
27
28 3.2.3 Summary of Evidence of the Effect of Short-Term NOi Exposure on
29 Cardiovascular Morbidity
30 The available evidence on the effect of short-term exposure to NO2 on cardiovascular
31 health effects is inadequate to infer the presence or absence of a causal relationship at this time.
32 Evidence from epidemiologic studies of HRV, repolarization changes, and cardiac rhythm
March 2008 3-70 DRAFT-DO NOT QUOTE OR CITE
-------
1 disorders among heart patients with ICDs are inconsistent. In most studies, observed
2 associations with PM were similar or stronger than associations with NO2. Generally positive
3 associations between ambient NO2 concentrations and hospital admissions or ED visits for CVD
4 have been reported in single-pollutant models; however, most of the effect estimates were
5 diminished in multipollutant models also containing CO and PM indices. Mechanistic evidence
6 of a role for NO2 in the development of CVDs from studies of biomarkers of inflammation, cell
7 adhesion, coagulation, and thrombosis is lacking. Furthermore, the effects of NO2 on various
8 hematological parameters in animals are inconsistent and, thus, provide little biological
9 plausibility for effects of NO2 on the cardiovascular system. However, there is limited evidence
10 from controlled human exposure studies suggesting a reduction in hemoglobin with NO2
11 exposure at concentrations of 1.0 to 2.0 ppm (with 3-h exposures) that requires confirmation.
12
13
14 3.3 MORTALITY ASSOCIATED WITH SHORT-TERM NO2
15 EXPOSURE
16 There was no epidemiologic study reviewed in the 1993 AQCD that examined the
17 mortality effects of ambient NO2. Since the 1993 AQCD, a number of studies, mostly using
18 time-series analyses, reported short-term mortality risk estimates for NO2 (see Annex Table
19 AX6.3-19). However, since most of these studies' original focus or hypothesis was on PM, a
20 quantitative interpretation of the NO2 mortality risk estimates requires caution. Risk estimates
21 are summarized across studies after reviewing individual multicity studies.
22
23 3.3.1 Multicity Studies and Meta-Analyses
24 In reviewing the range of mortality risk estimates, multicity studies provide the most
25 useful information because they analyze multiple cities data in a consistent method, avoiding
26 potential publication bias. Risk estimates from multicity studies usually are reported for
27 consistent lag days, further reducing potential bias caused by choosing the "best" lag in
28 individual studies. There have been several multicity studies from the United States, Canada,
29 and Europe. Meta-analysis studies also provide useful information on describing heterogeneity
30 of risk estimates across studies, but unlike multicity studies, the heterogeneity of risk estimates
31 seen in meta-analysis may also reflect the variation in analytical approaches across studies.
32 Thus, we focus our review mainly on the results from multicity studies, and effect estimates from
March 2008 3-71 DRAFT-DO NOT QUOTE OR CITE
-------
1 these studies are summarized. Discussion will focus on the studies that were not affected by
2 GAMs with convergence issues (Dominici et al., 2002; Ramsay et al., 2003) unless otherwise
3 noted when the studies raise relevant issues.
4
5 3.3.1.1 National Morbidity, Mortality, and Air Pollution Study (NMMAPS)
6 The time-series analysis of the largest 90 U.S. cities (Samet et al., 2000; reanalysis
7 Dominici et al., 2003) in the National Morbidity, Mortality, and Air Pollution Study (NMMAPS)
8 is by far the largest multicity study conducted to date to investigate the mortality effects of air
9 pollution, but its primary interest was PM (i.e., PMi0), and NO2 was not measured in 32 of the 90
10 cities. This study's model adjustment for weather effects employs more terms than other time-
11 series studies in the literature, suggesting that the model adjusts for potential confounders more
12 aggressively than the models in other studies. PMio and 63 (in summer) appeared to be more
13 strongly associated with mortality than the other gaseous pollutants. Regarding NO2, SO2, and
14 CO, the authors stated, "The results did not indicate associations of these pollutants with total
15 mortality." PMio, NO2, SO2, and CO showed the strongest association at lag 1 day (for Os, it
16 was lag 0 day), and the addition of other copollutants in the model at lag 1 day hardly affected
17 the mortality risk estimates for PMio or the gaseous pollutants. Figure 3.3-1 shows the total
18 mortality risk estimates for NO2 from Dominici et al. (2003). The NO2 risk estimates in the
19 multipollutant models were about the same or larger. Thus, these results do not indicate that the
20 NO2-mortality association was confounded by PMio or other pollutants (and vice versa).
21
22 3.3.1.2 Canadian Multicity Studies
23 There have been four Canadian multicity studies conducted by the same group of
24 investigators (Burnett et al., 1998, 2000, 2004; Brook et al., 2007). This section focuses on
25 Burnett et al. (2004) and Brook et al. (2007), as these studies are most extensive both in terms of
26 the length and coverage of cities.
27 Total (nonaccidental), cardiovascular, and respiratory mortality were analyzed in the
28 Burnett et al. (2004) study of 12 Canadian cities from 1981 to 1999. Daily 24-h avg as well as
29 1-h max values were analyzed for all the gaseous pollutants and coefficient of haze (CoH). For
30 PM2.5, coarse PM (PMio-2.s), PMio, CoH, SO2, and CO, the strongest mortality association was
31 found at lag 1, whereas for NO2, it was the 3-day moving average (i.e., average of 0-, 1-, and 2-
32 day lags), and for O3, it was the 2-day moving average. Of the single-day lag estimates for NO2,
March 2008 3-72 DRAFT-DO NOT QUOTE OR CITE
-------
.a
a. 1.0
Q.
? .1
j_ '-jjj
a. £ 0.5
C
— u
1 o
« o n
o *•' u
5 C
fl)
&8 nr
w g> -0.5
to o
_C C
U
^ -1.0
f
i
;
Lag
1
;
o
:
:
1
i
' i
;
;
t
>
u
La<
:
:
,
;
31
;
1
i
;
( >
.
:
:
K
ii
1
La<
(
•
:
32
i
;
'
:
:
i
»
i
;
;
i
ABCDEABCDEABCDE
Models
Figure 3.3-1. Posterior means and 95% posterior intervals of national average estimates
for NO2 effects on total mortality from nonexternal causes at lags 0,1, and
2 within sets of the 90 cities with pollutant data available. Models A = NO2
alone; B = NO2 + PMi0; C = NO2 + PMio + O3; D = NO2 + PMio + SO2;
E = SO2 + PM10 + CO.
Source: Dominici et al. (2003).
1 lag 1 day showed the strongest associations, which is consistent with the NMMAPS result, but
2 its risk estimate was more than 4 times larger than that for the NMMAPS study. The 24-h avg
3 values showed stronger associations than the 1-h max values for all the gaseous pollutants and
4 CoH except for 63. The pooled NC>2 mortality risk estimate in a single-pollutant model (for all
5 available days) was 2.0% (95% CI: 1.1, 2.9) per 20-ppb increase in the 3-day moving average of
6 NO2. The magnitudes of the effect estimates were similar for total, cardiovascular, and
7 respiratory mortality. Larger risk estimates were observed for warmer months. NO2 was most
8 strongly correlated with CoH (r = 0.60), followed by PM2.5 (r = 0.48). The NO2-mortality
9 association was not sensitive to adjustment for these or any of other pollutants in the two-
10 pollutant models. However, Burnett et al. (2004) noted that simultaneous inclusion of daily
11 PM2.5 data (available for 1998 and 2000; sample size comparable to the main analysis [every 6th
12 day from 1981 to 1999] but more recent years) and NO2 in the model resulted in a considerable
March 2008
3-73
DRAFT-DO NOT QUOTE OR CITE
-------
1 reduction of the NC>2 risk estimates. Authors discussed that reducing combustion would result in
2 public health benefits because NO2 or its products originate from combustion sources, but
3 cautioned that they could not implicate NO2 as a specific causal pollutant.
4 Brook et al. (2007) further examined data from 10 Canadian cities with a special focus on
5 NC>2 and the role of other traffic-related air pollutants. Again, NO2 showed the strongest
6 associations with mortality among the pollutants examined including NO, and none of the other
7 pollutants substantially reduced NC>2 risk estimates in multipollutant models. The analysis also
8 confirmed the Burnett et al. (2004) study results that NC>2 risk estimate was larger in the warm
9 season. Generally, NO showed stronger correlation with the primary VOCs (e.g., benzene,
10 toluene, xylenes) than NO2 or PM2.5. NO2 was more strongly correlated with the organic
11 compounds than it was with the PM mass indices or trace metals in PM2.s. Brook et al. (2007)
12 concluded that the strong NO2 effects seen in Canadian cities could be a result of it being the best
13 indicator, among the pollutants monitored, of fresh combustion as well as photochemically
14 processed urban air.
15 In summarizing the Canadian multicity studies, NO2 was most consistently associated
16 with mortality among the air pollutants examined, especially in the warm season. Adjustments
17 for PM indices and its components generally did not reduce NO2 risk estimates. NO2 also was
18 shown to be associated with organic compounds that are indicative of combustion products
19 (traffic-related air pollution) and photochemical reactions.
20
21 3.3.1.3 Air Pollution and Health: A European Approach (APHEA) Studies
22 The APHEA project is a European multicity effort, analyzing data from multiple studies
23 using a standardized methodology. This section focuses on the more recent APHEA2 studies
24 which included 29 European cities.
25 Samoli et al. (2006) analyzed 29 APHEA2 cities to estimate NO2 associations for total,
26 cardiovascular, and respiratory deaths. The average of lags 0-1 days were chosen a priori to
27 avoid potential bias with the "best" lag approach. In addition, the association of total mortality
28 with NO2 over 6 days (lags 0-5) were summarized over all cities using a cubic polynomial
29 distributed lag model. Results from this model suggested multiday effects, with the strongest
30 association shown at lag 1 day, which is consistent with the results from NMMAPS and
31 Canadian multicity studies. The risk estimates for total, cardiovascular, and respiratory causes
32 were comparable. In the two-pollutant models with black smoke, PMio, SO2, and Os, the risk
March 2008 3-74 DRAFT-DO NOT QUOTE OR CITE
-------
1 estimates for total and cardiovascular mortality were not affected. The second-stage analysis
2 examined possible effect modifiers. For total and cardiovascular mortality, the geographical area
3 (defined as western, southern, and central eastern European cities) was the most important effect
4 modifier (estimates were lower in eastern cities), followed by smoking prevalence (NO2 risk
5 estimates were higher in cities with a lower prevalence of smoking). The authors concluded that
6 the results showed effects of NO2 on mortality, but that the role of NO2 as a surrogate of other
7 unmeasured pollutants could not be completely ruled out.
8 In an earlier study, Katsouyanni et al. (2001; reanalysis, 2003) analyzed data from 29
9 European cities and reported risk estimates for PMio and not for NC>2, but found that the cities
10 with higher NC>2 levels tended to have larger PMio risk estimates. Furthermore, simultaneous
11 inclusion of PMio and NO2 reduced the PMio risk estimate by half. An analysis of the elderly
12 mortality in the same 28 cities (Aga et al., 2003) also found a similar effect modification of PM
13 by NC>2. Thus, PM and NC>2 risk estimates in these European cities may be reflecting the health
14 effects of the same air pollution source and/or act as effect modifiers of each other.
15
16 3.3.1.4 The Netherlands Study
17 While the Netherlands studies for the 1986 to 1994 data (Hoek et al., 2000, 2001;
18 reanalysis in Hoek, 2003) are not multicity studies and the Netherlands data were also analyzed
19 as part of APHEA2 (Samoli et al., 2006), the results from the reanalysis (Hoek, 2003) are
20 discussed here, because the database comes from a large population (14.8 million for the entire
21 country) and a more extensive analysis was conducted than in the multicity studies. PMio, black
22 smoke, O3, NO2, SO2, CO, sulfate (SO42 ), and nitrate (N(V) were analyzed at lags 0, 1, and
23 2 days and the average of lags 0-6 days. All the pollutants were associated with total mortality,
24 and for single-day models, lag 1 day showed strongest associations for all the pollutants. NC>2
25 was most highly correlated with black smoke (r = 0.87), and the simultaneous inclusion of NC>2
26 and black smoke reduced both pollutants' risk estimates (the NC>2 estimate was reduced by more
27 than 50%). PMio was less correlated with NC>2 (r = 0.62), and the simultaneous inclusion of
28 these pollutants resulted in an increase in the NC>2 risk estimate.
29
30 3.3.1.5 Other Multicity Studies
31 Other European multicity studies, conducted in eight Italian cities (Biggeri et al., 2005),
32 nine French cities (Le Tertre et al., 2002) and seven Spanish cities (Saez et al., 2002) provide
March 2008 3-75 DRAFT-DO NOT QUOTE OR CITE
-------
1 evidence for a short-term NO2 effect on mortality. An additional multicity study was conducted
2 in Australian cities (Simpson et al., 2005b). The studies by Biggeri et al. (2005) and Simpson
3 et al. (2005b) are summarized in this section. The studies by Le Tertre et al. (2002) and Saez
4 et al. (2002), conducted using Generalized Additive Model (GAM) methods with the default
5 convergence setting, are presented in Annex Table AX6.3-19.
6 Biggeri et al. (2005) analyzed eight Italian cities (Turin, Milan, Verona, Ravenna,
7 Bologna, Florence, Rome, and Palermo) from 1990 to 1999. Only single-pollutant models were
8 examined in this study. Statistically significant positive associations were observed between
9 NO2 and total, cardiovascular, and respiratory mortality, with the largest effect estimate observed
10 for respiratory mortality. Since all the pollutants showed positive association and the
11 correlations among the pollutants were not presented, it is not clear how much of the observed
12 associations are shared or confounded. The mortality risk estimates were not heterogeneous
13 across cities for all the gaseous pollutants.
14 Simpson et al. (2005b) analyzed data from four Australian cities (Brisbane, Melbourne,
15 Perth, and Sydney) using methods similar to the APHEA2 approach. They also examined
16 sensitivity of results to alternative regression models. Associations between mortality and NC>2,
17 63, and nephelometer readings (a measure of PM) were examined at single-day lag 0, 1,2, and
18 3 days and using the average of 0- and 1-day lags. Among the three pollutants, correlation was
19 strongest between NO2 and nephelometer readings, ranging from (r ~ 0.62 among the four
20 cities). Of the three pollutants, NO2 showed the largest mortality risk estimates per interquartile
21 range. Similar to the study by Biggeri et al. (2005), the strongest association was observed
22 between NO2 and respiratory mortality, compared to total or cardiovascular mortality. The three
23 alternative regression models yielded similar results. The NO2 risk estimates were not sensitive
24 to the addition of nephelometer readings in the two-pollutant models for total mortality, but the
25 nephelometer risk estimate was greatly reduced in the model with NO2.
26
27 3.3.1.6 Meta-Analyses of NO2 Mortality Studies
28 Stieb et al. (2002) reviewed time-series mortality studies published between 1985 and
29 2000, and conducted a meta-analysis to estimate combined effects for each of PMio, CO, NO2,
30 O3, and SO2. Since many of the studies reviewed in that analysis were affected by the GAM
31 convergence issue, Stieb et al. (2003) updated the estimates by separating the GAM versus non-
32 GAM studies and by single- versus multipollutant models. There were more GAM estimates
March 2008 3-76 DRAFT-DO NOT QUOTE OR CITE
-------
1 than non-GAM estimates for all the pollutants except 862. For NO2, there were 11 estimates
2 from single-pollutant models and only 3 estimates from multipollutant models. The lags and
3 multiday averaging used in these estimates varied. The combined estimate for total mortality
4 was 0.8% (95% CI: 0.2, 1.5) per 20-ppb increase in the daily average NO2 from the single-
5 pollutant models and 0.4% (95% CI: -0.2, 1.1) per 20-ppb increase in the 24-h average from the
6 multipollutant models. Note that, although the estimate from the multipollutant models was
7 smaller than that from the single-pollutant models, the number of the studies for the
8 multipollutant models was small (3), also, the data extraction procedure of this meta-analysis for
9 the multipollutant models was to extract from each study the multipollutant model that resulted
10 in the greatest reduction in risk estimate compared with that observed in single-pollutant models.
11 It should be noted that all the multeity studies whose combined estimates have been discussed
12 above were published after this meta-analysis.
13
14 3.3.2 Summary of Evidence of the Effect of Short-Term NOi Exposure on
15 Mortality
16 The epidemiologic evidence on the effect of short-term exposure to NC>2 on total
17 nonaccidental and cardiopulmonary mortality is suggestive but not sufficient to infer a causal
18 relationship. The epidemiologic studies are generally consistent in reporting positive
19 associations. However, there is little evidence available to evaluate coherence and plausibility
20 for the observed associations, particularly for cardiovascular and total mortality.
21 In the short-term exposure studies, the range of NC>2 total mortality risk estimates is 0.5
22 to 3.6% per 20-ppb increase in the 24-h average NC>2 or 30-ppb increase in daily 1-h max (Figure
23 3.3-2). The use of various lag periods, averaging days, and distributed lags does not appear to
24 alter the estimates substantially. The heterogeneity of estimates in these studies may be due to
25 several factors, including the differences in (1) model specification, (2) NC>2 levels, and (3) effect
26 modifying factors. Interestingly, the Canadian 12-city study showed combined risk estimates
27 (average of 0-1 day or single 1-day lag) about 4 times larger than that for the U.S. estimate,
28 despite the fact that the range of Canadian NO2 concentrations (10 to 26 ppb) was somewhat
29 lower than that for the U.S. data (9 to 39 ppb for the 10%-trimmed data). In fact, the NMMAPS
30 estimate is the smallest among the multicity studies. Since a similar pattern (i.e., the NMMAPS
31 estimate being the smallest among multicity studies) was seen for PMio mortality risk estimates
March 2008 3-77 DRAFT-DO NOT QUOTE OR CITE
-------
U.S. 90 cities study (Dominici et al.. 2003}
24-h average, lag 1 day
with PMioandSO2
Canadian 12 cities study (Burnett et al., 2004)
24-h average, average of lag 0-2 days
with 03
24-h average, lag 1 day (every-6th-day)
with PM2.5
European 30 cities study (Samoli et al., 2006}
1-hr daily max.average of lag 0-1 days
with SO2
Italian 8 cities study (Bigger) et al., 2005)
24-h average.average of lag 0-1 days
The Netherlands study (Hoek. 2003)
24-hr average, lag 1 day
average of lag 0-6 days
with black smoke
Australian 4 cities study (Simpson et al., 2005t»
1-h daily max.average of lag 0-1 days
with fine particles by nephelorneter
Mela-analysis (Stieb et al., 2003)
24-h average, lag and multiday averages mixed
with copollutants that showed largest reduction
Percent Axcess Mortality
0246
Figure 3.3-2. Combined NOi mortality risk estimates from multicity and meta-analysis
studies. Risk estimates are computed per 20-ppb increase for 24-h average
or 30-ppb increase for 1-h daily maximum NOi concentrations. For
multipollutant models, results from the models that resulted in the greatest
reduction in NOi risk estimates are shown.
(U.S. Environmental Protection Agency, 2004), it is possible that this may be due to the
difference in model specifications.
Several multicity studies provided risk estimates for broad cause-specific categories
(typically all-cause, cardiovascular, and respiratory) using consistent lags/averaging for broad
March 2008
3-78
DRAFT-DO NOT QUOTE OR CITE
-------
1 causes (cardiovascular and respiratory), but the patterns were not always consistent. This
2 inconsistency was likely due to smaller sample size, or the lags reported not being consistent
3 across the specific causes examined (Figure 3.3-3). While the smaller multicity studies (the
4 Italian and Australian studies) reported larger risk estimates for respiratory mortality, the larger
5 Canadian and APHEA2 studies reported comparable risk estimates among the broad specific
6 causes of deaths. In addition, since other pollutants also showed similar associations with these
7 causes or categories, it is difficult to discuss consistency with causal inference that is specific to
8 NO2. The multipollutant models in these studies generally did not alter NO2 risk estimates,
9 except for the Netherlands study in which NO2 was highly correlated with the copollutant black
10 smoke. While the multipollutant results generally suggest a lack of confounding, it is difficult to
11 attribute the observed excess mortality risk estimates to NO2 alone.
12 While the multicity studies examining the relationship between short-term NO2 exposure
13 and mortality observed statistically significant associations for total, cardiovascular, and
14 respiratory causes, the issue of surrogacy of the role of NO2 and possible interactions with PM
15 and other pollutants remain unresolved. As reviewed in earlier sections, controlled human
16 exposure studies, by necessity, are limited to acute, fully reversible functional and/or
17 symptomatic responses in healthy or mildly asthmatic subjects. Animal studies have not used
18 mortality as an endpoint in acute exposure studies. However, a number of animal studies
19 (described in Section 3.1.3) have shown biochemical, lung host defense, permeability, and
20 inflammation effects with acute exposures and may provide limited biological plausibility for
21 mortality in susceptible individuals. A 5-ppm NO2 exposure for 24 h in rats caused increases in
22 blood and lung total glutathione (GSH) and a similar exposure resulted in impairment of alveolar
23 surface tension of surfactant phospholipids due to altered fatty acid content. A fairly large body
24 of literature describes the effects of NO2 on lung host defenses at low exposures. However, most
25 of these effects are seen only with subchronic or chronic exposure and, therefore, do not
26 correlate well with the short lag times evidenced in the epidemiologic studies. The
27 corresponding evidence of interaction between NO2 and other pollutants in controlled human and
28 toxicologic studies are also very limited. Thus, there is a gap between the observed associations
29 between short-term exposure to NO2 mortality reported in observational epidemiologic studies
30 and available evidence from controlled human and toxicologic studies in establishing a causal
31 link.
March 2008 3-79 DRAFT-DO NOT QUOTE OR CITE
-------
Canadian 12 cities study _
(Burnett et al., 2004) -
Avg. 0-2 days -
European 30 cities study
(Samoll et al., 2006)
Avg. 0-1 days
Italian 8 cities study
(Biggerl et al., 2005)
Avg. 0-1 days
Australian 4 cities study -
(Simpson et al., 2005b) -
Avg. 0-1 days -
o
i
Percent Excess Mortality
2 4 6 8 10 12
I I I I I I
-Xr
• Total
^ Cardiovascular
X Respiratory
-X-
Figure 3.3-3. Combined NOi mortality risk estimates for broad cause-specific categories
from multicity studies. Risk estimates are computed per 20-ppb increase for
24-h average or 30-ppb increase for 1-h daily maximum NOi concentrations.
1 Results from several large U.S. and European multicity studies and a meta-analysis study
2 observed positive associations between ambient NC>2 concentrations and risk of all-cause
3 (nonaccidental) mortality, with effect estimates ranging from 0.5 to 3.6% excess risk in mortality
4 per standardized increment1 (Section 3.3.1, Figure 3.3-2). In general, the NC>2 effect estimates
5 were robust to adjustment for copollutants. Both cardiovascular and respiratory mortality have
March 2008
3-80
DRAFT-DO NOT QUOTE OR CITE
-------
1 been associated with increased NO2 concentrations in epidemiologic studies (Figure 3.3-3);
2 however, similar associations were observed for other pollutants, including PM and 862. The
3 range of mortality excess risk estimates was generally smaller than that for other pollutants such
4 as PM.
5 While NC>2 exposure, alone or in conjunction with other pollutants, may contribute to
6 increased mortality, evaluation of the specificity of this effect is difficult. Clinical studies
7 showing hematologic effects and animal toxicologic studies showing biochemical, lung host
8 defense, permeability, and inflammation changes with short-term exposures to NC>2 provide
9 limited evidence of plausible pathways by which risks of morbidity and, potentially, mortality
10 may be increased, but no coherent picture is evident at this time.
11
12
13 3.4 RESPIRATORY EFFECTS ASSOCIATED WITH LONG-TERM
14 NO2 EXPOSURE
15 There was no epidemiologic evidence available in the 1993 AQCD on the respiratory
16 effects of long-term exposure (>2 weeks) to ambient NC>2. The 1993 AQCD reported that
17 chronic exposure to high NC>2 levels (>8 ppm) caused emphysema in several animal species.
18 Since the 1993 AQCD, a number of studies reported associations between long-term NC>2
19 exposure and respiratory effects (see Annex Tables AX6.3-15, AX6.3-16, and AX6.3-17).
20
21 3.4.1 Lung Function Growth
22
23 Epidemiologic Studies
24 Studies of lung function demonstrate some of the strongest effects of long-term exposure
25 to NC>2. Recent cohort studies have examined the effect of long-term exposure to NC>2 in both
26 children and adults (see Annex Table AX6.3-15). Forest plots of the results for FEVi and FVC
27 from the three major children's cohort studies (Gauderman et al., 2004; Rojas-Martinez et al.,
28 2007a,b; Oftedal et al., 2008) are presented in Figures 3.4-1 and 3.4-2.
29 The Children's Health Study (CHS) in southern California is a longitudinal cohort study
30 designed to investigate the effect of chronic exposure to several air contaminates (including
31 NC>2) on respiratory health in children. Twelve California communities were selected based on
32 historical data indicating different levels of specific pollutants. In each community, monitoring
March 2008 3-81 DRAFT-DO NOT QUOTE OR CITE
-------
Study
GsudBrrnsnn
(2004)
Oftedal (2008)
Oftedal (2008)
Oftedal (2008)
Oftedal (2008)
Oftedal (2008)
Oftedal (2008)
Rojas Martinez
(2007a,b)
Rojas Martinez
(2007a,b)
Location
Southern
California
Oslo
Oslo
Oslo
Oslo
Oslo
Oslo
Mexico Oily
Mexico City
Gender
Both
Both
Boys
Girls
Both
Boys
Girts
Boys
Girts
Period
1993-2001
1991-1992
1991-1992
1991-1992
1992-2002
1991-2002
1991-2002
1996-1999
1996-1999
Baseline age
g
1st yr of life
Istyroflife
Istyrof life
9-10
9-10
9-10
10
10
N
1759 -
1 ft A -J
1847 •
938 -
909 -
1847-
938 -
909 -
1103 -
1115 -
:
I
g
1 1 1
-25 -20 -15 -10
i
1
-505
A
l
10
FEV, (ml) per 20 ppb N02 per year
Study
Gaudermann
(2004)
Ottedal (2008)
Oftedal (2008)
Ottedal (2008)
Oftedal (2008)
Oftedal (2008)
Oftedal (2008)
Rojas Martinez
(2007a,b)
Rojas Martinez
(2007a,b)
Location
Southern
California
Oslo
Oslo
Oslo
Oslo
Oslo
Oslo
Mexico City
Mexico City
Gender
Both
Both
Boys
Girts
Both
Boys
Girts
Boys
Girts
Period
1993-2001
1991-1992
1991-1992
1991-1992
1992-2002
1992-2002
1992-2002
1996-1999
1996-1999
Baseline age
8
1 st yr of life
Istyroflife
Istyroflife
9-10
9-10
9-10
10
10
N
1759 -
1A47 -
IO*H
938 -
909 -
1847 •
938-
909 -
1103 -
1115 -
1 1 1
-»•
-
_
1-
1
1 1
B_
I
-100 -75 -50 -25 0 25 50 75 100
FVC (ml) per 20 pg/mL of PM10 per year
Figure 3.4-1. Decrements in forced expiratory volume in 1 s (FEVi) associated with a
20-ppb increase in NO2 (A) and a 20-ug/m3 increase in PMi0 (B) in children,
standardized per year of follow-up. Results from three major children's
long-term cohort studies are presented.
Source: Gauderman et al. (2004); Oftedal et al. (2008), Rojas-Martinez et al. (2007a,b).
March 2008
3-82
DRAFT-DO NOT QUOTE OR CITE
-------
Study Location
Gaudermann (2004) Southern
California
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Rojas Martinez (2007a,b) Mexico City
Rojas Martinez (2007a,b) Mexico City
Gaudermann (2004) Southern
California
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Oftedal (2008) Oslo
Rojas Martinez (2007a,b) Mexico City
Rojas Martinez (2007a,b) Mexico City
Gender
Both
Both
Boys
Girls
Both
Boys
Girls
Boys
Girls
Both
Both
Boys
Girls
Both
Boys
Girls
Boys
Girls
Period Baseline N
age
1993-2001 8 1759 -
1991-1992 1st yr of life 1847 -
1991-1992 1st yr of life 938 -
1991-1992 1st yr of life 909 -
1992-2002 9-10 1847 -
1991-2002 9-10 938 -
1991-2002 9-10 909 -
1996-1999 10 1103 -
1996-1999 10 1115 •
-3
H
f
ft
}
-•-
^_
i i i i i i i i i i
5 -30 -25 -20 -15 -10 -5 0 5 10 15 20
FVC (ml) per 20 ppb N02 per year
1993-2001 8 1759 -
1991-1992 Istyroflife 1847 -
1991-1992 Istyroflife 938 -
1991-1992 Istyroflife 909 -
1992-2002 9-10 1847 -
1992-2002 9-10 938 -
1992-2002 9-10 909 -
1996-1999 10 1103 -
1996-1999 10 1115 '
-4 H
1
J
:-
fl
1
-100 -75 -50 -25 0 25 50 75 100
FVC (mL) per 20 (jg/mL of PM10 per year
Figure 3.4-2. Decrements in forced vital capacity (FVC) associated with a 20-ppb increase
in NOi (A) and a 20-ug/m3 increase in PMio (B) in children, standardized per
year of follow-up. Results from three major children's long-term cohort
studies are presented.
Source: Gauderman et al. (2004); Oftedal et al. (2008), Rojas-Martinez et al. (2007a,b).
March 2008
3-83
DRAFT-DO NOT QUOTE OR CITE
-------
1 sites were set up to measure hourly 63, NO2, and PMio and 2-week averages of PM2.5, and acid
2 vapor. Children in grades 4, 7, and 10 were recruited though local schools. The study followed
3 children for 10 years, with annual questionnaires and lung function measurement. The study had
4 several important characteristics: it was prospective and exposure and outcome data were
5 collected in a consistent manner over the duration of the study, and confounding by SES was
6 controlled in the models by selecting communities similar in demographic characteristics at the
7 outset.
8 Peters et al. (1999) reported the initial results from the CHS: a cross-sectional analysis of
9 lung function tests conducted on 3,293 children in the first year of the study. Both NO2 and
10 PMio were associated with decreases in FVC, FEVi, and MMEF. Avol et al. (2001) then studied
11 the effect of relocating to areas of differing air pollution levels in 110 children 10 years of age
12 who were participating in the CHS. As a group, subjects who had moved to areas of lower NO2
13 showed increased growth in lung function, but the effects did not reach statistical significance.
14 In general, the authors focused on associations with PM, where larger and statistically significant
15 effects were observed.
16 In 2004, Gauderman et al. reported results for an 8-year follow up of the children
17 enrolled in grade 4 (n = 1,759). Exposure to NO2 was significantly associated with deficits in
18 lung growth over the 8-year period. The difference in FVC for children exposed to the lowest
19 versus the highest levels of NO2 (34.6 ppb) was -95.0 mL (95% CI: -189.4 to-0.6). For FEVi,
20 the difference was -101.4 mL (95% CI: -165.5 to -38.4), and for MMEF, -221.0 mL/s (95%
21 CI: -377.6,-44.4). Results were similar for boys and girls and among children without a
22 history of asthma. These deficits in growth of lung function resulted in clinically significant
23 differences in FEVi at age 18. In addition, the NO2 concentration associated with deficits in lung
24 growth was 34.6 ppb (range of means across communities: 4.4-39.0 ppb), a level below the
25 current standard. Similar results were reported for acid vapor (resulting primarily from
26 photochemical conversions of NOx to HNOs). These results are depicted in Figure 3.4-3. The
27 authors concluded that the effects of NO2 could not be distinguished from the effects of particles
28 (PM2.s and PMio) as NO2 was strongly correlated with these contaminants (0.79, and 0.67,
29 respectively).
30 More recently, Gauderman et al. (2007) has reported results of an 8-year follow-up on
31 3,677 children who participated in the CHS. Children living <500 m from a freeway (n = 440)
March 2008 3-84 DRAFT-DO NOT QUOTE OR CITE
-------
2
"re
T3
«
0
TJ
iL
'o
0
CO
V
!>
10
8
6
2
0
~ R = 0.04 4up
P = 0.89
-
~ •» LB *ML RW
T Kv
SD
*AT *AL
*SM * LM *LE
LN
i TI i i i
HI
35
45
55
65
O3 from 10 a.m. to 6 p.m. (ppb)
75
4UP
10 20 30
N02 (ppb)
40
Acid Vapor (ppb)
*UP
*ML
10
8
6
P = 0.002
AT
SM
LN
10
15
20
25
30
Elemental Carbon (pg/m3)
Figure 3.4-3. Proportion of 18-year olds with a FEVi below 80% of the predicted value
plotted against the average levels of pollutants from 1994 through 2000 in the
12 southern California communities of the Children's Health Study.
AL = Alpine; AT = Atascadero; LA = Lake Arrowhead; LB = Long Beach; LE = Lake Elsinore; LM = Lompoc;
LN = Lancaster; ML = Mira Loma; RV = Riverside; SD = San Dimas; SM = Santa Maria; UP = Upland
Source: Derived from Gauderman et al. (2004).
March 2008
3-85
DRAFT-DO NOT QUOTE OR CITE
-------
1 had significant deficits in lung growth over the 8-year follow-up compared to children who lived
2 at least 1500 m from a freeway. The difference in FVC was -63 mL (-131 to 5); the difference
3 in FEVi -81 mL (-143 to -18); and the difference in MMEF -127 mL/s (-243 to -11). This
4 study did not attempt to measure specific pollutants near freeways or to estimate exposure to
5 specific pollutants for study subjects. Thus, while the study presents important findings with
6 respect to traffic pollution and respiratory health in children, it does not provide evidence that
7 NO2 is responsible for these deficits in lung growth.
8 Further evaluation of exposure estimation was done in this cohort of schoolchildren
9 (Molitor et al., 2007). Several models of interurban air pollution exposure were used to classify
10 and predict FVC in an integrated Bayesian modeling framework using three interurban
11 predictors: distance to a freeway, traffic density, and predicted average NO2 exposure from the
12 California line source dispersion (CALINE4) model. Results suggested that the inclusion of
13 residual spatial terms can reduce uncertainty in the prediction of exposures and associated health
14 effects.
15 In Mexico City, Rojas-Martinez et al. (2007a,b) evaluated the association between long-
16 term exposure to PMio, Os, and NO2 and lung function growth in a cohort of 3,170 children aged
17 8 years at baseline in 31 schools from April 1996 through May 1999. Ten air-quality monitoring
18 stations within 2 km of the schools provided exposure data. Figure 3.4-4 shows the results for
19 FEVi, by gender and pollutant with adjustments noted for copollutants. The results of this
20 3-year study support the hypothesis that long-term exposure to ambient air pollutants is
21 associated with deficit in lung growth in children. The results are, in part, consistent with
22 previous results from the CHS. Similar to the CHS, the high correlation among the three
23 pollutants studied did not allow independent effects to be accurately estimated in this long-term
24 exposure study.
25 Another cohort study in Oslo, Norway, examined short- and long-term NO2 and other
26 pollutant exposure effects on lung function (PEF, forced expiratory flow at 25% of forced vital
27 capacity [FEF25], forced expiratory flow at 50% of forced vital capacity [FEF50]) in 2,307 nine-
28 and ten-year-old children (Oftedal et al., 2008). The EPISODE dispersion model (Slordal et al.,
29 2003) was used for the exposure estimate and evaluation concluded that the modeled NO2 and
30 PM levels represent the long- and short-term exposure reasonably well. An incremental change
31 equal to the IQR of lifetime exposure to NO2, PMio, and PM2 5 was associated with changes in
March 2008 3-86 DRAFT-DO NOT QUOTE OR CITE
-------
O, (Girls)
3,2
2,9
6-monlh mean concentrations
64.3,69.3, 75.7 ppb
56.42,67.63,92.22 (jg/m3
28.92, 34.57 40.85 ppb
- « p25
-• pSO
• A - - - p75
3,2
2,9
o> 2,6
03 (Boys)
P25
pSO
p75
3 4
Phase
PM10 (Girls)
567
Adjust with PMIO and NO2
4
Phase
Adjust with PM10 and NO2
PM10 (Boys)
Adjust with O3 and NO2
Adjust with O3 and NO2
NO, (Girls)
NO, (Boys)
_ .»_ . P25
P50
*- - p75
Adjust with O3 and PM,0
4567
Phase
Adjust with O3 and PM)0
Figure 3.4-4. Estimated annual growth in FEVi, of long-term ozone (Os), particulate
matter ^10 jim in diameter (PMio), and nitrogen dioxide (NOi) in girls and
boys. Mexico City, 1996 to 1999 (multipollutant models). Adjusted for age,
body mass index, height, height by age, weekday time spent in outdoor
activities, environmental tobacco smoke exposure, pervious-day mean air
pollutant concentration, and study phase.
Source: Derived from Rojas-Martinez et al. (2007a,b).
March 2008
3-87
DRAFT-DO NOT QUOTE OR CITE
-------
1 adjusted peak flow of-79 mL/s (95% CI: -128,-31),-66 mL/s (95% CI: -110,-23), and
2 -58 mL/s (95% CI: -94, -21), respectively. Examining short- and long-term NC>2 exposures
3 simultaneously yielded only the long-term effects. Adjusting for a contextual socioeconomic
4 factor diminished the association. Comparable PEF to the CHS were found but forced volumes
5 were considerably weaker.
6 In another European study, Moseler et al. (1994) measured NO2 outside the homes of 467
7 children, including 106 who had physician-diagnosed asthma, in Freiburg, Germany. Five of six
8 lung function parameters were reduced among asthmatic children exposed to NO2 at
9 concentrations of >21 ppb. No significant reductions in lung function were detected among
10 children without asthma.
11 To examine the effect of lifetime exposure to air pollutants in young adults, lung function
12 in students attending the University of California (Berkeley) who had been lifelong residents of
13 the Los Angeles or San Francisco areas was assessed (Tager et al, 2005). Using geocoded
14 address histories, a lifetime exposure to air pollution was constructed for each student.
15 Increasing lifetime exposure to NC>2 was associated with decreased FEF75 and FEF25-75. In
16 models including Os and PMio as well as NC>2, the effect of NC>2 diminished significantly while
17 the 63 effect remained robust.
18 The SAPALDIA (Study of Air Pollution and Lung Diseases in Adults) study
19 (Ackermann-Liebrich et al., 1997) compared 9,651 adults (age 18 to 60) in eight different
20 regions in Switzerland. Significant associations of NC>2, SC>2, and PMio with FEVi and FVC
21 were found with a 10-|ig/m3 (5.2 ppb) increase in annual average exposure. Due to the high
22 correlations between NO2 and the other pollutants (862: r = 0.86; PMi0: r = 0.91), it was
23 difficult to assess the effect of a specific pollutant. A random subsample of 560 adults from
24 SAPALDIA recorded personal measurements of NC>2 and measurements of NC>2 outside their
25 homes (Schindler et al., 1998). Using the personal and home measurements of NC>2, similar
26 associations were reported between NC>2 with FEVi and FVC. Downs et al. (2007) reported the
27 relationship in this group of long-term reduced exposure to PMio and age-related decline in lung
28 function, but they did not examine NC>2 or other pollutants.
29 Goss et al. (2004) examined the relationship of ambient pollutants on individuals with
30 cystic fibrosis using the Cystic Fibrosis Foundation National Patient Registry in 1999 and 2000.
31 Exposure was assessed by linking air pollution values from the Aerometric Information Retrieval
March 2008 3-88 DRAFT-DO NOT QUOTE OR CITE
-------
1 System with the patient's home ZIP code. Associations were reported between PM and
2 exacerbations or lung function changes, but no clear associations were found for 63, SO2, NO2,
3 or CO. The odds of patients with cystic fibrosis having two or more pulmonary exacerbations
4 during 2000 per 10-ppb NO2 is 0.98 (95% CI: 0.91, 1.01).
5
6 Toxicological Studies
1 A limited number of animal studies, especially those using spikes of NO2, have shown
8 decrements in vital capacity and lung distensibility, which may provide biological plausibility for
9 these lung function findings. NO2 concentrations in many urban areas of the United States and
10 elsewhere consist of spikes superimposed on a relatively constant background level. As
11 discussed in the 1993 AQCD, Miller et al. (1987) evaluated this urban pattern of NO2 exposure
12 in mice using continuous 7-days/week, 23-h/day exposures to 0.2 ppm NO2 with twice daily
13 (5 days/week) 1-h spike exposures to 0.8-ppm NO2 for 32 and 52 weeks. Mice exposed to clean
14 air and to the constant background concentration of 0.2-ppm NO2 served as controls. Vital
15 capacity tended to be lower (p = 0.054) in mice exposed to NO2 with diurnal spikes than in mice
16 exposed to air. Lung distensibility, measured as respiratory system compliance, also tended to
17 be lower in mice exposed to diurnal spikes of NO2 compared with constant NO2 exposure or air
18 exposure. These changes suggest that <52 weeks of low-level NO2 exposure with diurnal spikes
19 may produce a subtle decrease in lung distensibility, although part of this loss in compliance may
20 be a reflection of the reduced vital capacity. Vital capacity appeared to remain suppressed for at
21 least 30 days after exposure. Lung morphology in these mice was evaluated only by light
22 microscopy (a relatively insensitive method) and showed no exposure-related lesions. The
23 decrease in lung distensibility suggested by this study is consistent with the thickening of
24 collagen fibrils in monkeys (Bils, 1976) and the increase in lung collagen synthesis rates of rats
25 (Last et al., 1983) after exposure to higher levels of NO2.
26 Tepper et al. (1993) exposed rats to 0.5-ppm NO2, 22 h/day, 7 days/week, with a 2-h
27 spike of 1.5-ppm NO2, 5 days/week for up to 78 weeks. No effects on pulmonary function were
28 observed between 1 and 52 weeks of exposure. However, after 78 weeks of exposure, flow at
29 25% FVC was decreased, perhaps indicating airways obstruction. A significant decrease in the
30 frequency of breathing was also observed at 78 weeks that was paralleled by a trend toward
March 2008 3-89 DRAFT-DO NOT QUOTE OR CITE
-------
1 increased expiratory resistance and expiratory time. Taken together, these results suggest that
2 few, if any, significant effects were seen that suggest incipient lung degeneration.
3 There were no effects on pulmonary function (lung resistance, dynamic compliance) in
4 NO2-exposed rabbits that were immunized intraperitoneally within 24-h of birth until 3 months
5 of age to either Alternaria tennis or house dust mite antigen. The rabbits were given
6 intraperitoneal injections once weekly for 1 month, and then every 2 weeks thereafter, and
7 exposed to 4-ppm NC>2 for 2 h daily (Douglas et al., 1994).
8 A number of epidemiologic studies examined the effects of long-term exposure to NO2
9 and observed associations with decrements in lung function and partially irreversible decrements
10 in lung function growth. Results from the Southern California Children's Health Study indicated
11 that decrements were similar for boys compared to girls, and among children who did not have a
12 history of asthma (Gauderman et al., 2004). As shown in Appendix Table 5B, the mean NC>2
13 concentrations in these studies range from 21.5 to 34.6 ppb; thus, all have been conducted in
14 areas where NC>2 levels are below the level of the NAAQS. The epidemiologic studies of long-
15 term exposure to NC>2, however, are likely confounded by other ambient copollutants. In
16 particular, similar associations have also been found for PM and proximity to traffic (<500 m).
17
18 3.4.2 Asthma Prevalence and Incidence
19 Several publications from the CHS in southern California report results on the
20 associations of NC>2 exposure with asthma prevalence and incidence. Gauderman et al. (2005)
21 conducted a study of children randomly selected from the CHS with exposure measured at
22 children's homes. Although only 208 were enrolled, exposure to NC>2 was strongly associated
23 with both lifetime history of asthma and asthma medications use. Gauderman et al. (2005)
24 measured ambient NC>2 with Palmes tubes attached to the subjects' homes at the roofline eaves,
25 signposts, or rain gutters at an approximate height of 2 m above the ground. Samplers were
26 deployed for 2-week periods in both summer and fall. Traffic-related pollutants were
27 characterized by three metrics: (1) proximity of home to freeway, (2) average number of
28 vehicles within 150 meters, and (3) model-based estimates. Yearly average NC>2 levels within
29 the 10 communities ranged from 12.9 to 51.5 ppb. The average NC>2 concentration measured at
30 home was associated with asthma prevalence (OR = 8.33 [95% CI: 1.15, 59.87] per 20 ppb)
31 with similar results by season and when taking into account several potential confounders. In
March 2008 3-90 DRAFT-DO NOT QUOTE OR CITE
-------
1 each community studied, NO2 was more strongly correlated with estimates of freeway-related
2 pollution than with non-freeway-related pollution. In a related CHS study, McConnell et al.
3 (2006) studied the relationship of proximity to major roads and asthma and also found a positive
4 relationship.
5 Islam et al. (2007) studied whether lung function is associated with new onset asthma and
6 whether this relationship varies by exposure to ambient air pollutants by examining a cohort of
7 2,057 fourth-grade children who were asthma- and wheeze-free at the start of the CHS and
8 following them for 8 years. A hierarchal model was used to evaluate the effect of individual air
9 pollutants (NO2, PMio, PM2.5, and acid vapor, NO2, EC, and OC) on the association of lung
10 function with asthma. This study shows that better airflow, characterized by higher FEF25_75 and
11 FEVi during childhood was associated with decreased risk of new-onset asthma during
12 adolescence. However, exposure to high levels of ambient pollutants (NO2 and others)
13 attenuated this protective association of lung function on asthma occurrence.
14 Millstein et al. (2004) studied the effects of ambient air pollutants on asthma medication
15 use and wheezing among 2,034 fourth-grade schoolchildren from the CHS. Included in the
16 pollutants examined were NO2 and HNOs. They observed that monthly average pollutant levels
17 produced primarily by photochemistry (i.e., HNOs, acetic acid), but not NO2, were suggestive of
18 a positive association with asthma medication use among children with asthma—especially
19 among children who spent more than the calculated median time outdoors. The March-August
20 ORforHNO3(IQR1.64ppb)was 1.62 (95% CI: 0.94, 2.80) and for NO2 (IQR 5.74 ppb), 0.96
21 (95% CI: 0.68, 1.37).
22 Kim et al. (2004a) reported associations with both NO2 and NOx for girls in the San
23 Francisco bay area. They studied 1,109 students (grades 3 to 5) at 10 school sites for bronchitis
24 symptoms and asthma in relation to ambient pollutant levels to include NO, NO2, and NOx
25 measured at the school site. Mean levels ranged for schools from 33 to 69 ppb for NOx; 19 to 31
26 for NO2; and 11 to 38 ppb for NO. NOx and NO2 measurements at school sites away from
27 traffic were similar to levels measured at the regional site. They found associations between
28 traffic-related pollutants and asthma and bronchitis symptoms, which is consistent with previous
29 reports of traffic and respiratory outcomes. The higher effect estimates with black carbon, NOx,
30 and NO compared with NO2 and PM2 5 suggest that primary or fresh traffic emissions may play
March 2008 3-91 DRAFT-DO NOT QUOTE OR CITE
-------
1 an etiologic role in these relationships and that, while NOx and NO may serve as indicators of
2 traffic exposures, they also may act as etiologic agents themselves.
3 Brauer et al. (2007) assessed the development of asthmatic/allergic symptoms and
4 respiratory infections during the first 4 years of life in a birth cohort study in the Netherlands
5 (n = 4,000, but the number of participants decreased over the study to -3,500). Air pollution
6 concentrations at the home address at birth were calculated by a validated model combining air
7 pollution measurements with a Geographic Information System (GIS). Wheeze, physician-
8 diagnosed asthma, and flu and serious colds were associated with air pollutants (considered
9 traffic-related: NO2, PM2.5, soot) after adjusting for other potential confounding variables; for
10 example, NO2 was associated with physician-diagnosed asthma (OR = 1.28 [95% CI: 1.04,
11 1.56]) as a cumulative lifetime indicator. In comments to this study, Jerrett (2007) observed that
12 the effects were larger and more consistent than in participants of the same study at age 2 and
13 that these effects suggested that onset and persistence of respiratory disease formation begins at
14 an early age and continues. He further noted that the more sophisticated method for exposure
15 assessment based on spatially and temporally representative field measurements and land use
16 regression was capable of capturing small area variations in traffic pollutants.
17 Other studies (see Annex Table AX6.3-16) also have investigated asthma prevalence and
18 incidence in children associated with NO2 exposure. Although several of these studies have
19 reported positive associations, the large number of comparisons made and the limited number of
20 positive results do not suggest a strong relationship between long-term NO2 exposure and
21 asthma. Several studies used the International Study of Asthma and Allergies in Children
22 (ISAAC) protocol. Children were interviewed in school and results of the questionnaire were
23 compared with air pollution measurements in their communities. These studies included
24 thousands of children in several European countries and Taiwan, and the results in all but one
25 study were nonsignificant. Exposure in these studies varied, but medians were often greater than
26 20 ppb. Most of the studies did not report correlations of NO2 exposure with other air pollutants;
27 therefore, it is not possible to determine whether some of these associations were related to other
28 air contaminants.
29 Overall, results from the available epidemiologic evidence investigating the association
30 between long-term exposure to NO2 and increases in asthma prevalence and incidence are
31 inconsistent. Two major cohort studies, the Children's Health Study in southern California
March 2008 3-92 DRAFT-DO NOT QUOTE OR CITE
-------
1 (Gauderman et al., 2005) and a birth cohort study in the Netherlands (Brauer et al., 2007)
2 observed significant associations; however, several other studies did not find consistent
3 associations between long-term NO2 exposure and asthma outcomes.
4
5 3.4.3 Respiratory Symptoms
6 Annex Table AX6.3-17 lists studies examining the association between long-term
7 exposure to NO2 and respiratory symptoms. Most of the studies reported some positive
8 associations with NO2 exposure and symptoms, but all reported a large number of negative
9 results. Only one of these studies (Peters et al., 1999) reported an association of NO2 exposure
10 with wheeze, and in boys. This was despite the fact that wheeze was investigated in a large
11 number of studies, including several studies that included thousands of children.
12 McConnell et al. (2003) studied the relationship between bronchitis symptoms and air
13 pollutants in the CHS. Symptoms assessed yearly by questionnaire from 1996 to 1999 were
14 associated with the yearly variability for the pollutants for NO2 (OR = 1.071 [95% CI: 1.02,
15 1.13). In two-pollutant models, the effects of yearly variation in NO2 were only modestly
16 reduced by adjusting for other pollutants except for OC and NO2 (Figure 3.4-5). McConnell
17 et al. (2006) further evaluated whether the association of exposure to air pollution with annual
18 prevalence of chronic cough, phlegm production, or bronchitis was modified by dog or cat
19 ownership indicators or allergen and endotoxin exposure. Subjects consisted of 475 children
20 from the CHS. Among children owning a dog, there was a strong association between bronchitis
21 symptoms and all pollutants studied. Odds ratio for NO2 were 1.49 (95% CI: 1.14, 1.95),
22 indicating that dog ownership may worsen the relationship between air pollution and respiratory
23 symptoms in asthmatic children.
24 Two studies of infants were conducted in Germany and the Netherlands using the same
25 exposure protocol (Gehring et al., 2002; Brauer et al., 2002). In Munich, 1,756 infants were
26 enrolled and followed for 2 years. Outcomes of interest were asthma, bronchitis, and respiratory
27 symptoms including wheeze, cough, and nasal symptoms. To determine exposure, 40 measuring
28 sites were selected in Munich, including sites along main roads and side streets and background
29 sites. At each site, NO2 was measured four times (once in each season) for 14 days using Palmes
30 tubes. Regression modeling was used to relate annual average pollutant concentrations to a set
March 2008 3-93 DRAFT-DO NOT QUOTE OR CITE
-------
Risk of Bronchitic Symptoms as a Function
of Yearly Deviation in NO2
i.o -
1 4 -
£
Q.
^ 1.3 -
*
.2 1.2-
« 1-1 -
TJ
o 1.0
no.
I_L_L_L_LJ
i * ? i
Adjustment Air Pollutants
Figure 3.4-5. Odds ratios for within-community bronchitis symptoms associations with
NO2, adjusted for other pollutants in two-pollutant models for the 12
communities of the Children's Health Study.
Source: McConnell et al. (2003).
1 of predictor variables (i.e., traffic density, heavy vehicle density, household density, population
2 density) obtained from GIS. The percentage of variability explained by the model (R2) was
3 0.62 for NO2. Using geocoded birth addresses, values for the predictor variables were obtained
4 for each child, and the model was used to assign an estimate of NO2 exposure. At 1 year of age,
5 an increase of 8.5 |ig/m3 (4.5 ppb) of NO2 was associated with cough (OR = 1.40 [95% CI: 1.12,
6 1.75]) and dry cough at night (OR =1.36 [95% CI: 1.07, 1.74]). NO2 exposure was not
7 associated with wheeze, bronchitis, or respiratory infections. Estimated PM2 5 exposure was also
8 associated with cough and dry cough at night, with nearly identical odds ratios.
9 In the Netherlands (Brauer et al., 2002), the same protocol was used to estimate NO2
10 exposure in a birth cohort of 3,730 infants. However, these study subjects lived in many
11 different communities from rural areas to large cities in northern, central, and western parts of
12 the Netherlands. Forty sites were selected to represent different exposures and measurements
13 were taken as in the Gehring et al. (2002) study. In this study, ear, nose, and throat infections
14 (OR =1.16 [95% CI: 1.00, 1.34]) and physician-diagnosed flu (OR = 1.11 [95% CI: 1.00,
March 2008 3-94 DRAFT-DO NOT QUOTE OR CITE
-------
1 1.23]) were marginally significant. The association of NC>2 with dry cough at night could not be
2 replicated, nor was NO2 associated with asthma, wheeze, bronchitis, or eczema.
3 In both of these studies, the 40 monitoring sites set up to measure NO2 also measured
4 PM2.5 with Harvard Impactors. Estimates of NC>2 and PM2.5 were highly correlated in Brauer
5 et al. (r = 0.97). The correlation was not reported in Gehring et al. (2002); however, the
6 similarity of odds ratios for each pollutant suggests that the estimated exposures were also highly
7 correlated. Thus, a major limitation of these studies is the inability to distinguish the effects of
8 different pollutants.
9 In a study of 3,946 Munich schoolchildren, Nicolai et al. (2003) assessed traffic exposure
10 using two different methods. First, all street segments within 50 m of each child's home were
11 identified and the average daily traffic counts were totaled. Second, a model was constructed
12 based on measurement of NC>2 at 34 sites throughout the city using traffic counts and street
13 characteristics (R2 = 0.77). The model was then used to estimate NC>2 exposure at each child's
14 home address. When traffic counts of < 50m were used as an exposure variable, a significant
15 association was found with current asthma (OR = 1.79 [95% CI: 1.05, 3.05]), wheeze
16 (OR = 1.66 [95% CI: 1.07, 2.57]), and cough (OR = 1.62 [95% CI: 1.16, 2.27]). Similar results
17 were found when modeled NO2 exposure was substituted as the exposure variable (current
18 asthma OR = 1.65 [95% CI: 0.94, 2.90], wheeze OR = 1.58 [95% CI: 1.05, 2.48], cough
19 OR= 1.60 [95% CI: 1.14,2.23]). Asthma, wheeze, and cough were also associated with
20 estimated exposures to soot and benzene derived from models, suggesting that some component
21 of traffic pollution is increasing risk of respiratory conditions in children, but making it difficult
22 to determine whether NO2 is the cause of these conditions.
23 In summary, epidemiologic studies conducted in both the United States and Europe have
24 observed inconsistent results regarding an association between long-term exposure to NO2 and
25 respiratory symptoms. While some positive associations were noted, a large number of symptom
26 outcomes were examined and the results across specific outcomes were inconsistent.
27
28 3.4.4 Animal Studies of Long-Term Morphological Effects to the
29 Respiratory System
30 Animal toxicology studies demonstrate morphological changes to the respiratory tract
31 from exposure to NO2 that may provide further biological plausibility for the decrements in lung
32 function growth observed in epidemiologic studies discussed above. Several investigators have
March 2008 3-95 DRAFT-DO NOT QUOTE OR CITE
-------
1 studied the temporal progression of early events due to NO2 exposure in the rat (e.g., Freeman
2 et al., 1966, 1968, 1972; Stephens et al., 1971, 1972; Evans et al., 1972, 1973a,b, 1974, 1975,
3 1976, 1977; Cabral-Anderson et al., 1977; Rombout et al., 1986) and guinea-pig (Sherwin and
4 Carlson, 1973). The results of these studies were summarized in the 1993 AQCD. Overall,
5 animal toxicological studies demonstrated that NO2 exposure resulted in permanent alterations
6 resembling emphysema-like disease, morphological changes in the centriacinar region of the
7 lung and in bronchiolar epithelial proliferation, which might provide biological plausibility for
8 the observed epidemiologic associations between long-term exposure to NO2 and respiratory
9 morbidity.
10
11 3.4.5 Summary and Integration of Evidence on Long-Term NOi Exposure
12 and Respiratory Illness and Lung Function Decrements
13 Overall, the epidemiologic and experimental evidence is suggestive but not sufficient to
14 infer a causal relationship between long-term NO2 exposure and respiratory morbidity. The
15 available database evaluating the relationship between respiratory illness in children associated
16 with long-term exposures to NO2 has increased. Three recent studies in large cohorts in three
17 countries have examined this relationship. The CHS, examining NO2 exposure in children over
18 an 8-year period, demonstrated deficits in lung function growth (Gauderman et al., 2004). This
19 has been observed also in Mexico City, Mexico (Rojas-Martinez et al., 2007a,b), and in Oslo,
20 Norway (Oftedal et al., 2008).
21 Deficit in lung function growth is a known risk factor for chronic respiratory disease and
22 possibly for premature mortality in later life stages. Lung growth continues from early
23 development through early adulthood, reaches a plateau, and then eventually declines with
24 advancing age. Dockery and Brunekreef (1996) have hypothesized that the risk for chronic
25 respiratory disease is associated with maximum lung size, the length of time the lung size has
26 been at the plateau, and the rate of decline of lung function. Therefore, exposures to NO2 and
27 other air pollutants in childhood may reduce maximum lung size by limiting lung growth and
28 subsequently increase the risk in adulthood for chronic respiratory disease.
29 Models and/or mechanisms of action for decrements in lung function growth and other
30 respiratory effects from long-term exposure to air pollution are not clearly established. Figure
31 3.4-6 is adapted from an earlier model discussed by Gilliland et al. (1999), reflective of efforts of
32 the CHS research. Gilliland et al. proposed that respiratory effects in children from exposure to
March 2008 3-96 DRAFT-DO NOT QUOTE OR CITE
-------
Ambient NO,
Dietary
antioxidants
Antioxidant
enzymes
Total
personal
exposure
to NO,
Total
personal
dose
Oxidative/radical
damage
*/ V
Dietary PUFA Physical activity
MPO
Asthma
AtopyTNFa
Neutrophilic
inflammation
J, Lung function
growth
T Asthma
Indoor NO,
Tissue
damage
Figure 3.4-6. Biologic pathways of long-term NOi exposure on morbidity.
MPO = myeloperoxidase; PUFA = polyunsaturated fatty acids; TNF-a = tumor necrosis factor-alpha.
Source: Adapted from Gilliland et al. (1999).
1 gaseous and particulate air pollutants result from chronically increased oxidative stress,
2 alterations in immune regulation, and repeated pathologic inflammatory responses that overcome
3 lung defenses to disrupt the normal regulatory and repair processes. Rojas-Martinez et al.
4 (2007a,b) noted that oxidative stress resulting from increased exposure to oxidized compounds
5 (63, NO2, and particle components) has been identified as a major feature underlying the toxic
6 effects of air pollutants (Kelly et al., 2003; Saxon and Diaz-Sanchez, 2005; Cross et al., 2002).
7 They further noted that the resulting increased expression of enhanced proinflammatory
8 cytokines leads to enhanced inflammatory response (Saxon and Diaz-Sanchez, 2005) and
9 potential chronic lung damage. If this results in permanent loss, it is not clear whether repeated
10 versus average exposure is the major factor. Current data and the nonlinear pattern of childhood
11 lung function growth (Perez-Padilla et al., 2003) are noted by Rojas-Martinez et al. (2007a,b) as
12 limitations on estimating the impact on lung function attained in early adulthood.
13 Other important biochemical mechanisms examined in animals may provide biological
14 plausibility for the chronic effects of NO2 observed in epidemiologic studies. The main
15 biochemical targets of NO2 exposure appear to be antioxidants, membrane polyunsaturated fatty
16 acids, and thiol groups. Reactions of NO2 with these species in the extracellular lining fluid of
March 2008
3-97
DRAFT-DO NOT QUOTE OR CITE
-------
1 the lung leads to the formation of nitrite (NO2 ) and hydrogen (H+) ions. NO2 effects include
2 changes in oxidant/antioxidant homeostasis and chemical alterations of lipids and proteins.
3 Lipid peroxidation has been observed at NO2 exposures as low as 0.04 ppm for 9 months and at
4 exposures of 1.2 ppm for 1 week, suggesting lower effect thresholds with longer durations of
5 exposure. Other studies show decreases in formation of key arachidonic acid metabolites in
6 AMs following NO2 exposures of 0.5 ppm. NO2 has been shown to increase collagen synthesis
7 rates at concentrations of as low as 0.5 ppm. This could indicate increased total lung collagen,
8 which is associated with pulmonary fibrosis, or increased collagen turnover, which is associated
9 with remodeling of lung connective tissue. Morphological effects following chronic NO2
10 exposures have been identified in animal studies that link to these increases in collagen synthesis
11 and may provide plausibility for the deficits in lung function growth described in epidemiologic
12 studies.
13 An alternative explanation for the decrease in lung function growth observed in the CHS
14 needs to be considered. Since this response was associated with both NO2 and HNOs exposure,
15 ambient levels of NO may also have been involved. Three groups have reported emphysematous
16 changes in animal studies following prolonged exposure to NO. In the Mercer study (1995), a
17 decreased number of interstitial cells and thinning of the alveolar septa was observed. Other
18 studies in vitro and in animal models have demonstrated that NO inhibits protein synthesis and
19 cellular proliferation. Whether NO plays a role in maintaining the alveolar interstitial
20 compartment requires further investigation. Furthermore, the formation of NO or NO-related
21 species may have occurred following complex reactions of NO2 and HNOs with components of
22 the extracellular lining fluid. The role of MV, H+, NO and other metabolites in modulating
23 responses to NO2 and/or HNOs is unknown.
24 In regard to asthma prevalence incidence associated with NO2 long-term exposure, two
25 major cohorts, the CHS in southern California and birth cohort in the Netherlands, and several
26 other studies provide the evidence for this outcome. Again, the studies are well designed and
27 implemented. However, these results are not consistent with a number of other studies that have
28 investigated this relationship.
29 Animal toxicologic studies provide biological plausibility for the observed increased
30 incidence of respiratory illness among children. A number of defense system components such
31 as AMs and humoral and cell-mediated immunity have been demonstrated to be targets for
March 2008 3-98 DRAFT-DO NOT QUOTE OR CITE
-------
1 inhaled NC>2. The animal studies described above show that NO2 exposure impairs the host
2 defense system, increasing susceptibility to respiratory infections. Morphological changes are
3 elicited in ciliated epithelial cells at NO2 concentrations of as low as 0.5 ppm for 7 months;
4 however, early studies showed that mucocilary clearance, the first line of defense, is not affected
5 by exposures of <5 ppm. A more recent study in guinea pigs showed a concentration-dependent
6 decrease in ciliary activity at 3-ppm NC>2. The AMs, a second line of defense in the lung, are
7 affected by NC>2 in a concentration- and species-dependent manner with both acute and chronic
8 exposures. Mechanisms whereby NC>2 affects AM function include membrane lipid
9 peroxidation, decreased ability to produce superoxide anion, inhibition of migration, and
10 decreased phagocytic activity. Decreases in bactericidal and phagocytic activities are likely
11 related to increased susceptibility to pulmonary infections. More recent studies have confirmed
12 that AMs are a primary target for NC>2 at exposure levels of <1 ppm. Humoral and cell-mediated
13 immunity form a third line of defense that has been shown to be suppressed by NC>2 exposure.
14 The use of animal infectivity studies provides key evidence for the effects of NC>2 on respiratory
15 morbidity and mortality. For these studies, the animals are exposed to NC>2, and subsequently to
16 an aerosol containing the infectious agent. This body of work shows that NC>2 decreases
17 intrapulmonary bactericidal activity in mice in a concentration-dependent manner, with no
18 concurrent changes in mucociliary clearance.
19 Thus, evidence indicates that the reduced efficacy of lung defense systems may be an
20 important mechanism for the observed increase in incidence and severity of respiratory
21 infections. Overall, the NC>2 toxicologic literature suggests a linear concentration-response
22 relationship that exists in an exposure range of 0.5 to >5ppm and mortality resulting from
23 pulmonary infection. NC>2 exposure reduces the efficiency of defense against infections at
24 concentrations of as low as 0.5 ppm. The exposure protocol is important, with concentration
25 being more important than duration of exposure and with peak exposures being important in the
26 overall response. The effect of concentration is stronger with intermittent exposure than with
27 continuous exposure. Repeated exposures of low levels of NC>2 are necessary for many
28 respiratory effects. The animal toxicologic studies also demonstrate differences in species
29 sensitivity to NC>2 and differences in responses to the microbes used for the infectivity tests.
30 Animal to human extrapolation is limited by a poor understanding of the quantitative relationship
31 between NC>2 concentrations and effective doses between animals and humans. However,
March 2008 3-99 DRAFT-DO NOT QUOTE OR CITE
-------
1 animals and humans share many host defense components, making the infectivity model useful
2 for understanding the mechanisms whereby NC>2 elicits adverse respiratory health effects.
3
4
5 3.5 OTHER MORBIDITY EFFECTS ASSOCIATED WITH
6 LONG-TERM NO2 EXPOSURE
7 The current review includes a number of studies published since 1993 characterizing the
8 effect of long-term NOx exposure on cancer, CVD, reproductive, and developmental morbidity.
9 These studies form a new body of literature that was unavailable in 1993, when the previous
10 AQCD was published.
11
12 3.5.1 Cancer Incidence Associated with Long-Term NO2 Exposure
13 Two studies (see Annex Table AX6.3-18) have investigated the relationship between
14 NC>2 exposure and lung cancer and reported positive associations. Although this literature
15 review has concentrated on studies that measured exposure to NO2, modeled exposures will be
16 considered for cancer studies. This is necessary because the relevant exposure period for lung
17 cancer may be 30 years or more.
18 Nyberg et al. (2000) reported results of a case control study of 1,043 men age 40 to
19 75 years with lung cancer and 2,364 controls in Stockholm County. They mapped residence
20 addresses to a GIS database indicating 4,300 traffic-related line sources and 500 point sources of
21 NC>2 exposure. Exposure was derived from a model validated by comparison to actual
22 measurements of NC>2 at six sites. Exposure to NC>2 at 10 |ig/m3 (5.2 ppb) was associated with
23 an OR of 1.10 (95% CI: 0.97, 1.23). Exposure to the 90th percentile (>29.26 |ig/m3
24 [15.32 ppb]) of NO2 was associated with an OR of 1.44 (95% CI: 1.05, 1.99).
25 Very similar results were reported in a Norwegian study (Nafstad et al., 2003). The study
26 population is a cohort of 16,209 men who enrolled in a study of CVD in 1972. The Norwegian
27 cancer registry identified 422 incident cases of lung cancer. Exposure data was modeled based
28 on residence, estimating exposure for each person in each year from 1974 to 1998. Each
29 10 |ig/m3 (5.2 ppb) of NO2 was associated with an OR of 1.08 (95% CI: 1.02, 1.15). Cancer
30 incidence with exposure of >30 |ig/m3 (15.7 ppb) was associated with an OR of 1.36 (95% CI:
31 1.01, 1.83); however, controlling for SO2 exposure did appreciably change the effect estimates
32 forNO2.
March 2008 3-100 DRAFT-DO NOT QUOTE OR CITE
-------
1 What is particularly striking in these two studies is the similarity in the estimate of effect.
2 Despite the fact that these two studies were conducted by different investigators, in different
3 countries, using different study designs and different methods for modeling exposure, the odds
4 ratios and confidence intervals for exposure per 10 |ig/m3 (5.2 ppb) and above 30 |ig/m3
5 (15.7 ppb) are virtually identical.
6
7 Animal and In Vitro Carcinogenicity and Genotoxicity Studies
8 There is no clear evidence that NO2 or gaseous nitrogen oxides act as a complete
9 carcinogen. No studies were found on NC>2 using classical carcinogenesis whole-animal
10 bioassays. Of the existing studies that have evaluated the carcinogenic and cocarcinogenic
11 potential of NO2, results are often unclear or conflicting. Witschi (1988) critically reviewed
12 some of the important theoretical issues in interpreting these types of studies. NC>2 does appear
13 to act as a tumor promoter at the site of contact (i.e., in the respiratory tract from inhalation
14 exposure), possibly due to its ability to produce cellular damage and, thus, promote regenerative
15 cell proliferation. This hypothesis is supported by observed hyperplasia of the lung epithelium
16 from NC>2 exposure (see Lung Morphology section, U.S. Environmental Protection Agency,
17 1993), which is a common response to lung injury, and enhancement of endogenous retrovirus
18 expression (Roy-Burman et al., 1982). However, these findings were considered by EPA (1993)
19 to be inconclusive.
20 When studied using in vivo assays, no inductions of recessive lethal mutations were
21 observed in Drosophila exposed to NC>2 (Inoue et al., 1981; Victorin et al., 1990). NC>2 does not
22 increase chromosomal aberrations in lymphocytes and spermatocytes or micronuclei in bone
23 marrow cells (Gooch et al., 1977; Victorin et al., 1990). No increased stimulation of poly (ADP-
24 ribose) synthetase activity (an indicator of DNA repair, suggesting possible DNA damage) was
25 reported in AMs recovered from BAL of rats continuously exposed to 1.2-ppm NC>2 for 3 days
26 (Bermudez, 2001).
27 NC>2 has been shown to be positive when tested for genotoxicity in vitro assays. NC>2 is
28 mutagenic in bacteria and in plants. In cell cultures, three studies showed chromosomal
29 aberrations, sister chromatid exchanges (SCEs), and DNA single-strand breaks. However, a
30 fourth study (Isomura et al., 1984) concluded that NO, but not NO2, was mutagenic in hamster
31 cells (see Annex Tables AX4.11 A, 4.1 IB, and 4.11C).
32
March 2008 3-101 DRAFT-DO NOT QUOTE OR CITE
-------
1 Toxicological Studies of Coexposure with NO2 and Known Carcinogens
2 Rats were injected with N-bis (2-hydroxy-propyl) nitrosamine (BHPN) and continuously
3 exposed to 0.04-, 0.4-, or 4-ppm NO2 for 17 months. Although the data indicated 5 times as
4 many lung adenomas or adenocarcinomas in the rats injected with BHPN and exposed to 4-ppm
5 NO2 (5/40 compared to 1/10), the results failed to achieve statistical significance (Ichinose et al.,
6 1991). In a later study, Ichinose and Sagai (1992) reported increased lung tumors in rats injected
7 with BHPN, followed the next day by either clean air (0%), 0.05-ppm NO2 (8.3%), 0.05-ppm
8 NO2 + 0.4-ppm O3 (13.9%), or 0.4-ppm O3 + 1 mg/m3 H2SO4-aerosol (8.3%) for 13 months, and
9 then maintained for another 11 months until study termination. Exposure to NO2 was
10 continuous, while the exposures to O3 and H2SO4-aerosol were intermittent (exposure for
11 10 h/day). The increased lung tumors from combined exposure of NO2 and O3 were statistically
12 significant.
13 Ohyama et al. (1999) coexposed rats to diesel exhaust particle extract-coated carbon
14 black particles (DEPcCBP) once a week for 4 weeks by intratracheal instillation and to either 6-
15 ppm NO2, 4-ppm SO2, or 6-ppm NO2 + 4-ppm SO2 16 h/day for 8 months, and thereafter
16 exposed to clean air for 8 months. Alveolar adenomas were increased in animals exposed to
17 DEPcCBP and either NO2 and/or SO2 compared to animals in the DEPcCBP-only group and to
18 controls. The incidences of lung tumors for the NO2, SO2, and NO2 and/or SO2 groups were 6/24
19 (25%), 4/30 (13%), and 3/28 (11%), respectively. No alveolar adenomas were observed in
20 animals exposed to DEPcCBP alone or in the controls. Increased alveolar hyperplasia was
21 elevated in all groups compared to controls. In addition, DNA adducts, as determined by 32P
22 postlabelling, were observed in the animals exposed to both DEPcCBP and either NO2 and/or
23 SO2, but not in animals exposed to DEPcCBP alone or controls. The authors concluded that the
24 cellular damage induced by NO2 and/or SO2 may have resulted in increased cellular permeability
25 of the DEPcCBP particles into the cells.
26
27 Studies in Animals with Spontaneously High Tumor Rates
28 The frequency and incidence of spontaneously occurring pulmonary adenomas was
29 increased in strain A/J mice (with spontaneously high tumor rates) after exposure to 10-ppm NO2
30 for 6 h/day, 5 days/week for 6 months (Adkins et al., 1986). These small, but statistically
31 significant, increases were only detectable when the control response from nine groups (n = 400)
32 were pooled. Exposure to 1-and 5-ppm NO2 had no effect. In contrast, Richters and Damji
March 2008 3-102 DRAFT-DO NOT QUOTE OR CITE
-------
1 (1990) found that an intermittent exposure to 0.25-ppm NO2 for up to 26 weeks decreased the
2 progression of a spontaneous T cell lymphoma in AKR/cum mice and increased survival rates.
3 The investigators attributed this effect to an NO2-induced decrease in the proliferation of T
4 lymphocyte subpopulation in the spleen (especially T-helper/inducer CD+ lymphocytes) that
5 produces growth factors for the lymphoma. A study by Wagner et al. (1965) suggested that NO2
6 may accelerate the production of tumors in CAFl/Jax mice (a strain that has spontaneously high
7 pulmonary tumor rates) after continuous exposure to 5-ppm NO2. After 12 months of exposure,
8 7/10 mice in the exposed group had tumors, compared to 4/10 in the controls. No differences in
9 tumor production were observed after 14 and 16 months of exposure. A statistical evaluation of
10 the data was not presented.
11
12 Facilitation ofMetastases
13 Whether NO2 facilitates metastases has been the subject of several experiments by
14 Richters and Kuraitis (1981, 1983), Richters and Richters (1983), and Richters et al. (1985).
15 Mice were exposed to several concentrations and durations of NO2 and were injected
16 intravenously with a cultured-derived melanoma cell line (B16) after exposure, and subsequent
17 tumors in the lung were counted. Although some of the experiments showed an increased
18 number of lung tumors, statistical methods were inappropriate. Furthermore, the experimental
19 technique used in these studies probably did not evaluate metastases formation as the term is
20 generally understood, but more correctly, colonization of the lung by tumor cells.
21 Production ofN-Nitroso Compounds and other Nitro Derivatives
22 Because of evidence that NO2 could produce NO2 and NOs in the blood and the fact
23 that NO2 is known to react with amines to produce animal carcinogens (nitrosamines), the
24 possibility that NO2 could produce cancer via nitrosamine formation has been investigated. Iqbal
25 et al. (1980) were the first to demonstrate a linear time- and concentration-dependent relationship
26 between the amount of 7V-nitrosomorpholine (NMOR, an animal carcinogen) found in whole-
27 mouse homogenates after the mice were gavaged with 2 mg of morpholine (an exogenous amine
28 that is rapidly nitrosated) and exposure to 15- to 50-ppm NO2 for between 1 and 4 h. In a
29 follow-up study at more environmentally relevant exposures, Iqbal et al. (1981) used
30 dimethylamine (DMA), an amine that is slowly nitrosated to dimethylnitrosamine (DMN). They
31 reported a concentration-related increase in biosynthesis of DMN at NO2 concentrations of as
March 2008 3-103 DRAFT-DO NOT QUOTE OR CITE
-------
1 low as 0.1 ppm; however, the rate was significantly greater at concentrations above 10-ppm NC>2.
2 Increased length of exposure also increased DMN formation between 0.5 and 2 h, but synthesis
3 of DMN was less after 3 or 4 h of exposure than after 0.5 h.
4 Mirvish et al. (1981) concluded that the results of Iqbal et al. (1980) were technically
5 flawed, but they found that in vivo exposure to NO2 could produce a nitrosating agent (NSA)
6 that would nitrosate morpholine only when morpholine was added in vitro. Further experiments
7 showed that the NSA was localized in the skin (Mirvish et al., 1983) and that mouse skin
8 cholesterol was a likely NSA (Mirvish et al., 1986). It has also been reported that only very
9 lipid-soluble amines, which can penetrate the skin, would be available to the NSA. Compounds
10 such as morpholine, which are not lipid-soluble, could only react with NC>2 when painted directly
11 on the skin (Mirvish et al., 1988). Iqbal (1984), responding to the Mirvish et al. (1981)
12 criticisms, verified their earlier (Iqbal et al., 1980) studies.
13 The relative significance of NC>2 from NC>2 compared with other NC>2 sources such as
14 food, tobacco, and nitrate-reducing oral bacteria is uncertain. Nitrosamines have not been
15 detected in tissues of animals exposed by inhalation to NC>2 unless precursors to nitrosamines
16 and/or inhibitors of nitrosamine metabolism are coadministered. Rubenchik et al. (1995) could
17 not detect 7V-nitrosodimethylamine (NDMA) in tissues of mice exposed to 7.5- to 8.5-mg/m3
18 NC>2 for 1 h. NDMA was found in tissues, however, if mice were simultaneously given oral
19 doses of amidopyrine and 4-methylpyrazole, an inhibitor of NDMA metabolism. Nevertheless,
20 the main source of NC>2 in the body is endogenously formed, and food is also a contributing
21 source of nitrite (from nitrate conversion).
22
23 Summary of Evidence on the Effects of Long-Term NO2 Exposure on Cancer Incidence
24 In summary, two epidemiologic studies conducted in Europe showed an association
25 between long-term NC>2 exposure and incidence of cancer (Nyberg et al., 2000; Nafstad et al.,
26 2003); however, the animal toxicologic studies have provided no clear evidence that NC>2
27 directly acts as a carcinogen, though it does appear to act as a tumor promoter at the site of
28 contact (Section 3.5.1). There are no in vivo studies that suggest that NC>2 causes teratogenesis
29 or malignant tumors. Only very high exposure studies, i.e., levels not relevant to ambient NO2
30 levels, demonstrate increased chromosomal aberrations and mutations in vitro studies. A more
31 likely pathway for NC>2 involvement in cancer induction is through secondary formation of nitro-
March2008 3-104 DRAFT-DO NOT QUOTE OR CITE
-------
1 polycylic aromatic hydrocarbons (nitro-PAHs), as nitro-PAHs are known to be more mutagenic
2 than their parent compounds. The evidence for a causal relationship between NO2 and increased
3 cancer risk is inadequate to infer the presence or absence of a causal relationship at this time.
4 The information presented in this section is relevant to potential mechanisms by which
5 exposure to products formed by reaction of gaseous nitrogen oxides with organic compounds can
6 be carcinogenic. As discussed previously in Section 2.2, nitro-PAHs and other nitrated organic
7 compounds can be produced through reactions of NC>2 or NO with organic compounds in the
8 atmosphere. Nitro-PAHs are largely found on particles, and they can also be including in direct
9 emissions of particles, such as diesel exhaust particles. Effects of paniculate nitrogen
10 compounds have been considered in previous reviews of the PM NAAQS.
11 In addition, it is possible that the products of NO2 (NO2 and NOs~) could produce
12 carcinogens (e.g., N-nitrosomorpholine) from exposure from an environmentally occurring
13 precursor compound (e.g., morpholine) within the body. The studies do demonstrate that this is
14 a possible mechanism; however, it should be pointed out that (1) that these studies are limited to
15 a single precursor compound whereas humans would be exposed to multiple precursor
16 compounds thus producing an array of nitrosamines and other nitrated compounds. (2) The level
17 of nitrosamines per se produced in this fashion would be small compared to the nitrosamines that
18 come from cigarette smoke, smoked meats, and other food sources and from the atmospheric
19 transformation of products in the ambient air, (3) a wide array of nitrated products are produced
20 in the ambient air with a number of these products known to be carcinogens and/or mutagens.
21
22 3.5.2 Cardiovascular Effects Associated with Long-Term NOi Exposure
23 One epidemiologic study examined the association of cardiovascular effects with long-
24 term exposure to NO2. Miller et al. (2007) studied 65,893 postmenopausal women between the
25 ages of 50 and 79 years without previous CVD in 36 U.S. metropolitan areas from 1994 to 1998.
26 They examined the association between one or more fatal or nonfatal cardiovascular events and
27 the women's exposure to air pollutants. Subject's exposures to air pollution were estimated by
28 assigning the annual mean levels of air pollutants in 2000 measured at the monitor nearest the
29 residence based on its five-digit ZIP Code centroid, which resulted in a more spatially resolved
30 exposure estimate. A total of 1,816 women had one or more fatal or nonfatal cardiovascular
31 events, including 261 deaths from cardiovascular causes. The main focus of the study was
March 2008 3-105 DRAFT-DO NOT QUOTE OR CITE
-------
1 PM2.5, but the overall CVD events (but not results for death events only) using all the
2 copollutants (PMio, PMio-2.5, 862, NO2, CO, and 63) in both single- and multipollutant models
3 were presented. The results for the models only including subjects with non-missing exposure
4 data (n = 28,402 subjects resulting in 879 CVD events) are described here. In the single-
5 pollutant model results, PM2.5 showed the strongest associations with the CVD events by far
6 among the pollutants, followed by SC>2. NC>2 did not show any association with the overall CVD
7 events (heart rate [HR] = 0.98 [95% CI: 0.89, 1.08] per 10-ppb increase in the annual average).
8 In the multipollutant model, which included all the pollutants, the association of PM2.5 and 862
9 with overall CVD events became even stronger. NC>2 became negatively associated with the
10 overall CVD events (HR = 0.82 [95% CI: 0.70,0.95]). Correlations among these pollutants
11 were not described; therefore, it is not possible to estimate the extent of confounding among
12 these pollutants in these associations, but it is clear that PM2.5 was the best predictor of the CVD
13 events.
14 Limited toxicology data exist on the effect of NC>2 on the heart. Alterations in vagal
15 responses have been shown to occur in rats exposed to 10-ppm NC>2 for 24 h; however, exposure
16 to 0.4-ppm NC>2 for 4 weeks revealed no change (Tsubone and Suzuki, 1984). MVinduced
17 effects on cardiac performance are suggested by a significant reduction in the pressure of oxygen
18 in arterial blood (PaO2) in rats exposed to 4.0-ppm NC>2 for 3 months. When exposure was
19 decreased to 0.4-ppm NC>2 over the same exposure period, PaC>2 was not affected (Suzuki et al.,
20 1981). In addition, a reduction in HR has been shown in mice exposed to both 1.2- and 4.0-ppm
21 NO2 for 1 month (Suzuki et al., 1984). Whether these effects are the direct result of NO2
22 exposure or secondary responses to lung edema and changes in blood hemoglobin content is not
23 known (U.S. Environmental Protection Agency, 1993). A more recent study (Takano et al.,
24 2004) using an obese rat strain found changes in blood triglycerides, HDL, and HDL/total
25 cholesterol ratios with a 24-week exposure to 0.16-ppm NC>2.
26 No effects on hematocrit and hemoglobin have been reported in squirrel monkeys
27 exposed to 1.0-ppm NC>2 for 16 months (Fenters et al., 1973) or in dogs exposed to <5.0-ppm
28 NO2 for 18 months (Wagner et al., 1965). There were, however, polycythemia and an increased
29 ratio of PMNs to lymphocytes in rats exposed to 2.0 + 1.0 ppm NC>2 for 14 months (Furiosi et al.,
30 1973).
March 2008 3-106 DRAFT-DO NOT QUOTE OR CITE
-------
1 The few available epidemiologic and toxicological evidence do not suggest that long-
2 term exposure to NO2 has cardiovascular effects. The U.S. Women's Health Initiative study
3 (Miller et al., 2007) did not find any associations between long-term NO2 exposure and
4 cardiovascular events. The toxicological studies observed some effects of NO2 on cardiac
5 performance and heart rate, but only at exposure levels of as high as 4 ppm. Overall, these data
6 are inadequate to infer the presence or absence of a causal relationship.
7
8 3.5.3 Reproductive and Developmental Effects Associated with Long-Term
9 NO2 Exposure
10
11 Epidemiologic Studies
12 The effects of maternal exposure during pregnancy to air pollution have been examined
13 by several investigators in recent years (2000 through 2006). These outcomes were not
14 evaluated in the 1993 AQCD. The most common endpoints studied are low birth weight,
15 preterm delivery, and measures of intrauterine growth (e.g., small for gestational age [SGA]).
16 Generally, these studies have used routinely collected air pollution data and birth certificates
17 from a given area for their analysis.
18 While most studies analyzed average NO2 exposure for the whole pregnancy, many also
19 considered exposure during specific trimesters or other time periods. Fetal growth, for example,
20 is much more variable during the third trimester. Thus, studies of fetal growth might anticipate
21 that exposure during the third trimester would have the greatest likelihood of an association, as is
22 true for the effect of maternal smoking during pregnancy. However, growth can also be affected
23 through placentation, which occurs in the first trimester. Similarly, preterm delivery might be
24 expected to be related to exposure early in pregnancy affecting placentation, or through acute
25 effects occurring just before delivery.
26 Of the three studies conducted in the United States, one (Bell et al., 2007) reported a
27 significant decrease in birthweight associated with exposure to NO2 among mothers in
28 Connecticut and Massachusetts. The two studies conducted in California did not find
29 associations between NO2 exposure with any adverse birth outcome (Ritz et al., 2000; Salam
30 et al., 2005). Differences in these studies that may have contributed to the differences in results
31 include the following: sample size, exposure assessment methods, average NO2 concentration,
32 and different pollution mixtures. The results reported by Bell et al. (2007) had the largest sample
March 2008 3-107 DRAFT-DO NOT QUOTE OR CITE
-------
1 size and, therefore, greater power to assess small increases in risk. The two California studies
2 reported higher mean concentrations of NO, but also strong correlations of NO2 exposure with
3 PM mass and CO.
4 Annex Table AX6.3-12 lists seven studies that investigated the relationship of ambient
5 NO2 exposure with birth weight. Since low birth weight may result from either inadequate
6 growth in utero or delivery before the usual 40 weeks of gestation, three of the authors only
7 considered low birth weight (<2500 g) in full-term deliveries (>37 weeks); the other four
8 controlled for gestational age in the analysis. When correlations with other pollutants were
9 reported in these studies, they ranged from 0.5 to 0.8. All of these studies reported strong effects
10 for other pollutants.
11 Lee et al. (2003) reported a significant association between NO2 and low birth weight,
12 and the association was only for exposure in the second trimester. It is difficult to hypothesize
13 any biological mechanism relating NO2 exposure and fetal growth specifically in the second
14 trimester. Bell et al. (2007) reported an increased risk of low birth weight with NO2 exposure
15 averaged over pregnancy (OR = 1.027 [95% CI: 1.002, 1.051]) and a deficit in birthweight
16 specific to the first trimester. In addition, the deficit in birthweight appeared to be greater among
17 black mothers (-12.7 g per IQR increase in NO2 [95% CI: -18.0, -7.5]) than for white mothers
18 (-8.3 g per IQR increase in NO2 [95% CI: -10.4,-6.3]).
19 Six studies investigated NO2 exposure related to preterm delivery (Annex Table
20 AX6.3-13). Three reported positive associations (Bobak, 2000; Maroziene and Grazuleviciene,
21 2002; Leem et al., 2006) and three reported no association (Liu et al., 2003; Ritz et al., 2000;
22 Hansen et al., 2006). Among the studies reporting an association, two (Bobak, 2000; Leem
23 et al., 2006) reported significant associations for both the first trimester and the third trimester
24 of pregnancy. The third (Maroziene and Grazuleviciene, 2002) reported significant increases in
25 risk for exposure in the first trimester and averaged over all of pregnancy. In two (Bobak, 2000;
26 Leem et al., 2006) of the positive studies, NO2 exposure was correlated with SO2 exposure
27 (r = 0.54, 0.61 for the two studies); the third study did not report correlations.
28 Three studies (see details in Annex Table AX6.3-14) specifically investigated fetal
29 growth by comparing birth weight for gestational age with national standards. Two of these
30 studies reported associations of small for gestational age with NO2 exposure. Mannes et al.
31 (2005) determined increased risk for exposure in trimesters 2 and 3, while Liu et al. (2003)
March 2008 3-108 DRAFT-DO NOT QUOTE OR CITE
-------
1 reported risks associated only with NC>2 exposure in the first month of pregnancy. In all three
2 studies, NO2 exposure was correlated with CO exposure (r = 0.69, 0.57, 0.72 in the three studies)
3 (Mannes et al., 2004; Liu et al., 2003).
4 Two additional studies found that NO2 concentrations were associated with
5 hospitalization for respiratory disease in the neonatal period (Dales et al., 2006) and sudden
6 infant death syndrome (SIDS) (Dales et al, 2004).
7
8 Toxicological Studies
9 Only a few studies have investigated the effects of NO2 on reproduction and development
10 of NO2. Exposure to 1-ppm NO2 for 7 h/day, 5 days/week for 21 days resulted in no alterations
11 in spermatogenesis, germinal cells, or interstitial cells of the testes of 6 rats (Kripke and Sherwin,
12 1984). Similarly, breeding studies by Shalamberidze and Tsereteli (1971) found that long-term
13 NO2 exposure had no effect on fertility. However, there was a statistically significant decrease
14 in litter size and neonatal weight when male and female rats exposed to 1.3-ppm NO2, 12 h/day
15 for 3 months were bred. In utero death due to NO2 exposure resulted in smaller litter sizes, but
16 no direct teratogenic effects were observed in the offspring. In fact, after several weeks,
17 NO2-exposed litters approached weights similar to those of controls.
18 Following inhalation exposure of pregnant Wistar rats to 0.5- and 5.3-ppm NO2 for
19 6 h/day throughout gestation (21 days), maternal toxic effects and developmental disturbances in
20 the progeny were reported (Tabacova et al., 1985; Balabaeva and Tabacova, 1985; Tabacova and
21 Balabaeva, 1988). Maternal weight gain during gestation was significantly reduced at 5.3 ppm,
22 with findings of pathological changes, e.g., desquamative bronchitis and bronchiolitis in the
23 lung, mild parenchymal dystrophy and reduction of glycogen in the liver, and blood stasis and
24 inflammatory reaction in the placenta. At gross examination, the placentas of the high-dose
25 dams were smaller in size than those of control rats. A marked increase of lipid peroxides was
26 found in maternal lungs and particularly in the placenta at both exposure levels by the end of
27 gestation (Balabaeva and Tabacova, 1985). Disturbances in the prenatal development of the
28 progeny were registered, such as 2- to 4-fold increase in late post-implantation lethality at 0.5
29 and 5.3 ppm, respectively, as well as reduced fetal weight at term and stunted growth at 5.3 ppm.
30 These effects were significantly related to the content of lipid peroxides in the placenta, which
31 was suggestive of a pathogenetic role of placental damage. Teratogenic effects were not
32 observed, but dose-dependent morphological signs of embryotoxicity and retarded intrauterine
March 2008 3-109 DRAFT-DO NOT QUOTE OR CITE
-------
1 development, such as generalized edema, subcutaneous hematoma, retarded ossification, and
2 skeletal aberrations, were found at both exposure levels.
3 In a developmental neurotoxicity study, Wistar rats were exposed by inhalation to 0,
4 0.025-, 0.05-, 0.5-, or 5.3-ppm NO2 during gestational days 0 through 21. Maternal toxicity was
5 not reported. Viability and physical development (i.e., incisor eruption and eye opening) were
6 significantly affected in the group exposed only to 5.3 ppm. There was a concentration-
7 dependent change in neurobehavioral endpoints such as disturbances in early neuromotor
8 development, including coordination deficits, retarded locomotor development, and decreased
9 activity and reactivity. Statistical significance was observed in some or all of the endpoints at
10 the time point(s) measured in the 0.05-, 0.5-, and 5.3-ppm exposure groups.
11 Di Giovanni et al. (1994) investigated whether in utero exposure of rats to NO2 changed
12 ultrasonic vocalization, a behavioral response indicator of the development of emotionality.
13 Pregnant Wistar female rats were exposed by inhalation to 0-, 1.5-, and 3-ppm NO2 from day 0
14 to 20 of gestation. Dam weight gain, pregnancy length, litter size at birth, number of dams
15 giving birth, and postnatal mortality were unaffected by NO2. There was a significant decrease
16 in the duration of ultrasonic signals elicited by the removal of the pups from the nest in the
17 10-day and 15-day-old male pups in the 3-ppm NO2-exposed group. No other parameters of
18 ultrasonic emission, or of motor activity, were significantly affected in these prenatally exposed
19 pups. Since prenatal exposure to NO2 did not significantly influence the rate of calling, the
20 authors concluded that this decrease in the duration of ultrasounds in the 3-ppm NO2 exposed
21 group does not necessarily indicate altered emotionality, and the biological significance of these
22 findings remains to be determined.
23
24 Summary of Evidence on the Effects of Long-Term NO2 Exposure on Reproductive and
25 Developmental Effects
26 In summary, the epidemiologic evidence does not consistently report associations
27 between NO2 exposure and growth retardation; however, some evidence is accumulating for
28 effects on preterm delivery. Similarly, scant animal evidence supports a weak association
29 between NO2 exposure and adverse birth outcomes and provides little mechanistic information or
30 biological plausibility for the epidemiology findings.
31
March 2008 3-110 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.5.4 Summary of Other Morbidity Effects Associated with Long-Term
2 NO2 Exposure
3 This section has presented epidemiologic and toxicological studies evaluating limited
4 evidence of cancer incidence, cardiovascular effects, and reproductive and developmental effects
5 linked to long-term NO2 exposure. The epidemiologic evidence is limited but suggestive for
6 effects of long-term NO2 exposure on adverse birth outcomes and cancer incidence. Animal
7 studies do not provide mechanistic information to support these observational findings. Some
8 toxicological studies have demonstrated an effect of NO2 exposure on cardiovascular endpoints.
9 However, whether these effects are the direct result of NO2 exposure or secondary responses to
10 lung edema and changes in blood hemoglobin content are not known. Similar findings have
11 been reported in the epidemiologic literature for short-term exposures only. Overall, these data
12 are inadequate to infer the presence or absence of a causal relationship.
13
14
15 3.6 MORTALITY ASSOCIATED WITH LONG-TERM EXPOSURE
16 No studies of mortality associated with long-term NO2 exposure were evaluated in the
17 1993 AQCD. More recently, there have been several studies that examined mortality
18 associations with long-term exposure to air pollution, including NO2, using Cox proportional
19 hazards regression models with adjustment for potential confounders. The U.S. studies tended to
20 focus on effects of PM, while the European studies tended to investigate the influence of traffic-
21 related air pollution.
22
23 3.6.1 U.S. Studies on the Long-Term NO2 Exposure Effects on Mortality
24 Dockery et al. (1993) conducted a prospective cohort study to examine the effects of air
25 pollution, focusing on PM components, in six U.S. cities, which were chosen based on the levels
26 of air pollution (with Portage, WI being the least polluted and Steubenville, OH, the most
27 polluted). In this study, a 14-to-16-year mortality follow-up of 8,111 adults in the six cities was
28 conducted. Fine particles were the strongest predictor of mortality; NO2 was not analyzed in this
29 study. Krewski et al. (2000) conducted sensitivity analysis of the Harvard Six Cities study and
30 examined associations between gaseous pollutants (i.e., Os, NO2, SO2, CO) and mortality. NO2
31 showed risk estimates similar to those for PM2 5 per "low to high" range increment with total
32 (1.15 [95% CI: 1.04, 1.27] per 10-ppb increase), cardiopulmonary (1.17 [95% CI: 1.02, 1.34]),
March 2008 3-111 DRAFT-DO NOT QUOTE OR CITE
-------
1 and lung cancer (1.09 [95% CI: 0.76, 1.57]) deaths; however, in this dataset NO2 was highly
2 correlated with PM2.5 (r = 0.78), SO42 (r = 0.78), and SO2 (r = 0.84).
3 Pope et al. (1995) examined PM effects on mortality using the American Cancer Society
4 (ACS) cohort. Air pollution data from 151 U.S. metropolitan areas in 1980 were linked with
5 individual risk factors in 552,138 adults who resided in these areas when enrolled in the study in
6 1982. Mortality was followed up until 1989. As with the Harvard Six Cities Study, the main
7 hypothesis of this study was focused on fine particles and SO42 , and gaseous pollutants were not
8 analyzed. Krewski et al. (2000) examined association between gaseous pollutants (means by
9 season) and mortality in the Pope et al. (1995) study data set. NO2 showed weak but negative
10 associations with total and cardiopulmonary deaths using either seasonal means. An extended
11 study of the ACS cohort doubled the follow-up time (to 1998) and tripled the number of deaths
12 compared to the original study (Pope et al., 2002). In addition to PM2 5, all the gaseous
13 pollutants were examined. SO2 was associated with all the mortality outcomes (including all
14 other cause of deaths), but NO2 showed no associations with the mortality outcomes (RR = 1.00
15 [95% CI: 0.98, 1.02] per 10-ppb increase in multiyear average NO2).
16 Lipfert et al. (2000a) conducted an analysis of a national cohort of-70,000 male U.S.
17 military veterans who were diagnosed as hypertensive in the mid 1970s and were followed up for
18 about 21 years (up to 1996). This cohort was 35% black and 81% had been smokers at one time.
19 Thus, unlike other cohort studies described in this section, this hypertensive cohort with a very
20 high smoking rate is not representative of the U.S. population. Total suspended particulates
21 (TSP), PMio, CO, O3, NO2, SO2, SO42 , PM2.5, and PMi0.2.5 were considered. The county of
22 residence at the time of entry to the study was used to estimate exposures. Four exposure periods
23 (1960-1974, 1975-1981, 1982-1988, and 1989-1996) were defined, and deaths during each of the
24 three most recent exposure periods were considered. Lipfert et al. (2000a) noted that the
25 pollution risk estimates were sensitive to the regression model specification, exposure periods,
26 and the inclusion of ecological and individual variables. The authors reported that indications of
27 concurrent mortality risks were found for NO2 (the estimate was not given with confidence
28 bands) and peak Os. Their subsequent analysis (Lipfert et al., 2003) reported that the air
29 pollution-mortality associations were not sensitive to the adjustment for blood pressure. Lipfert
30 et al. (2006a) also examined associations between traffic density and mortality in the same
31 cohort, whose follow-up period was extended to 2001. They reported that traffic density was a
March 2008 3-112 DRAFT-DO NOT QUOTE OR CITE
-------
1 better predictor of mortality than the ambient air pollution variables, with the possible exception
2 of 63. The log-transformed traffic density variable was moderately correlated with NO2
3 (r = 0.48) and PM2.5 (r = 0.50) in this data set. For the 1989 to 1996 data period (the period that
4 showed generally the strongest associations with exposure variables among the four periods), the
5 estimated mortality relative risk for NO2 was 1.025 (95% CI: 0.983, 1.068) per 10-ppb increase
6 in a single-pollutant model. The two-pollutant model with the traffic density variable reduced
7 NO2 risk estimates to 0.996 (95% CI: 0.954, 1.040). Interestingly, as the investigators pointed
8 out, the risk estimates due to traffic density did not vary appreciably across these four periods.
9 They speculated that other environmental factors such as particles from tire, traffic noise, spatial
10 gradients in socioeconomic status might have been involved. Lipfert et al. (2006b) further
11 extended analysis of the veteran's cohort data to include one year of the EPA's Speciation
12 Trends Network (STN) data, which collected chemical components of PM2.5. As in the previous
13 Lipfert et al. (2006a) study, traffic density was the most important predictor of mortality, but
14 associations were also seen for EC, vanadium, N(V, and nickel. NO2, 63, and PMio also
15 showed positive but weaker associations. The risk estimate for NO2 was 1.043 (95% CI: 0.967,
16 1.125) per 10-ppb increase in a single-pollutant model. Multipollutant model results were not
17 presented for NO2 in this updated analysis. The results from the series of studies by Lipfert et al.
18 are suggestive of traffic-related air pollution, but the study population (hypertensive with very
19 high smoking rate) was not representative of the general U.S. population.
20 Abbey et al. (1999) investigated associations between long-term ambient concentrations
21 of PMio, O3, NO2, SO2, and CO (1973 to 1992) and mortality (1977 to 1992) in a cohort of
22 6,338 nonsmoking California Seventh-day Adventists. Monthly indices of ambient air pollutant
23 concentrations at 348 monitoring stations throughout California were interpolated to ZIP code
24 centroids according to home or work location histories of study participants, cumulated, and then
25 averaged over time. They reported associations between PMio and total mortality for males and
26 nonmalignant respiratory mortality for both sexes. NO2 was not associated with all-cause,
27 cardiopulmonary, or respiratory mortality for either sex. Lung cancer mortality showed large
28 risk estimates for most of the pollutants in either or both sexes, but the number of lung cancer
29 deaths in this cohort was very small (12 for female and 18 for male); therefore, it is difficult to
30 interpret these estimates.
March 2008 3-113 DRAFT-DO NOT QUOTE OR CITE
-------
1 When comparing the results of the U.S. studies mentioned above, differences in study
2 population characteristics and geographic unit of averaging for pollution exposure estimates need
3 to be considered. Most of the U.S. studies used a "semi-individual" study design, in which
4 information on health outcomes and potential confounders are collected and adjusted for on an
5 individual basis, but community-level air pollution exposure estimates are used. It is not clear to
6 what extent exposure error affects these types of studies. Unlike regional air pollutants (e.g.,
7 SC>42 , PM2.5) in the eastern United States whose levels are generally uniform within the scale of
8 the metropolitan area, the within-city variation for more locally-impacted pollutants such as NO2,
9 SO2, and CO are likely to be larger and, therefore, are more likely to have larger exposure errors
10 in the semi-individual studies. The smaller number of monitors available for NO2 in the United
11 States may make the relative error worse for NO2 compared to other pollutants. Exposure error
12 in these long-term studies likely contributes to the inconsistencies observed across studies. For
13 example, the ACS study found no associations with NO2; however, NO2 was among the
14 pollutants that showed associations with mortality in the veterans' study, with traffic density
15 showing the strongest association. The geographic resolution of air pollution exposure
16 estimation varied in these studies: the Metropolitan Statistical Area (MSA)-level averaging in
17 the ACS study and county-level averaging in the veterans' study. Traffic density and other
18 pollutants that showed mortality associations in the veterans study, including EC and NO2, are
19 more localized pollutants; therefore, using county-level aggregation, rather than MSA-level, may
20 have resulted in smaller exposure misclassification.
21
22 3.6.2 European Studies on the Long-Term NOi Exposure Effects on
23 Mortality
24 In contrast to the U.S. studies described above, the European studies described below,
25 have more spatially-resolved exposure estimates, because their hypotheses or study aims
26 involved mortality effects of traffic-related air pollution. Only one study from France (Filleul
27 et al., 2005) used a design similar to the Harvard Six Cities study or ACS in that it did not study
28 traffic-related air pollution and the exposure estimate was not done on an individual basis.
29 Hoek et al. (2002) investigated a random sample of 5,000 subjects from the Netherlands
30 Cohort Study on Diet and Cancer (NLCS) ages 55 to 69 from 1986 to 1994. Long-term exposure
31 to traffic-related air pollutants (black smoke and NO2) was estimated using 1986 home
32 addresses. Exposure was estimated with the measured regional and urban background
March 2008 3-114 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentration and an indicator variable for living near major roads. Cardiopulmonary mortality
2 was associated with living near a major road (RR = 1.95 [95% CI: 1.09, 3.52]) and less strongly
3 with the estimated air pollution levels (e.g., for NO2, RR = 1.32 [95% CI: 0.88, 1.98] per 10-ppb
4 increase). The risk estimate for living near a major road was 1.41 (95% CI: 0.94, 2.12) for total
5 mortality. For estimated NO2 (incorporating both background and local impact), the RR was
6 1.15(95%CI: 0.60, 2.23) per 10 ppb). Because the NO2 exposure estimates were modeled,
7 interpretation of their risk estimates is not straightforward. However, these results do suggest
8 that NO2, as a marker of traffic-related air pollution, was associated with these mortality
9 outcomes.
10 Filleul et al. (2005) investigated long-term effects of air pollution on mortality in 14,284
11 adults who resided in 24 areas from seven French cities when enrolled in the PAARC survey (for
12 air pollution and chronic respiratory diseases) in 1974. Models were run before and after
13 exclusion of six area monitors influenced by local traffic as determined by the NO/NO2 ratio of
14 >3. Before exclusion of the six areas, none of the air pollutants were associated with mortality
15 outcomes. After exclusion of these areas, analyses showed associations between total mortality
16 and TSP, black smoke, NO2, and NO. The estimated NO2 risks were 1.28 (95% CI: 1.07, 1.55),
17 1.58 (95% CI: 1.07, 2.33), and 2.12 (95% CI: 1.11, 4.03) per 10-ppb increase in NO2 mean over
18 the study period for total, cardiopulmonary, and lung cancer mortality, respectively. From these
19 results, the authors noted that inclusion of air monitoring data from stations directly influenced
20 by local traffic could overestimate the mean population exposure and bias the results. This point
21 raises a concern for NO2 exposure estimates used in other studies (e.g., ACS) in which the
22 average of available monitors was used to represent the exposure of each city's entire population.
23 Nafstad et al. (2004) investigated the association between mortality and long-term air
24 pollution exposure in a cohort of Norwegian 16,209 men followed from 1972/1973 through
25 1998. PM was not considered in this study because measurement methods changed during the
26 study period. NOx, rather than NO2, was used. Exposure estimates for NOx and SO2 were
27 constructed using models based on subjects' addresses and emission data for industry, heating,
28 and traffic and measured concentrations. Addresses linked to 50 of the busiest streets were given
29 an additional exposure based on estimates of annual average daily traffic. The adjusted risk
30 estimate for total mortality was 1.16 [95% CI: 1.12, 1.22] for a 10 ppb) increase in the estimated
31 exposure to NOx. Corresponding mortality risk estimates for respiratory causes other than lung
March 2008 3-115 DRAFT-DO NOT QUOTE OR CITE
-------
1 cancer was 1.16 (95% CI: 1.06, 1.26); for lung cancer, 1.11 (95% CI: 1.03, 1.19); and for
2 ischemic heart diseases, 1.08 (95% CI: 1.03,1.12). 862 did not show similar associations. The
3 risk estimates presented for categorical levels of these pollutants showed mostly monotonic
4 exposure-response relationships for NOx. These results are suggestive of the effects of traffic-
5 related air pollution on long-term mortality, but NOx likely represented the combined effects of
6 that source, possibly including PM, which could not be analyzed in this study. A case-control
7 study of 1,043 men aged 40 to 75 with lung cancer and 2,364 controls in Stockholm County
8 (Nyberg et al., 2000) reported similar results to this study. They mapped residence addresses to
9 a GIS database indicating 4,300 traffic-related line sources and 500 point sources of NO2
10 exposure. Exposure was derived from a model validated by comparison to actual measurements
11 of NO2 at six sites. Exposure to NO2 at 10 ppb was associated with an OR of 1.20 (95% CI:
12 0.94 1.49). Exposure to the 90th percentile (>29.26 |ig/m3) of NO2 was associated with an OR
13 of 1.44 (95% CI: 1.05, 1.99).
14 Naess et al. (2007) investigated the concentration-response relationships between air
15 pollution (i.e., NO2, PMi0, PM2 5) and cause-specific mortality using all the inhabitants of Oslo,
16 Norway, aged 51 to 90 years on January 1, 1992 (n = 143,842), with follow-up of deaths from
17 1992 to 1998. An air dispersion model was used to estimate the air pollution levels for 1992
18 through 1995 in all 470 administrative neighborhoods. Correlations among these pollutants were
19 high (ranged 0.88 to 0.95). All causes of deaths, cardiovascular causes, lung cancer, and COPD
20 were associated with all indicators of air pollution for both sexes and both age groups. The
21 investigators reported that the effects appeared to increase at NO2 levels higher than 40 |ig/m3
22 (21 ppb) in the younger age (51 to 70 years) group and with a linear effect in the interval of 20 to
23 60 |ig/m3 (10 to 31 ppb) for the older age group (see Figure 3.6-1). However, they also noted
24 that a similar pattern was found for both PM2.5 and PMi0. Thus, the apparent threshold effect
25 was not unique to NO2. NO2 risk estimates for all-cause mortality were presented only in a
26 figure. The findings are generally consistent with those from Nafstad et al. (2003, 2004) studies,
27 in which a smaller number of male-only subjects were analyzed. While NO2 effects were
28 suggested, the high correlation among the PM indices and NO2 or NOX makes it difficult to
29 ascribe these associations to NO2/NOx alone.
30 Gehring et al. (2006) investigated the relationship between long-term exposure to air
31 pollution originating from traffic and industrial sources and total and cause-specific mortality in
March 2008 3-116 DRAFT-DO NOT QUOTE OR CITE
-------
Ages 51-70 years
06
0.4
i/i
TJ 0.2 -
1
5 0.0
0.2
All causes
0.2
Ages 71-90 years
0,4
20
40
60
40
60
Nitrogen dioxide (fig/rn^)
Nitrogen dioxide (jjg/m1)
Figure 3.6-1. Age-adjusted, nonparametric smoothed relationship between NOi and
mortality from all causes in Oslo, Norway, 1992 through 1995.
Source: Nsess et al. (2007).
1 a cohort of women living in North Rhine-Westphalia, Germany. The area includes the Ruhr
2 region, one of Europe's largest industrial areas. Approximately 4,800 women (age 50 to
3 59 years) were followed for vital status and migration. Exposure to air pollution was estimated
4 by GIS models using the distance to major roads, NO2, and PMio (estimated from 0.71 x TSP,
5 based on available PMio and TSP data in the area) concentrations from air monitoring station
6 data. Cardiopulmonary mortality was associated with living within a 50-m radius of a major
7 road (RR= 1.70 [95% CI: 1.02, 2.81]) and NO2 (RR= 1.72 [95% CI: 1.28, 2.29] per 10-ppb
8 increase in annual average). Exposure to NO2 was also associated with all-cause mortality (1.21
9 [95% CI: 1.03, 1.42] per 10 ppb). NO2 was generally more strongly associated with mortality
10 than the indicator for living near a major road (within versus beyond a 50-m radius) or PMi0.
11 Most of the European cohort studies estimated an individual subject's exposure based on
12 spatial modeling using emission and concentration data. These studies may provide more
13 accurate exposure estimates than the community-level air pollution estimates typically used in
14 the U.S. studies. However, because they generally involve modeling with such information as
15 traffic volume and other emission estimates in addition to monitored concentrations, additional
16 uncertainties may be introduced. Thus, validity and comparability of various methods may need
17 to be examined. In addition, because the relationship between the concentration measured at the
18 community monitors and the health effects is ultimately of interest in this review, interpreting the
March 2008
3-117
DRAFT-DO NOT QUOTE OR CITE
-------
1 risk estimates based on individual-level exposures will require an additional step to translate the
2 difference. Finally, a more accurate exposure estimate does not solve the problem of the
3 surrogate role that NO2 may play. Most of these studies do acknowledge this issue and generally
4 treat NO2 as a surrogate marker, but the extent of such surrogacy and confounding with other
5 traffic- or combustion-related pollutant is not clear at this point. In the Hoek et al. study (2002),
6 the indicator of living near a major road was a better predictor of mortality than the estimated
7 NO2 exposures. In the Gehring et al. (2006) study, the estimated NO2 exposure was a better
8 predictor of total and cardiopulmonary mortality than the indicator of living near a major road.
9 Comparing the results for the indicators of living near a major road and the estimated NO2 or
10 NOx exposures is not straightforward, but it is possible that, depending on the presence of other
11 combustion sources (e.g., the North Rhine-Westphalia area included highly industrial areas),
12 NO2 may represent more than traffic-related pollution.
13
14 3.6.3 Summary of Evidence of the Effect of Long-Term NOi Exposure on
15 Mortality
16 Figure 3.6-2 summarizes the NO2 relative risk estimates for total mortality from the
17 studies reviewed in the previous sections. The relative risk estimates are grouped by those that
18 used community- or ecologic-level exposure estimates and those that used individual-level
19 exposure estimates, but because of the small number of studies listed, no systematic pattern
20 could be elucidated. The relative risk estimates for total mortality ranged from 0 to 1.28 per
21 10-ppb increase in annual or longer averages of NO2.
22 Potential confounding by copollutants needs to be considered in the interpretation of the
23 NO2 risk estimates. Not all of the studies presented correlations between NO2 and other
24 pollutants, but those that did indicated generally moderate to high correlations. For example, in
25 the Harvard Six Cities study (Krewski et al, 2000), the French study (Filleul et al., 2005), and the
26 German study (Gehring et al., 2006), the correlation between NO2 and PM indices ranged from
27 0.72 to 0.8. The high correlations between NO2 and PM suggest possible confounding between
28 these pollutants. Further, the results from the Netherlands study (Hoek et al., 2002), that living
29 near major roads was more strongly associated with mortality than NO2, supports a possible
30 surrogate role of NO2 as a marker of traffic-related pollution. However, this does not preclude
31 the possibility of NO2 playing a role in interactions among the traffic-related pollutants.
March 2008 3-118 DRAFT-DO NOT QUOTE OR CITE
-------
Relative risk per 10 ppb N02
0,5 1.0 1.5 2,0 2.5
Seventh-day Adventist (Abbey et al., 1999)
Male
Female
Harvard six cities (Krewski et al., 2000)
ACS (Pope et al., 2002)
Veterans* cohort study
French PAAC survey (Filleul et al, 2005)
The Netherlands NLCS (Hoek et al., 2002)
North Rhine-Westphalia, Germany; female
(Gehring et al., 2006)
Studies with ecologic exposure estimates
Studies with individual exposure estimates
Figure 3.6-2. Total mortality relative risk estimates from long-term studies. The original
estimate for the Norwegian study was estimated for NOx. Conversion of
NO2 = 0.35 x NOX was used.
1 Essentially no information is available on the possible effect modification of apparent NO2-
2 mortally associations.
March 2008
3-119 DRAFT-DO NOT QUOTE OR CITE
-------
1 Available information on risk estimates for more specific causes of death with long-term
2 exposure to NO2 is limited. Among the studies with larger number of subjects, the ACS study
3 (Pope et al., 2002) examined cardiopulmonary and lung cancer deaths, but as with the all-cause
4 deaths, they were not associated with NC>2. In the Naess et al. (2007) analysis of all inhabitants
5 of Oslo, Norway, NC>2 relative risk estimates for COPD were higher than those for other causes,
6 but the same pattern was seen for PM2.5 and PMio. In the German study by Gehring et al. (2006),
7 NC>2 relative risk estimates for cardiopulmonary mortality were larger than those for all-cause
8 mortality, but, again, the same pattern was seen for PMio. Thus, higher risk estimates seen for
9 specific causes of deaths were not specific to NC>2 in these studies.
10 In long-term studies, different geographic scales were used to estimate air pollution
11 exposure estimates across studies. Since the relative strength of association with health
12 outcomes among various air pollutant indices may be affected by the spatial distribution of the
13 pollutants (i.e., regional versus local), the numbers of monitors available, and the scale of
14 aggregation in the study design, it is not clear how these factors affected the apparent difference
15 in results.
16 In the U.S. and European cohort studies examining the relationship between long-term
17 exposure to NC>2 and mortality, results were generally not consistent. Further, when associations
18 were suggested, they were not specific to NC>2, also implicating PM and other traffic indicators.
19 The relatively high correlations reported between NC>2 and PM indices (r ~ 0.8) and the
20 unresolved issue of surrogacy and interactions make it difficult to interpret the observed
21 associations; thus, these data are inadequate to infer the presence or absence of a causal
22 relationship.
March 2008 3-120 DRAFT-DO NOT QUOTE OR CITE
-------
i 4. PUBLIC HEALTH SIGNIFICANCE
2
O
4 This chapter discusses several issues relating to the broader public health significance of
5 exposure to nitrogen oxides (NOX). Topics discussed are (1) defining adverse health effects, (2)
6 the shape of the concentration-response relationship for nitrogen dioxide (NO2) and evidence for
7 thresholds, (3) potentially susceptible subpopulations and both intrinsic and extrinsic factors that
8 influence susceptibility, and (4) the size of potentially susceptible population in the United
9 States. Exposure to ambient NC>2 is associated with a variety of outcomes including increases in
10 respiratory symptoms, particularly among asthmatic children, and emergency department (ED)
11 visits and hospital admissions for respiratory diseases among children and older adults (65+
12 years).
13
14
15 4.1 DEFINING ADVERSE HEALTH EFFECTS
16 The American Thoracic Society (ATS) published an official statement titled "What
17 Constitutes an Adverse Health Effect of Air Pollution?" (ATS, 2000b). This statement updated
18 the guidance for defining adverse respiratory health effects published 15 years earlier (ATS,
19 1985), taking into account new investigative approaches used to identify the effects of air
20 pollution and reflecting concern for impacts of air pollution on specific susceptible groups. In
21 the 2000 update, there was an increased focus on quality-of-life measures as indicators of
22 adversity and a more specific consideration of population risk. As shown in Figure 4.1-1, a shift
23 in the population mean may or may not result in clinically significant health consequences for
24 individuals within the population. However, an increased risk to the entire population is viewed
25 as adverse, even though it may not increase the risk of any identifiable individual to an
26 unacceptable level (ATS, 2000b). For example, a population of asthmatics could have a
27 distribution of lung function such that no identifiable single individual has a level associated with
28 significant impairment, and exposure to air pollution could shift the distribution to lower levels
29 that still do not bring any identifiable individual to a level that is associated with clinically
30 relevant effects. This shift to a lower level would be considered adverse because individuals
31 within the population would have diminished reserve function and, therefore, would be at
March 2008 4-1 DRAFT-DO NOT QUOTE OR CITE
-------
*- C
o o
^ IP
C 15
-------
TABLE 4.1-1. GRADATION OF INDIVIDUAL RESPONSES TO SHORT-TERM NO2
EXPOSURE IN PERSONS WITH IMPAIRED RESPIRATORY SYSTEMS
Symptomatic
Response
Wheeze
Cough
Chest pain
Duration of response
Functional
Response
FEVi change
Bronchial
responsiveness
Specific airways
resistance (SRaw)
Duration of response
Impact of
Responses
Interference with
normal activity
Medical treatment
Normal
None
Infrequent
Cough
None
None
None
Decrements of
Within normal
range
Within normal
range (± 20%)
None
Normal
None
No change
Mild
With otherwise
normal breathing
Cough with deep
breath
Discomfort just
noticeable on
exercise or deep
breath
<4h
Small
Decrements of
3 to < 10%
Increases of
<100%
SRaw increased
<100%
<4h
Mild
Few persons
choose to limit
activity
Normal medication
as needed
Moderate
With shortness of
breath
Frequent spontaneous
cough
Marked discomfort
on exercise or deep
breath
>4 h, but <24 h
Moderate
Decrements of >10 but
<20%
Increases of <300%
SRaw increased up
to 200% or up to 15cm
H2Os
>4hbut<24h
Moderate
Many persons choose to
limit activity
Increased frequency
of medication use or
additional medication
Severe
Persistent with
shortness of breath
Persistent
uncontrollable cough
Severe discomfort
on exercise or deep
breath
>24h
Large
Decrements of
>20%
Increases of >300%
SRaw increased
>200% or more than
15cmH2O-s
>24h
Severe
Most persons choose
to limit activity
Physician or
emergency
department visit
An increase in bronchial responsiveness of 100% is equivalent to a 50% decrease in provocative dose that produces a 20% decrease in FEVi
(PD20) or provocative dose that produces a 100% increase in SRaw (PD100).
Source: This table is adapted from the 1996 O3 AQCD (Table 9-2, page 9-25) (U.S. Environmental Protection
Agency, 1996).
1 the approaches taken to define their relative adversity are valid and reasonable in the context of
2 the new ATS (2000b) statement.
March 2008
4-3
DRAFT-DO NOT QUOTE OR CITE
-------
1 As assessed in detail in earlier chapters of this document and briefly recapitulated in
2 preceding sections of this chapter, exposures to a range of NO2 concentrations have been
3 reported to be associated with increased severity of health effects, such as respiratory symptoms,
4 ED visits and hospital admission for respiratory causes. Respiratory effects associated with
5 short-term NC>2 exposures have been extensively studied and are clearly causally related to NO2
6 exposure.
7
8
9 4.2 CONCENTRATION-RESPONSE FUNCTIONS AND POTENTIAL
10 THRESHOLDS
11 An important consideration in characterizing the public health impacts associated with
12 NO2 exposure is whether the concentration-response relationship is linear across the full
13 concentration range encountered or if nonlinear departures exist along any part of this range. Of
14 particular interest is the shape of the concentration-response curve at and below the level of the
15 current annual average standard of 53 parts per billion (ppb) (0.053 parts per million [ppm]).
16 Identifying possible "thresholds" in air pollution epidemiologic studies is challenging.
17 Various factors tend to linearize the concentration-response relationship, obscuring any threshold
18 that may exist. Factors that complicate determining the shape of the concentration-response
19 curve included: interindividual variation in susceptibility and response, additivity of pollutant-
20 induced effects to naturally occurring background disease processes, the extent to which
21 additional health effects are due to other environmental insults having a mode of action similar to
22 NC>2, exposure error, response error, and low data density in the lower concentration range.
23 Additionally, if the concentration-response relationship is shallow, identification of any existing
24 threshold will be more difficult.
25 The slope of the NC>2 concentration-response relationship has been explored in several
26 studies. To examine the shape of the concentration-response relationship between NC>2 and daily
27 physician consultations for asthma and lower respiratory disease in children, Hajat et al. (1999)
28 used bubble plots to examine residuals of significant models plotted against moving averages of
29 NC>2 concentration. They noted a weak trend for asthma and 0-1 day moving average of NC>2
30 and suggested that effects are weaker at lower concentrations and stronger at higher
31 concentrations than predicted by the linear model. These departures are in accord with the
32 sigmoidal dose-response models. A number of epidemiologic studies have reported no evidence
March 2008 4-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 for nonlinear relationships or a threshold response in relationships between NO2 and mortality or
2 morbidity. One multicity time-series study (Samoli et al., 2006) examined the relationship
3 between mortality and NO2 in 29 European cities. There was no indication of a response
4 threshold, and the concentration-response curves were consistent with a linear relationship. Kim
5 et al. (2004b) investigated the presence of a threshold in relationships between air pollutants and
6 mortality in Seoul, Korea, by analyzing data using a log-linear Generalized Additive Model
7 (GAM; linear model), a cubic natural spline model (nonlinear model), and a B-mode splined
8 model (threshold model). There was no evidence NO2 had a nonlinear association with
9 mortality. Burnett et al. (1997a) used the locally estimated smoothing splines (LOESS)
10 smoothing curves to describe the concentration-response for respiratory and cardiac
11 hospitalizations. The curves appeared linear, and there was no significant nonlinearity detected
12 by the inclusion of a quadratic in the models (Burnett et al., 1997b).
13 In general, positive associations were observed between ambient NO2 concentrations and
14 ED visits and hospitalizations for asthma in various epidemiologic studies conducted in different
15 study locations and during varying time periods. The effect was strongest when subjects of all
16 ages were included in the analyses. Several of these studies demonstrated a concentration-
17 response function. Jaffe et al. (2003) found a positive association between ambient NO2 and
18 asthma ED visits among Medicaid-enrolled asthmatics in two urban cities in Ohio. When a
19 concentration-response relationship was examined by quintile of NO2 concentration, risk
20 decreased in the second quintile in both cities and increased monotonically in the third and fourth
21 quintiles in Cleveland, but decreased in the third quintile in Cincinnati. The lack of consistency
22 in results may be due to the uncontrolled effects of copollutants, or other factors. Tenias et al.
23 (1998) reported a positive and significant association between ambient NO2 and ED visits in
24 Valencia's Hospital Clinic Universitari from 1994 to 1995. Castellsague et al. (1995) found a
25 small but significant association of NO2 and ED visits due to asthma in Barcelona. Specifically,
26 the adjusted risk estimates of asthma visits for each quartile of NO2 showed increased risks in
27 each quartile for the summer months, but not the winter months. Together these four studies
28 indicate some disagreement in the trend of the concentration-response curve from about 30 to
29 50 ppb 24-h NO2 and indicate increased risk above 50 ppb.
March 2008 4-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 4.3 POTENTIALLY SUSCEPTIBLE POPULATIONS TO HEALTH
2 EFFECTS RELATED TO SHORT-TERM AND LONG-TERM
3 EXPOSURE TO NO2
4 Many factors such as genetic (Kleeberger et al., 2005) and social (Gee and Payne-
5 Sturges, 2006) determinants of disease may contribute to interindividual variability and
6 heightened susceptibility to NO2 among persons within populations. The previous AQCD for
7 Oxides of Nitrogen (U.S. Environmental Protection Agency, 1993) identified certain groups
8 within the population that may be more susceptible to the effects of NO2 exposure, including
9 persons with preexisting respiratory disease, children, and older adults. Findings from new
10 studies support the conclusions from the previous assessment with regard to susceptibility.
11
12 4.3.1 Preexisting Disease as a Potential Risk Factor
13 A recent report of the National Research Council (NRC) emphasized the need to evaluate
14 the effect of air pollution on susceptible groups including those with respiratory illnesses and
15 cardiovascular disease (CVD) (NRC, 2004). Generally, chronic obstructive pulmonary disease
16 (COPD), conduction disorders, congestive heart failure (CHF), diabetes, and myocardial
17 infarction (MI) are conditions believed to put persons at greater risk for adverse events
18 associated with air pollution. In addition, epidemiologic evidence indicates persons with
19 bronchial hyperresponsiveness (BHR) as determined by methacholine provocation may be at
20 greater risk for symptoms such as phlegm and lower respiratory symptoms than subjects without
21 BHR (Boezen et al., 1998). Several researchers have investigated the effect of air pollution
22 among potentially sensitive groups with preexisting medical conditions.
23
24 Asthmatics
25 There is evidence from epidemiologic studies for an association between NO2 exposure
26 and children's hospital admissions, ED visits, and calls to doctors for asthma. This evidence
27 comes from large longitudinal studies, panel studies, and time-series studies. NO2 exposure is
28 associated with aggravation of asthma effects that include symptoms, medication use, and lung
29 function. Effects of NO2 on asthma were most evident with a cumulative lag of 2 to 6 days,
30 rather than same-day levels of NO2. Time-series studies also demonstrated a relationship in
31 children between hospital admissions or ED visits for asthma and NO2 exposure, even after
32 adjusting for copollutants such as particulate matter (PM) and carbon monoxide (CO). Important
March 2008 4-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 evidence is also available from epidemiologic studies of indoor NO2 exposures. A number of
2 recent studies show associations with wheeze, chest tightness, and length of symptoms (Belanger
3 et al., 2006); respiratory symptom rates (Nitschke et al., 2006); school absences (Pilotto et al.,
4 1997a); respiratory symptoms, likelihood of chest tightness, and asthma attacks (Smith et al.,
5 2000); and severity of virus-induced asthma (Chauhan et al., 2003). However, several studies
6 (Mukala et al., 1999, 2000; Farrow et al., 1997) evaluating younger children found no
7 association between indoor NC>2 and respiratory symptoms.
8 Airways hyperresponsiveness in asthmatics to both nonspecific chemical and physical
9 stimuli and to specific allergens appears to be the most sensitive indicator of response to NO2
10 (U.S. Environmental Protection Agency, 1993). Responsiveness is determined using a challenge
11 agent, which causes an abnormal degree of constriction of the airways as a result of smooth
12 muscle contraction. This response ranges from mild to severe (spanning orders of magnitude)
13 and is often accompanied by production of sputum, cough, wheezing, shortness of breath, and
14 chest tightness. Though some asthmatics do not have this bronchoconstrictor response
15 (Pattemore et al., 1990), increased airways responsiveness is correlated with asthma symptoms
16 and increased asthma medication usage. Clinical studies have reported increased airways
17 responsiveness to allergen challenge in asthmatics following exposure to 0.26-ppm NC>2 for
18 30 min during rest (Barck et al., 2002; Strand et al., 1997, 1998).
19 Toxicological studies provide biological plausibility that asthmatics are likely susceptible
20 to the effects of NO2 exposure. Numerous animal studies provide evidence that NO2 can
21 produce inflammation and lung permeability changes. These studies provide evidence for
22 several mechanisms by which NC>2 exposure can induce effects, including reduced mucociliary
23 clearance, and alveolar macrophage function such as depressed phagocytic activity and altered
24 humoral- and cell-mediated immunity. These are all mechanisms that can provide biological
25 plausibility for the NC>2 effects in asthmatic children observed in epidemiologic studies. One
26 limitation of this work is that effects on markers of inflammation, such as bronchoalveolar
27 lavage fluid levels of total protein and lactate dehydrogenase and recruitment or proliferation of
28 leukocytes, occur only at exposure levels of >5 ppm. Studies conducted at these higher exposure
29 concentrations may elicit mechanisms of action and effects that do not occur at near-ambient
30 levels of NC>2. Chauhan et al. (2003) reviewed potential mechanisms by which NC>2 exacerbates
31 asthma in the presence of viral infections. These mechanisms include "direct effects on the
March 2008 4-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 upper and lower airways by ciliary dyskinesis, epithelial damage, increases in pro-inflammatory
2 mediators and cytokines, rises in IgE concentration, and interactions with allergens, or indirectly
3 through impairment of bronchial immunity."
4
5 Cardiopulmonary Disease and Diabetes
6 While less evidence is available for these conditions, it is possible that preexisting
7 cardiovascular-related conditions may lead to heightened susceptibility to the effects of NO2
8 exposure. Some recent epidemiologic studies have reported that persons with preexisting
9 conditions may be at increased risk for adverse cardiac health events associated with ambient
10 NO2 concentrations (Peel et al., 2006; Mann et al., 2002; D'Ippoliti et al., 2003; von Klot et al.,
11 2005). Peel et al. (2006) reported evidence of effect modification by comorbid hypertension and
12 diabetes on the association between ED visits for arrhythmia and NO2 exposure. In another
13 study, a statistically significant positive relationship was reported between NO2 concentrations
14 and hospitalizations for ischemic heart disease (IHD) among those with prior diagnoses of CHF
15 and arrhythmia (Mann et al., 2002). However, Mann et al (2002) notes the vulnerability in the
16 secondary CHF group could be due to increased prevalence of MI as the primary diagnosis in
17 this group. In addition, these authors state they were unable to distinguish the effects of NO2
18 from other traffic pollutants (Mann et al., 2002). Von Klot et al. (2005) reported cardiac
19 readmission among MI survivors was associated with NO2 and this association was robust to
20 adjustment for PMi0. Modification of the association between NO2 and MI by conduction
21 disorders but not diabetes or hypertension was observed by D'Ippoliti et al. (2003). Park et al.
22 (2005b) examined the relationship of NO2 and heart rate variability (HRV) among those with
23 IHD, hypertension and diabetes but did not find an association.
24 There is limited evidence from clinical or toxicological studies on potential susceptibility
25 to NO2 in persons with CVDs; however, the limited epidemiologic evidence suggests that these
26 individuals may be more sensitive to effects of NO2 exposure or air pollution in general.
27 Reductions in blood hemoglobin (-10%) have been reported in healthy subjects following
28 exposure to NO2 (1 to 2 ppm) for a few hours during intermittent exercise (Frampton et al.,
29 2002). The clinical significance of hemoglobin reduction in persons with significant underlying
30 lung disease, heart disease, or anemia has not been evaluated, but the reductions could lead to
31 adverse cardiovascular consequences. These consequences would be exacerbated by
March 2008 4-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 concomitant exposure to CO, a combustion copollutant of NOx that binds to hemoglobin and
2 reduces oxygen availability to tissues and organs.
3
4 4.3.2 Age-Related Variations in Susceptibility
5 Children and older adults (65+ years) are often considered at increased risk from air
6 pollution compared to the general population. The American Academy of Pediatrics (2004)
7 concludes that children and infants are among the most susceptible to many air pollutants,
8 including NC>2. Because 80% of alveoli are formed postnatally and changes in the lung continue
9 through adolescence, the developing lung is highly susceptible to damage from exposure to
10 environmental toxicants (Dietert et al., 2000). In addition to children, older adults frequently are
11 classified as being particularly susceptible to air pollution. The basis of the increased sensitivity
12 in the elderly is not known, but one hypothesis is that it may be related to changes in the
13 respiratory tract lining fluid antioxidant defense network and/or to a decline in immune system
14 surveillance or response (Kelly et al., 2003). The generally declining health status of many older
15 adults may also increase their risks to air pollution-induced effects.
16 There is evidence that associations of NO2 with both respiratory ED visits and
17 hospitalizations are stronger among children (Peel et al., 2005; Atkinson et al., 1999b; Fusco
18 et al., 2001; Hinwood et al., 2006; Anderson et al., 1998) and older adults (Migliaretti et al.,
19 2005; Atkinson et al., 1999b; Schouten et al., 1996; Ponce de Leon et al., 1996; Prescott et al.,
20 1998). However, two studies (Sunyer et al., 1997; Migliaretti et al., 2005) found no difference
21 in the rates of ED visits associated with NO2 concentrations for children (<15 years) and adults
22 (15 to 64 years). Luginaah et al. (2005) and Wong et al. (1999) found no statistically significant
23 difference in the elderly and adult age groups.
24 Many field studies focused on the effect of NO2 on the respiratory health of children,
25 while fewer field studies have compared the effect of NC>2 in adults and other age groups. In
26 general, children and adults experienced decrements in lung function associated with short-term
27 ambient NO2 exposures (see Section 3.1.5). Importantly, a number of long-term exposure
28 studies suggest effects in children that include impaired lung function growth, increased
29 respiratory symptoms and infections, and onset of asthma (see Section 3.4).
30 In elderly populations, associations between NC>2 and hospitalizations or ED visits for
31 CVD, including stroke, have been observed in several studies (Anderson et al., 2007a; Atkinson
March 2008 4-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 et al., 1999b; Jalaludin et al., 2006; Hinwood et al., 2005; Wong et al., 1999; Barnett et al., 2006;
2 Zanobetti and Schwartz, 2006; Simpson et al., 2005a; Wellenius et al., 2005b; Morgan et al.,
3 1998; Morris et al., 1995). However, some results were inconsistent across cities (Morris et al.,
4 1995), and investigators could not distinguish the effect of NO2 from the effect of other traffic-
5 related pollutants such as PM and CO (Simpson et al., 2005a; Barnett et al., 2006; Morgan et al.,
6 1998b; Jalaludin et al., 2006; Zanobetti and Schwartz, 2006).
7 Several mortality studies investigated age-related differences in NO2 effects. Among the
8 studies that observed positive associations between NO2 and mortality, a comparison of all-age-
9 or <64-years-of-age-group NO2-mortality risk estimates to that of the >65-years-of-age group
10 indicates that, in general, the elderly population is more susceptible to NO2 effects (Biggeri et al.,
11 2005; Burnett et al., 2004). One study (Simpson et al., 2005a) found no difference in increases
12 in CVD mortality associated with NO2 concentrations between all ages and those participants of
13 > 65 years of age.
14
15 4.3.3 Gender
16 A limited number of studies stratified results by gender. Lugninaah et al. (2005) found
17 increases in hospital admissions associated with NO2 among females but not males. In a study of
18 children in Toronto, Canada, NO2 was positively associated with asthma admissions among both
19 boys and girls (Lin et al., 2005). However, in a study of asthma admissions among children in
20 Vancouver, NO2 was significantly and positively associated with asthma hospitalization only for
21 boys in the low socioeconomic group (Lin et al., 2004). An increased association with asthma
22 with exposure to traffic pollutants was observed for girls (Kim et al., 2004a). Decrements in
23 forced vital capacity (FVC) and forced expiratory volume in 1 s (FEVi) growth associated with
24 NO2 were reported in male and female children in Mexico (Rojas-Martinez et al., 2007a,b).
25
26 4.3.4 Genetic Factors for Oxidant and Inflammatory Damage from Air
27 Pollutants
28 A consensus now exists among epidemiologists that genetic factors related to health
29 outcomes and ambient pollutant exposures merit serious consideration (Kauffmann et al., 2004;
30 Gilliland et al., 1999). Interindividual variation in human responses to air pollutants suggests
31 that some subpopulations are at increased risk of detrimental effects from pollutant exposure, and
32 it has become clear that genetic background is an important susceptibility factor (Kleeberger,
March 2008 4-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 2005). Several criteria must be satisfied in selecting and establishing useful links between
2 polymorphisms in candidate genes and adverse respiratory effects. First, the product of the
3 candidate gene must be significantly involved in the pathogenesis of the adverse effect of
4 interest, often a complex trait with many determinants. Second, polymorphisms in the gene must
5 produce a functional change in either the protein product or in the level of expression of the
6 protein. Third, in epidemiologic studies, the issue of confounding by other environmental
7 exposures must be carefully considered.
8 Several glutathione S-transferase (GST) families have common, functionally important
9 polymorphic alleles that significantly affect host defense function in the lung (e.g., homozygosity
10 for the null allele at the GSTM1 and GSTT1 loci, homozygosity for the A105G allele at the
11 GSTP1 locus). GST genes are inducible by oxidative stress. Exposure to radicals and oxidants
12 in air pollution induces decreased glutathione (GSH) that increases transcription of GSTs.
13 Individuals with genotypes that result in enzymes with reduced or absent peroxide activity are
14 likely to have reduced oxidant defenses and potentially increased susceptibility to inhaled
15 oxidants and radicals.
16 Studies of genotype, respiratory health, and air pollution in general have been conducted
17 (Lee et al., 2004; Gilliland et al., 2002; Gauderman et al., 2007). NO2-related genetic effects
18 have been presented primarily by Romieu et al. (2006) and indicate that asthmatic children with
19 GSTM1 null and GSTP1 Val/Val genotypes appear to be more susceptible to developing
20 respiratory symptoms related to O3, but not NO2, concentrations. It was suggested that ambient
21 NO2 concentrations may affect breathing in children regardless of their GSTM1 or GSTP1
22 genotypes. GSTM1-positive and GSTP1 lie/lie- and Ile/Val-genotype children were more likely
23 to experience cough and bronchodilator use in response to NO2 than GSTMl-null and GSTP1-
24 Val/Val children. Contrary to expectations, a 20-ppb increase in ambient NO2 concentrations
25 was associated with a decrease in bronchodilator use among GSTP1 Val/Val-genotype children.
26 It remains plausible that there are genetic factors that can influence health responses to NO2,
27 though the few available studies do not provide specific support for genetic susceptibility to NO2
28 exposure.
March 2008 4-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 4.3.5 Populations with Potentially High Exposure
2 Certain groups may experience relatively high exposure to NO2, thus forming a
3 potentially vulnerable or susceptible population. Many studies find that indoor, personal, and
4 outdoor NO2 levels are strongly associated with proximity to traffic or traffic density (see
5 Section 2.5.4). NO2 concentrations in heavy traffic or on freeways, which have been observed in
6 the range of 40 to 70 ppb, can be more than twice the residential outdoor or residential/arterial
7 road level (Lee et al., 2000; Westerdahl et al., 2005). Due to high air exchange rates, NO2
8 concentrations inside a vehicle could rapidly approach levels outside the vehicle during
9 commuting; the mean in-vehicle NO2 concentration has been observed to be between 2 to 3
10 times ambient levels (see Section 2.5.4). Those with occupations that require them to be in or
11 close to traffic or roadways (e.g., bus and taxi drivers, highway patrol officers, toll collectors) or
12 those with long commutes could be exposed to relatively high levels of NO2 compared to
13 ambient levels.
14
15 4.3.6 Socioeconomic Position
16 Social-economic position (SEP) is a known determinant of health, and there is evidence
17 that SEP modifies the effects of air pollution (O'Neill et al. 2003; Makri and Stilianakis, 2008).
18 Higher exposures to air pollution and greater susceptibility to its effects may contribute to a
19 complex pattern of risk among those with lower SEP. Conceptual frameworks have been
20 proposed to explain the relationship between SEP, susceptibility, and exposure to air pollution.
21 Common to these frameworks is the consideration of the broader social context in which persons
22 live, and its effect on health in general (O'Neill et al., 2003; Gee and Payne-Sturges, 2004), as
23 well as on maternal and child health (Morello-Frosch and Shenassa, 2006) and asthma (Wright
24 and Subramanian, 2007) specifically. Multilevel modeling approaches that allow
25 parameterization of community-level stressors such as increased life stress as well as individual
26 risk factors are considered by these authors. In addition, statistical methods that allow for
27 temporal and spatial variability in exposure and susceptibility have been discussed in the recent
28 literature (Jerrett and Finkelstein, 2005; Kunzli et al., 2005).
29 Most studies to date have examined modification by SEP indicators on the association
30 between mortality and PM (O'Neill et al., 2003; Martins et al., 2004; Jerrett et al., 2004;
31 Finkelstein et al., 2003; Romieu et al., 2004a) or other indices such as traffic density, distance to
March 2008 4-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 roadway or a general air pollution index (Ponce et al., 2003; Woodruff et al., 2003; Finkelstein
2 et al., 2004). However, modification of NO2 associations has been examined in a few studies.
3 For example, in a study conducted in Seoul, Korea, community-level SEP indicators modified
4 the association of air pollution with ED visits for asthma: of the five criteria air pollutants
5 evaluated, NO2 showed the strongest association in lower SEP districts compared to high SEP
6 districts (Kim et al., 2007.) In addition, Clougherty et al. (2007) evaluated exposure to violence
7 (a chronic stressor) as a modifier of the effect of traffic-related air pollutants, including NC>2, on
8 childhood asthma. The authors reported an elevated risk of asthma with a 4.3-ppb increase in
9 NC>2 exposure solely among children with above-median exposure to violence in their
10 neighborhoods.
11
12
13 4.4 ESTIMATION OF POTENTIAL NUMBERS OF PERSONS IN
14 AT-RISK SUSCEPTIBLE POPULATION GROUPS IN THE
15 UNITED STATES
16 Although MVrelated health risk estimates may appear to be small, they may well be
17 biologically significant from an overall public health perspective owing to the large numbers of
18 persons in the potential risk groups. Several population groups have been identified as possibly
19 having increased susceptibility or vulnerability to adverse health effects from NO2, including
20 children, older adults, and persons with preexisting pulmonary diseases. One consideration in
21 the assessment of potential public health impacts is the size of various population groups that
22 may be at increased risk for health effects associated with NO2-related air pollution exposure.
23 Table 4.4.1 summarizes information on the prevalence of chronic respiratory conditions in the
24 U.S. population in 2004 and 2005 (National Center for Health Statistics, 2006a,b). Individuals
25 with preexisting cardiopulmonary disease constitute a fairly large proportion of the population,
26 with tens of millions of persons included in each disease category. Of most concern are those
27 persons with preexisting respiratory conditions, with approximately 10% of adults and 13% of
28 children having been diagnosed with asthma and 6% of adults with COPD (chronic bronchitis
29 and/or emphysema).
30 There are approximately 2.5 million deaths from all causes per year in the U.S.
31 population, with about 100,000 deaths from chronic lower respiratory diseases (Kochanek et al.,
32 2004) and 4,000 from asthma (NCHS, 2006c). For respiratory health diseases, there are
March 2008 4-13 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 4.4-1. PREVALENCE OF SELECTED RESPIRATORY DISORDERS BY AGE GROUP AND BY
GEOGRAPHIC REGION IN THE UNITED STATES (2004 [U.S. ADULTS] AND 2005 [U.S. CHILDREN] NATIONAL
HEALTH INTERVIEW SURVEY)
Age (years)
Chronic
Condition/Disease
Respiratory Conditions
Asthma
COPD
Chronic Bronchitis
Emphysema
Chronic
Condition/Disease
Respiratory Conditions
Asthma
Adults
(18+ years)
Cases
(x 106) %
14.4 6.7
8.6 4.2
3.5 1.7
Children
(<18 years)
Cases
(x 106) %
6.5 8.9
18-44 45-64
o/o o/o
6.4 7.0
3.2 4.9
0.3 2
Age (years)
0-4 5-11
o/o %
6.8 9.9
65-74 75+
o/o %
7.5 6.6
6.1 6.3
4.9 6.0
12-17
o/o
9.6
Region
Northeast Midwest South West
o/o % % %
6.8 6.8 6.0 7.5
4.0 4.7 4.4 3.5
1.5 1.7 2.0 1.1
Region
Northeast Midwest South West
o/o % % %
10.1 8.5 9.3 7.9
Source: National Center for Health Statistics (2006a,b).
March 2008
4-14
DRAFT-DO NOT QUOTE OR CITE
-------
1 nearly 4 million hospital discharges per year (DeFrances et al., 2005), 14 million ED visits
2 (McCaig and Burt, 2005), 112 million ambulatory care visits (Woodwell and Cherry, 2004), and
3 an estimated 700 million restricted-activity days per year due to respiratory conditions (Adams
4 et al., 1999). Of the total number of visits for respiratory disease, 1.8 million annual ED visits
5 are reported for asthma, including more than 750,000 visits by children. In addition, nearly
6 500,000 annual hospitalizations for asthma are reported (NCHS, 2006c).
7 Centers for Disease Control and Prevention (CDC) analyses have shown that the burden
8 of asthma has increased over the past two decades (NCHS, 2006c). In 2005, approximately 22.2
9 million (7.7% of the population) currently had asthma. The incidence was higher among
10 children (8.9% of children) compared to adults (7.2%) (Note: 2004 data is shown in Table 4.4-1,
11 with a prevalence of 6.7%). In addition, prevalence and severity is higher among certain ethnic
12 or racial groups such as Puerto Ricans, American Indians, Alaskan Natives, and African
13 Americans. The asthma hospitalization rate for black persons was 240% higher than for white
14 persons. Puerto Ricans were reported to have the highest asthma death rate (360% higher than
15 non-Hispanic white persons) and non-Hispanic black persons had an asthma death rate that was
16 200% higher than non-Hispanic white persons. Furthermore, a higher prevalence of asthma
17 among persons of lower SEP and an excess burden of asthma hospitalizations and mortality in
18 minority and inner-city communities have been observed in several studies (Wright and
19 Subramanian, 2007). Gender and age are also determinants of prevalence and severity: adult
20 females had a 40% higher prevalence than adult males; and boys, a 30% higher prevalence than
21 girls. Overall, females had a hospitalization rate about 35% higher than males.
22 In addition, population groups based on age group also comprise substantial segments of
23 the population that may be potentially at risk for NO2-related health impacts. Based on U.S.
24 census data from 2000, about 72.3 million (26%) of the U.S. population are under 18 years of
25 age, 18.3 million (7.4%) are under 5 years of age, and 35 million (12%) are 65 years of age or
26 older. Hence, large proportions of the U.S. population are in age groups that are likely to have
27 increased susceptibility and vulnerability for health effects from ambient NO2 exposure.
28 Based on data from the American Housing Survey, approximately 36 million persons live
29 within 300 feet (-90 meters) of a four-lane highway, railroad, or airport and 12.6% of U.S.
30 housing units are located within this distance (U.S. Census Bureau, 2006). Furthermore, several
31 exposure studies offer insight into differential exposures to NO2 from traffic in childhood. In
March 2008 4-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 California, 2.3% of schools, grades K-12, with a total enrollment of more than 150,000 students
2 were located within -500 feet (150 m) of high-traffic roads, and a higher proportion of nonwhite
3 and economically disadvantaged students attended schools within close proximity to these high-
4 traffic roadways (Green et al., 2004). Similar findings were reported for Detroit schoolchildren
5 (Wu and Batterman, 2006). Figure 4.4-1 shows the proportion of the population living within a
6 certain distance from major roadways as measured by field studies, the U.S. Census, and
7 population exposure models. It also presents results of air quality measurements showing the
8 decrease in concentration of black carbon, a traffic-related pollutant, with increasing distance
9 from the roadway. The considerable size of the population groups at risk indicate that exposure
10 to ambient NO2 could have a significant impact on public health in the United States.
11
12
13 4.5 SUMMARY
14 Both general and specific definitions of adversity are discussed. These general and
15 specific definitions of adversity are multifaceted, involving clinically observable effects, effects
16 on quality of life, loss of reserve capacity, and population distributions of effects.
17 In the limited studies that have specifically examined concentration-response
18 relationships between NO2 and health outcomes, there is little evidence of an effect threshold.
19 However, various factors, such as interindividual variation in response, additivity to background
20 of effect and/or exposure, and measurement error, tend to linearize the dose-response
21 relationship and obscure any population-level thresholds that might exist.
22 Persons with preexisting respiratory disease, children, and older adults may be more
23 susceptible to the effects of NO2 exposure. Individuals in sensitive groups may be affected by
24 lower levels of NO2 than the general population or experience a greater impact with the same
25 level of exposure. A number of factors may increase susceptibility to the effects of NO2.
26 Studies generally report a positive excess risk for asthmatics, and there is emerging evidence that
27 cardiovascular disease (CVD) may cause persons to be more susceptible, though it is difficult to
28 distinguish the effect of NO2 from other traffic pollutants. Children and older adults (65+ years)
29 may be more susceptible than adults, possibly due to physiological changes occurring among
30 these age groups. Evidence, albeit inconsistent, exists for a gender-age-based difference in
31 susceptibility, with the incidence of asthma differing for boys and girls at different ages (higher
32 for boys at younger ages, higher for girls at older ages).
March 2008 4-16 DRAFT-DO NOT QUOTE OR CITE
-------
£0
«
0
90 -
80 -
70 -
60 -
5Q -
40 -
30 -
20 '
10 -
n
Population Fraction
* HAPEM6(2007) .
• Garshicketal. (2003)
A McConnell et al. (2006)
• American Housing Survey (2005)
Black Carbon
o Zhuetal. (2002)
•
A
»
1
m
' A
0
0
o
0
- 90
- 80
- 70
' 60
- 50
- 40
- 30
- 20
- 10
M
E
c"
0
re
c
0)
0
o
o
c
O
.a
re
0
0
re
DO
50 100 150 200 250 300
Distance from Roadway, m
350
400 450
Figure 4.4-1. Fraction of the population living within a specified distance from roadways.
For comparison, concentrations of the traffic copollutant black carbon are
plotted as a function of distance from the roadway.
1 Although increases in risk associated with NC>2 exposure may be small in magnitude, the
2 population potentially affected by NC>2 is large. A considerable fraction of the population
3 resides, works, or attends school near major roadways, and these persons are likely to have
4 increased exposure to NO2. Of this population, those with physiological susceptibility will have
5 even greater risks of health effects related to NC>2. New studies of genetic determinants of
6 NO2-related health effects as well as community-level stressors that influence susceptibility may
7 inform future assessment of the health effects of NCh, but current evidence is limited as few
8 studies have been conducted. Furthermore, there may be interactions between factors that
9 influence susceptibility.
March 2008
4-17
DRAFT-DO NOT QUOTE OR CITE
-------
i 5. INTEGRATIVE SUMMARY AND CONCLUSIONS
2
3
4 5.1 INTRODUCTION
5 The previous chapters present the most policy relevant science pertaining to this National
6 Ambient Air Quality Standards (NAAQS) review. This chapter first summarizes and then draws
7 conclusions about atmospheric sciences, exposure assessment, nitrogen dioxide (NOz) exposure
8 indices, and health effects associated with exposure to oxides of nitrogen (NOx). These
9 conclusions have been derived based on explicit guidelines (Section 1.3) derived from the Hill
10 criteria (Hill, 1965) and modeled on other pertinent frameworks.
11 As discussed in the Integrated Plan for the Primary National Ambient Air Quality
12 Standard for Nitrogen Dioxide (U.S. Environmental Protection Agency, 2007), a series of policy
13 relevant questions was devised to frame this assessment of the scientific evidence, which will
14 form the scientific basis for a decision on whether the current primary NAAQS for N02 (0.053
15 parts per million [ppm], annual average) should be retained or revised. This draft Integrated
16 Science Assessment (ISA) focuses on evaluating the newly available scientific evidence to best
17 inform consideration of these framing questions:
18 • Has new information altered the scientific support for the occurrence of health effects
19 following short- and/or long-term exposure to levels of nitrogen oxides found in the
20 ambient air?
21 • What do recent studies focused on the near-roadway environment tell us about health
22 effects of nitrogen oxides?
23 • At what levels of nitrogen oxides exposure do health effects of concern occur?
24 • Has new information altered conclusions from previous reviews regarding the
25 plausibility of adverse health effects caused by exposure to nitrogen oxides?
26 • To what extent have important uncertainties identified in the last review been reduced
27 and/or have new uncertainties emerged?
28 • What are the air quality relationships between short- and long-term exposures to
29 nitrogen oxides?
30 The evidence relative to causality is summarized and integrated across disciplines and
31 conclusions about the health effects of N02 exposure are presented. The framework for the
March 2008 5-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 evaluation of evidence regarding causality is described in Chapter 1. The framework and
2 language draws from similar efforts across the Federal government and wider scientific
3 community, especially from the recent National Academy of Sciences (NAS) Institute of
4 Medicine (IOM) document Improving the Presumptive Disability Decision-Making Process for
5 Veterans (IOM, 2007). A five-level hierarchy is used here to be consistent with the Guidelines
6 for Carcinogen Risk Assessment (U.S. Environmental Protection Agency, 2005). Conclusions
7 concerning causality of association will be placed into one of five categories with regard to
8 weight of the evidence based on the Hill criteria (Hill, 1965). The five descriptors follow:
9 • Sufficient to infer a causal relationship,
10 • Sufficient to infer a likely causal relationship (i.e. more likely than not),
11 • Suggestive but not sufficient to infer a causal relationship,
12 • Inadequate to infer the presence or absence of a causal relationship, and
13 • Suggestive of no causal relationship.
14 This integrative discussion begins with some key conclusions from the atmospheric
15 sciences that are relevant to the interpretation of the health evidence and important
16 underpinnings for potential quantitative assessments, including information about ambient
17 concentrations and monitoring, and estimation of policy relevant background. Consideration of
18 exposure error and related issues is an essential component of this review, and Section 5.2.2
19 provides an overview of the findings that have informed our evaluation of the health evidence.
20 Conclusions regarding causality for different categories of health outcomes, using the framework
21 described previously, are presented along with highlights of the findings for more specific health
22 outcomes.
23
24
25 5.2 KEY FINDINGS RELATED TO THE SOURCE-TO-DOSE
26 RELATIONSHIP
27
28 5.2.1 Atmospheric Science and Ambient Concentrations
29 An understanding of atmospheric processes affecting a given pollutant is crucial for
30 understanding the causal chain linking its sources to health effects. NOz plays a key role in the
31 formation of ozone (Os) and photochemical smog. N02 is an oxidant and can react to form other
32 photochemical oxidants, including organic nitrates like the peroxyacyl nitrates (PANs) and
March 2008 5-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 inorganic acids like nitric acid (HN03). N02 also reacts with toxic compounds such as
2 polycyclic aromatic hydrocarbons (PAHs) to form nitro-PAHs, some of which are more toxic
3 than either reactant alone.
4 As noted in Chapter 2, nitric oxide (NO) and N02 interconvert rapidly in the atmosphere,
5 and so it is often convenient to refer to their sum (NOx) instead of to them individually. The
6 category definition of nitrogen oxides contains a number of nitrogen (N)-containing compounds
7 formed by the oxidation of N02 as described in Chapter 2.
8 • Major anthropogenic sources of NOx include motor vehicles, power plants, and fossil
9 fuel combustion in general. NOx is also emitted by burning biomass fuels.
10 • Natural NOx sources include wildfires, microbial activity in soils, and lightning.
11 • NOx is emitted by all of the above sources mainly as NO. Atmospheric reactions
12 oxidize NO to N02. Thus, most N02 in the atmosphere is the result of the oxidation
13 of primary NO.
14 • The current method of determining ambient NOx and then reporting NOz
15 concentrations by subtraction of NO is subject to interference by NOx oxidation
16 products, chiefly HN03, as well as peroxyacetyl nitrate (PAN) and other oxidized N-
17 containing compounds. Limited available evidence suggests that these compounds
18 and other reaction products result in an overestimation of N02 levels of as much as
19 25% at typical ambient levels (-15 ppb) during summer and in smaller
20 overestimations during winter.
21 • Measurements of these oxidation products in urban areas are sparse. Relationships
22 between these products and N02 are complex and difficult to predict. However,
23 products are expected to peak in the afternoon because of the continued oxidation of
24 N02 emitted during the morning rush hours.
25 • Within the urban core of metropolitan areas, where many of the ambient monitors are
26 sited close to strong NOx sources such as motor vehicles on busy streets and
27 highways, the positive artifacts are much smaller on a relative basis. Conversely, the
28 positive artifacts are larger in locations more distant from local NOx sources.
29 Therefore, variable, positive artifacts associated with measuring N02 using the
30 Federal Reference Method (FRM) severely limit its ability to serve as a precise
March 2008 5-3 DRAFT-DO NOT QUOTE OR CITE
-------
1 indicator of N02 concentrations at the typical ambient levels generally encountered
2 outside of urban cores.
3 • Because its dominant urban source is typically on-road vehicle emissions, ambient
4 NOz generally behaves with the temporal and spatial variability of other traffic-
5 generated pollutants in urban areas.
6 • Nitro-PAHs and other potentially toxic compounds are emitted directly from the
7 exhaust of on- and off-road vehicles and engines. In addition, nitro-PAHs also are
8 formed as products of atmospheric reactions of NOz.
9 • The annual average concentrations of NOz of ~15 parts per billion (ppb) reported by
10 the regulatory monitoring networks are well below the level of the current NAAQS
11 (53 ppb). However, daily maximum 1-h average concentrations can be greater than
12 100 ppb in some locations, e.g., areas with heavy traffic.
13 • Policy Relevant Background concentrations of N02 are much lower than average
14 ambient concentrations and are typically less than 0.1 ppb over most of the United
15 States, with highest values found in agricultural areas.
16
17 5.2.2 Exposure Assessment
18 In addition to ambient N02, people are also exposed to N02 produced by indoor sources
19 (such as gas stoves) and by other microenvironmental sources (such as vehicle exhaust while
20 commuting) and to the oxidation products of N02 either indoors or outdoors. Indoor and outdoor
21 microenvironmental sources of NOx, are often of greater importance in determining a person's
22 total exposure than the largest sources in the national emissions inventories. The amount of time
23 a person spends in different microenvironments and the infiltration characteristics (as a function
24 of the NOz penetration coefficient (P), air exchange rate (a), and the NOz decay rate (k) of these
25 microenvironments are strong determinants of a person's total exposure to NOz and of the
26 association between ambient N02 concentrations and personal exposures to ambient N02. Key
27 findings related to assessing N02 exposures are listed below.
28 • NOz concentrations are highly spatially and temporally variable in urban areas.
29 Intersite correlations for NOz concentrations range from slightly negative to highly
30 positive in examined cities. The range of spatial variation in NOz concentrations is
31 similar to that for 03, but larger than that of fine particulate matter (PM2.s). Twenty-
March 2008 5-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 four-hour concentration differences between individual paired sites in a metropolitan
2 statistical area (MSA) can be larger than the annual means at these sites.
3 • This variability can lead to exposure error in epidemiologic studies conducted in areas
4 for which N02 concentrations are not well correlated between ambient monitoring
5 sites and the community average, or in areas with differences in levels between
6 ambient monitoring sites and the community average.
7 • Rooftop N02 measurements, particularly in inner cities, likely underestimate levels
8 occurring at lower elevations, closer to motor vehicle emissions.
9 • Co-located samples show that passive N02 samplers generally correlate well with
10 FRM ambient samplers, and the concentration differences are generally within 10%.
11 However, personal passive samplers and the ambient samplers are both subject to
12 measurement artifacts.
13 • In the absence of indoor sources, indoor N02 levels are about one-half those found
14 outdoors. In the presence of indoor sources, particularly unvented combustion
15 sources, N02 levels can be much higher than reported ambient concentrations.
16 • Alpha (a), the ratio of personal exposure to N02 of ambient origin to the ambient
17 N02 concentration, ranged from -0.3 to -0.6 in studies where it was determined.
18 • Indoor exposures to N02 are accompanied by exposures to other products of indoor
19 combustion and to products of indoor N02 chemistry, such as nitrous acid (HONO).
20 • The evidence relating ambient levels to personal exposures is inconsistent. Some of
21 the longitudinal studies examined found that ambient levels of N02 were reliable
22 proxies of personal exposures to N02. However, a number of studies did not find
23 significant associations between ambient and personal levels of N02. The differences
24 in results are related in large measure to differences in study design and in exposure
25 determinants. Measurement artifacts and differences in analytical measurement
26 capabilities could also have contributed to the inconsistent results. Indeed, in a
27 number of the studies examined, the majority of measurements of personal N02
28 concentrations were beneath detection limits, and in all studies some personal
29 measurements were beneath detection limits.
30 • The collective variability in all of the above parameters, in general, contributes to
31 exposure measurement errors in air pollution-health outcome studies.
March 2008 5-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 • In two European studies, community averages of personal total exposures were highly
2 correlated with either ambient or outdoor concentrations. However, because of
3 limitations in these studies, caution should be exercised in using these results to
4 determine whether ambient concentrations of N02 can be used as surrogates for
5 community average exposures in epidemiologic studies.
6 Two points about ambient and personal exposures are crucial for interpreting the
7 epidemiologic findings reported in this ISA. First, ambient N02 contributes significantly to total
8 personal N02 exposure, with the ratio of personal N02 exposure of ambient origin to ambient
9 N02 concentrations, or a, ranging from 0.3 to 0.6. Second, the observational evidence relating
10 ambient N02 concentrations to community-average exposures is very limited. For example,
11 although two studies found strong associations between ambient or outdoor [N02] and
12 community-average personal exposures, the utility and universality of these results is
13 compromised by the designs of these studies. Moreover, treating ambient [N02] as a surrogate
14 for personal N02 exposures is additionally complicated by factors such as ambient [N02] spatial
15 variability, errors in ambient [N02] measurements, and variance in exposure factors within a
16 population. The first two of these additional complications are described above in Chapter 2 and
17 the third in Chapter 3.
18
19
20 5.3 KEY HEALTH EFFECTS FINDINGS
21
22 5.3.1 Findings from the Previous Review of the National Ambient Air
23 Quality Standard for Nitrogen Oxides
24 The 1993 Air Quality Criteria for Nitrogen Oxides (AQCD for Nitrogen Oxides)
25 concluded that there were two key health effects of greatest concern at ambient or near-ambient
26 concentrations of N02:
27 • Increases in airways responsiveness of asthmatic individuals after short-term
28 exposures.
29 • Increased occurrence of respiratory illness among children associated with longer-
30 term exposures to N02.
31 Evidence also was found for increased risk of emphysema, but this appeared to be of
32 major concern only with exposures to levels of N02 that were much higher than current ambient
March 2008 5-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 levels of NOz (U.S. Environmental Protection Agency, 1993). Qualitative evidence regarding
2 airways responsiveness and lung function changes was drawn from controlled human exposure
3 and animal toxicological studies; studies did not elucidate a concentration-response relationship.
4 Epidemiologic studies reported increased respiratory symptoms with increased indoor
5 exposures. Animal toxicological findings of lung host defense system changes with
6 exposure provided a biologically plausible basis for these results. Subpopulations considered
7 potentially more susceptible to the effects of N02 exposure included persons with preexisting
8 respiratory disease, children, and the elderly. In the 1993 AQCD, the epidemiologic evidence for
9 respiratory health effects was limited, and no studies had considered effects such as hospital
10 admissions, emergency department (ED) visits, or mortality.
11
12 5.3.2 New Findings on the Health Effects of Exposure to Nitrogen Oxides
13 New evidence developed since 1993 generally has confirmed and extended the
14 conclusions articulated in the 1993 AQCD. Since the 1993 AQCD, the epidemiologic evidence
15 has grown substantially, including new field or panel studies on respiratory health outcomes,
16 numerous time-series epidemiologic studies of effects such as hospital admissions, and a
17 substantial number of studies evaluating mortality risk with short-term NOz exposures. As noted
18 above, no epidemiologic studies were available in 1993 that assessed relationships between
19 nitrogen oxides and outcomes such as hospital admissions, ED visits, or mortality; in contrast,
20 dozens of epidemiologic studies on such outcomes are now included in this evaluation. Several
21 new studies have reported findings from prospective cohort studies on respiratory health effects
22 with long-term NOz exposure. In addition, significant new evidence characterizing the responses
23 of susceptible and vulnerable populations has developed since 1993, particularly concerning
24 children, asthmatics, and those living or working near roadways. While not as marked as the
25 growth in the epidemiologic literature, a number of new toxicological and controlled human
26 exposure studies provide further insights into relationships between NOz exposure and health
27 effects. The conclusions and findings of this evaluation are summarized in Table 5.3-1. Table
28 5.3-1 also summarizes the conclusions drawn in the previous NAAQS review along with those
29 from this draft ISA, and the contrast in available evidence discussed above is clearly illustrated
30 in this table. The marked increase in evidence from epidemiologic studies, along with additional
31 new evidence from human and animal experimental studies, has greatly increased the support for
March 2008 5-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 associations between short-term N02 exposures and respiratory effects compared with evidence
2 available in the previous review and has provided some suggestive evidence for other effects, as
3 highlighted below.
4
5 5.3.2.1 Short-Term Exposure to NOi and Respiratory Health Effects
6 Taken together, recent studies provide scientific evidence that N02 is associated with a
7 range of respiratory effects and are sufficient to infer a likely causal relationship between short-
8 term N02 exposure and adverse effects on the respiratory system. This finding is supported by
9 the large body of new epidemiologic evidence, in combination with findings from human and
10 animal experimental studies. The epidemiologic evidence for respiratory effects can be
11 characterized as consistent, in that associations are reported in studies conducted in numerous
12 locations with a variety of methodological approaches. Considering this large body of
13 epidemiologic studies alone, the findings are coherent in the sense that the studies report
14 associations with respiratory health outcomes that are logically linked together. The consistency
15 and coherence of findings for respiratory effects is illustrated in Figure 5.3-1; this figure
16 combines effect estimates for respiratory symptoms, hospitalizations or ED visits, and
17 respiratory mortality, drawn from figures presented in Chapter 3. Here it can be seen that there
18 are generally positive associations between N02 and respiratory symptoms and hospitalization or
19 ED visits, with a number being statistically significant, particularly the more precise effect
20 estimates. There is also a pattern of positive associations with respiratory mortality, though most
21 are not statistically significant. A number of the epidemiologic studies have been conducted in
22 locations where the ambient N02 levels are well below the level of the current NAAQS; some
23 descriptive statistics for the N02 concentrations used in those studies are included in Appendix
24 Tables 5A and 5B.
25 These health effects associations have been observed in epidemiologic studies reporting
26 maximum ambient concentrations of as high as 100 to 300 ppb, concentrations within the range
27 of the controlled animal and human exposures used in current toxicological and clinical studies
28 reporting respiratory effects. Tables 5.3-2 and 5.3-3 summarize the health endpoints that have
29 been linked with N02 exposure in human clinical and animal toxicological studies, respectively,
30 along with the lower range of doses or concentrations with which these effects have been
31 reported. To put the concentration and dose information in perspective, maximum ambient
32 concentrations from earlier years in the United States and elsewhere were substantially greater
March 2008 5-8 DRAFT-DO NOT QUOTE OR CITE
-------
Relative risk or Odds ratio
Figure 5.3-1. Summary of Epidemiologic Studies Examining Short-Term Exposures to
Ambient NO2 and Respiratory Outcomes. Circles represent effect estimates
and lines indicate the 95% confidence intervals. Effect estimates for studies
conducted in the United States or Canada are presented in black.
ED=emergency department visit. References are listed by study number in
Table 5.3-4.
March 2008
5-9
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-2. KEY HUMAN HEALTH EFFECTS OF EXPOSURE TO NITROGEN DIOXIDE—CLINICAL STUDIES"
t-O
o
o
oo
NO2
(ppm)
Exposure
Duration
Observed Effects
References
0.26 0.5 h Asthmatics exposed to N02 during rest experienced enhanced sensitivity to
challenge-induced decrements in lung function and increased allergen-induced
airways inflammatory response. Inflammatory response to allergen observed in the
absence of allergen-induced lung function response. No N02-induced change in
lung function.
Barcketal. (2002, 2005a)
Strand et al. (1996,1997, 1998)
m
o
?d
o
t—t
H
m
0.1-0.3
0.3-0.4
0.5-2.0 h
Meta-analysis showed increased airways responsiveness following N02 exposure in
asthmatics. Large variability in protocols and responses. Most studies used
nonspecific airways challenges. Airways responsiveness tended to be greater for
resting (mean 45 min) than exercising (mean 102 min) exposure conditions.
2-4 h Inconsistent effects on FVC and FEVi in COPD patients with mild exercise.
Folinsbee (1992)
Gong et al. (2005)
Morrow et al. (1992)
Vagaggini et al. (1996)
en
i
O
O
H
6
O
2
o
1.0-2.0
2-6 h
>2.00
1-3 h
Increased inflammatory response and airways responsiveness to nonspecific
challenge in healthy adults exposed during intermittent exercise. Effects on lung
function and symptoms in healthy subjects not detected by most investigators.
Small decrements in FEVi reported for asthmatics.
Lung function changes (e.g., increased airways resistance) in healthy subjects.
Effects not found by others at 2-4 ppm.
Azadniv et al. (1998)
Blomberg et al. (1997, 1999)
Devlin et al. (1999)
Frampton et al. (2002)
Jorresetal. (1995)
Beil and Ulmer (1976)
Niedingetal. (1979)
Nieding and Wagner (1977)
Niedingetal. (1980) _
N02 = Nitrogen dioxide.
= Functional expiratory volume in 1 s.
FVC = Forced vital capacity.
COPD = Chronic obstructive pulmonary disease.
-------
TABLE 5.3-3. SUMMARY OF TOXICOLOGICAL EFFECTS FROM NO2 EXPOSURE
(LOWEST-OBSERVED-EFFECT LEVEL BASED ON CATEGORY)
t^j Concentration Exposure
§ (ppm) Duration Species
00 0.2 From conception to 12 Rats
wks post delivery
0.5 Weanling period (from Rats
5 wks old to 1 2 wks)
0.5 0.5-10 days Rats
0.5 9 wks Rats
with spikes of 1.5
^ 0.8 1 or 3 days Rats
BALF = Bronchoalveolar lavage fluid.
ROS = Reactive oxygen species.
O
H
b
0
o
0
0
m
0
0
H
m
Effect Category
Increase in BALF Inflammation
lymphocytes
Suppression of ROS Lung host defense
Depressed activation of Lung host defense
arachidonic acid
metabolism and
superoxide production
Increase in the number of Morphological effects
fenestrae in the lungs
Increase in bronchiolar Morphological effects
epithelial proliferation
Reference
Kumae and Arakawa
(2006)
Kumae and Arakawa
(2006)
Robisonetal. (1993)
Mercer et al. (1995)
Earth et al. (1994a)
-------
1 than current levels; yet in the 3-year period 2003-2005, 1-h excursions in the United States have
2 been observed in the range of 100 to 200 ppb (see Chapter 2). The human and animal findings
3 underlying this causal judgment are summarized below.
4
5 Lung Host Defenses and Immunity
6 • Impaired host-defense systems and increased risk of susceptibility to both viral and
7 bacterial infections after N02 exposures have been observed in epidemiologic, human
8 clinical, and animal toxicological studies (Section 3.1.2). A recent epidemiologic
9 study (Chauhan et al., 2003) provided evidence that increased personal exposure to
10 NOz worsened virus-associated symptoms and decreased lung function in children
11 with asthma. The limited evidence from human clinical studies indicates that N02
12 may increase susceptibility to injury by subsequent viral challenge at exposures of as
13 low as 0.6 ppm for 3 h (Frampton et al., 2002). Toxicological studies have shown
14 that lung host defenses are sensitive to N02 exposure, with several measures of such
15 effects observed at concentrations of less than 1 ppm. The epidemiologic and
16 experimental evidence indicates coherence for effects of NOz exposure on host
17 defense (i.e., immune system effects). This group of outcomes also provides
18 plausibility and potential mechanistic support for other respiratory effects described
19 subsequently, such as respiratory symptoms or increased ED visits for respiratory
20 diseases.
21
22 A irways Inflammation
23 • Effects of NOz on airways inflammation have been observed in human clinical and
24 animal toxicological studies at higher than ambient levels The few available
25 epidemiologic studies are suggestive of an association between ambient N02
26 concentrations and inflammatory response in the airways in children, though the
27 associations were inconsistent in the adult populations examined (Section 3.1.3).
28 Human clinical studies provide evidence for increased airways inflammation at NOz
29 concentrations of <2.0 ppm; the onset of inflammatory responses in healthy subjects
30 appears to be between 100 and 200 ppm-min, i.e., 1 ppm for 2 to 3 h (Figure 3.1-1).
31 Increases in biological markers of inflammation were not observed consistently in
32 healthy animals at levels of less than 5 ppm; however, increased susceptibility to N02
March 2008 5-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations of as low as 0.4 ppm was observed when lung vitamin C was reduced
2 (by diet) to levels that were <50% of normal. Together, the findings of human and
3 animal studies provide suggestive evidence for airways inflammation with N02
4 exposure, particularly in the more sensitive groups such as children or asthmatics.
5
6 Airways Hyperresponsiveness
7 • The evidence from human and animal experimental studies provides suggestive
8 evidence for increased airways responsiveness to specific allergen challenges
9 following NOz exposure (Section 3.1.4.1). Recent human clinical studies involving
10 allergen challenge in asthmatics suggest that NOz exposure may enhance the
11 sensitivity to allergen-induced decrements in lung function and increase the allergen-
12 induced airways inflammatory response at exposures of as low as 0.26-ppm N02 for
13 30 min (Figure 3.1-2). Increased immune-mediated pulmonary inflammation was
14 also observed in rats exposed to house dust mite allergen following exposure to
15 5-ppm N02 for 3 h.
16 • Exposure to NO 2 also has been found to enhance the inherent responsiveness of the
17 airways to subsequent nonspecific challenges in human clinical studies (Section
18 3.1.4.2). In general, small but significant increases in nonspecific airways
19 responsiveness were observed in the range of 1.5 to 2.0 ppm for 3 h exposures in
20 healthy adults and between 0.2- and 0.3-ppm NOz for 30-min exposures in
21 asthmatics. Subchronic exposures (6 to 12 weeks) of animals to N02 also increase
22 responsiveness to nonspecific challenges at exposures of 1 to 4 ppm.
23
24 Respiratory Symptoms
25 • Consistent evidence has been observed for an association of respiratory effects with
26 indoor and personal NOz exposures in children at ambient concentration levels
27 (Section 3.1.5.1). In particular, the Pilotto et al. (2004) intervention study provided
28 evidence of improvement in respiratory symptoms with reduced N02 exposure in
29 asthmatic children. This study linked respiratory effects with exposure to N02 from
30 an indoor combustion source, i.e., unflued gas heaters, thus, increasing confidence
31 that NOz is not solely a marker for an air pollution mixture in observed associations
32 with NOz from outdoor sources (particularly traffic).
March 2008 5-13 DRAFT-DO NOT QUOTE OR CITE
-------
1 • The epidemiologic studies using community ambient monitors also found
2 associations between ambient N02 concentration and respiratory symptoms (Section
3 3.1.4.2, see Figure 3.1-6). The results of new multicity studies (Schildcrout et al.,
4 2006; Mortimer et al., 2002) provide further support for associations with respiratory
5 symptoms and medication use in asthmatic children. Positive associations were
6 observed in cities where the median (90th percentile) range was 18 to 26 (34 to 37)
7 ppb for a 24-h average (24-h avg) (Schildcrout et al., 2006) and the mean N02 level
8 (range) was 32 (7 to 96) ppb for a 4-h avg (Mortimer et al., 2002). These
9 concentrations are within the range of 24-h avg concentrations observed in recent
10 years. In the results of multipollutant models, N02 associations in these multicity
11 studies were generally robust to adjustment for copollutants including 03, carbon
12 monoxide (CO), and particulate matter with an aerodynamic diameter of < 10 pm
13 (PMio) (Figure 3.1-7).
14 • Most human clinical studies did not report or observe respiratory symptoms with N02
15 exposure, and animal toxicological studies do not measure effects that would be
16 considered symptoms. The experimental evidence on airways inflammation and
17 immune system effects discussed previously, however, provides some plausibility and
18 coherence for the observed respiratory symptoms in epidemiologic studies.
19
2 0 Lung Function
21 • Recent epidemiologic studies that examined the association between ambient N02
22 concentrations and lung function in children and adults generally produced
23 inconsistent results (Section 3.1.5.1). Human clinical studies did not generally find
24 direct effects of N02 on lung function in healthy adults at levels of as high as 4.0 ppm
25 (Section 3.1.5.2). For asthmatics, the direct effects of N02 on lung function have also
26 been inconsistent at exposure concentrations of less than 1-ppm N02.
27
28 Respiratory ED Visits and Hospitalizations
29 • Epidemiologic evidence exists for positive associations of short-term ambient N02
30 concentrations below the current NAAQS with increased numbers of ED visits and
31 hospital admissions for respiratory causes, especially asthma (Section 3.1.7). As
32 shown in Appendix Table 5B, a number of studies were conducted in locations where
March 2008 5-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 mean (maximum) 24-h concentrations were in the range of 15 to 20 (28 to 82) ppb.
2 These associations are particularly consistent among children and older adults (65+
3 years) when all respiratory outcomes are analyzed together (Figures 3.1-8 and 3.1-9),
4 and among children and subjects of all ages for asthma admissions (Figures 3.1-12
5 and 3.1-13). When examined with copollutant models, associations of NOz with
6 respiratory ED visits and hospital admissions were generally robust and independent
7 of the effects of copollutants (Figures 3.1-10 and 3.1-11). In preceding sections,
8 mechanistic evidence has been described related to host defense and immune system
9 changes, airways inflammation, and airways responsiveness that provide plausibility
10 and coherence for these observed effects.
11
12 5.3.2.2 Short-Term Exposure to NO2 and Cardiovascular Health Effects
13 The available evidence on the effects of short-term exposure to N02 or cardiovascular
14 health effects is inadequate to infer the presence or absence of a causal relationship at this time.
15 • Evidence from epidemiologic studies of heart rate variability (HRV), repolarization
16 changes, and cardiac rhythm disorders among heart patients with ischemic cardiac
17 disease are inconsistent (Section 3.2.1). In most studies, associations with PM were
18 found to be similar or stronger than associations with NOz. The mean 24-h
19 concentrations generally were in the range of 9 to 39 ppb (Annex Table AX6.3-6).
20 Generally positive associations between ambient N02 concentrations and hospital
21 admissions or ED visits for cardiovascular disease have been reported in single-
22 pollutant models where mean 24-h concentrations generally were in the range of 20 to
23 40 ppb (Section 3.2.2); however, most of these effect estimate values were
24 diminished in multipollutant models that also contained CO and PM indices.
25 • Mechanistic evidence of a role for NOz in the development of cardiovascular diseases
26 from studies of biomarkers of inflammation, cell adhesion, coagulation, and
27 thrombosis is lacking (Section 3.2.1.4; Seaton and Dennekamp, 2003). Furthermore,
28 the effects of NOz on various hematological parameters in animals are inconsistent
29 and, thus, provide little biological plausibility for effects of N02 on the cardiovascular
30 system. However, limited evidence from controlled human exposure studies is
March 2008 5-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 suggestive of a reduction in hemoglobin with N02 exposure at concentrations
2 between 1.0 and 2.0 ppm (with 3 h exposures).
3
4 5.3.2.3 Effects of Short-Term Exposure to NO2 on Mortality
5 The epidemiologic evidence is suggestive but not sufficient to infer a casual relationship
6 of short-term exposure to NOz with nonaccidental and cardiopulmonary-related mortality.
7 • Results from several large U.S. and European multicity studies and a meta-analysis
8 study indicated positive associations between ambient NO 2 concentrations and the
9 risk of all-cause (nonaccidental) mortality, with effect estimates ranging from 0.5 to
10 3.6% excess risk in mortality per standardized increment1 (Section 3.3.1,
11 Figure 3.3-2). In general, the N02 effect estimates were robust to adjustment for
12 copollutants. Both cardiovascular and respiratory mortality have been associated
13 with increased N02 concentrations in epidemiologic studies (Figure 3.3-3); however,
14 similar associations were observed for other pollutants, including PM and sulfur
15 dioxide (SO2). The range of risk estimates for mortality excess was generally smaller
16 than that for other pollutants such as PM.
17 • While NOz exposure, alone or in conjunction with other pollutants, may contribute to
18 increased mortality, evaluation of the specificity of this effect is difficult. Clinical
19 studies showing hematologic effects and animal toxicological studies showing
20 biochemical, lung host defense, permeability, and inflammation changes with short-
21 term exposures to N02 provide limited evidence of plausible pathways by which risks
22 of morbidity and, potentially, mortality may be increased, but no coherent picture is
23 evident at this time.
24
25 5.3.2.4 Effects of Long-Term Exposure to NOi on Respiratory Morbidity
26 The epidemiologic and toxicological evidence examining the effect of long-term
27 exposure to NO 2 on respiratory morbidity is suggestive but not sufficient to infer a casual
28 relationship at this time.
29 • A number of epidemiologic studies examined the effects of long-term exposure to
30 N02 and reported positive associations with decrements in lung function and partially
'Excess risk estimates are standardized to a 20-ppb incremental change in daily 24-h avg N02 or a 30-ppb
incremental change in daily 1-h max N02.
March 2008 5-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 irreversible decrements in lung function growth (Section 3.4.1, Figures 3.4-1 and
2 3.4-2). Results from the Southern California Children's Health Study indicated that
3 decrements were similar for boys and girls and among children who had no history of
4 asthma (Gauderman et al., 2004). The mean NOz concentrations in these studies
5 range from 21.5 to 34.6 ppb; thus, all have been conducted in areas where NOz levels
6 are below the level of the NAAQS. Similar associations have also been found for
7 PM, Os, and proximity to traffic (<500 m), though these studies did not report the
8 results of copollutant models. The high correlation among traffic-related pollutants
9 made it difficult to accurately estimate the independent effects in these long-term
10 exposure studies.
11 • Results from the available epidemiologic evidence investigating the association
12 between long-term exposure to N02 and increases in asthma prevalence and
13 incidence are suggestive (Section 3.4.2). Two major cohort studies, the Children's
14 Health Study in southern California (Gauderman et al., 2005) and a birth cohort study
15 in the Netherlands (Brauer et al., 2007) observed significant associations; however,
16 several other studies did not find consistent associations between long-term NOz
17 exposure and asthma outcomes.
18 • Epidemiologic studies conducted in both the United States and Europe also have
19 produced inconsistent results regarding an association between long-term exposure to
20 N02 and respiratory symptoms (Section 3.4.3). While some positive associations
21 were noted, a large number of symptom outcomes were examined and the results
22 across specific outcomes were inconsistent.
23 • Animal toxicological studies demonstrated that N02 exposure resulted in
24 morphological changes in the centriacinar region of the lung and in bronchiolar
25 epithelial proliferation (Section 3.4.4), which may provide some biological
26 plausibility for the observed epidemiologic associations between long-term exposure
27 to NOz and respiratory morbidity. Susceptibility to these morphological effects was
28 found to be influenced by many factors, such as age, compromised lung function, and
29 acute infections.
March 2008 5-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 5.3.2.5 Other Morbidity Effects Associated with Long-Term Exposure to
2 The available epidemiologic and toxicological evidence is inadequate to infer the
3 presence or absence of a causal relationship for carcinogenic, cardiovascular, and reproductive
4 and developmental effects related to long-term N02 exposure.
5 • Two epidemiologic studies conducted in Europe showed an association between long-
6 term N02 exposure and increased incidence of cancer (Nyberg et al., 2000; Nafstad
7 et al., 2003). However, the animal toxicological studies have provided no clear
8 evidence that N02 acts as a carcinogen, though it does appear to act as a tumor
9 promoter at the site of contact (Section 3.5.1). There are no in vivo studies
10 suggesting that N02 causes teratogenesis or malignant tumors. A more likely
11 pathway for N02 involvement in cancer induction is through secondary formation of
12 nitro-polycylic aromatic hydrocarbons (nitro-PAHs), as nitro-PAHs are known to be
13 more mutagenic than the parent compounds.
14 • The very limited epidemiologic and toxicological evidence does not suggest that
15 long-term exposure to N02 has cardiovascular effects (Section 3.5.2). The U.S.
16 Women's Health Initiative study (Miller et al., 2007) did not find any associations
17 between long-term N02 exposure and cardiovascular events. The toxicological
18 studies found some effects of N02 on cardiac performance and heart rate, but only at
19 exposure levels of above 4 ppm.
20 • The epidemiologic evidence is not consistent for associations between N02 exposure
21 and growth retardation; however, some evidence is accumulating for effects on
22 preterm delivery (Section 3.5.3). Similarly, scant animal evidence supports a weak
23 association between N02 exposure and adverse birth outcomes and provides little
24 mechanistic information or biological plausibility for the epidemiologic findings.
25
26 5.3.2.6 Effects of Long-Term Exposure to NOi on Mortality
27 The epidemiologic evidence is inadequate to infer the presence or absence of a causal
28 relationship between long-term exposure to N02 and mortality. In the U.S. and European cohort
29 studies examining the relationship between long-term exposure to N02 and mortality, results
30 were generally inconsistent (Section 3.6, Figure 3.6-2). Further, when associations were
31 suggested, they were not specific to N02, but also implicated PM and other traffic indicators.
March 2008 5-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 The relatively high correlations reported between NOz and PM indices (r ~ 0.8) make it difficult
2 to interpret these observed associations at this time.
3
4 5.3.2.7 Concentration-Response Relationships and Thresholds
5 In studies that have examined concentration-response relationships between N02 and
6 health outcomes specifically, there is little evidence of an effect threshold (Section 4.2). Factors
7 that make it difficult to identify any threshold that may exist include exposure error, response
8 measurement error, low data density in the lower concentration range, interindividual variation in
9 susceptibility to health effects, additivity of pollutant-induced effects to the naturally occurring
10 background disease processes, and the extent to which health effects are due to other
11 environmental insults having a mode of action similar to that of NOz. Additionally, if the
12 concentration-response relationship is shallow, identification of any threshold that may exist will
13 be more difficult.
14
15 5.3.2.8 NO2 Exposure Indices
16 The available NOz indices used to indicate short-term ambient NOz exposure are daily
17 maximum 1-h (1-h max); 24-h average (24-h avg); and 2-week average NOz concentrations.
18 New data on short-term exposures have been published since the 1993 AQCD for Nitrogen
19 Oxides. Some studies examined only one index, and these studies form an evidence base for that
20 individual index. A few studies used both 1-h and 24-h data and, thus, allow a comparison of
21 these averaging periods. These include studies of respiratory symptoms, ED visits for asthma,
22 hospital admissions for asthma, and mortality.
23 • Meta-analysis regression results for asthma ED visits comparing effect estimates for
24 the 1-h and 24-h time periods indicate that effect estimates are slightly, but not
25 significantly, larger with a 24-h avg compared with a 1-h max N02.
26 • Experimental studies in both animals and humans provided evidence that short-term
27 N02 exposure (i.e., <1 h to 2-3 h) can result in respiratory effects such as increased
28 airways responsiveness or inflammation, thereby, increasing the potential for
29 exacerbation of asthma. These findings generally support epidemiologic evidence on
30 short-term exposures, but do not provide evidence that distinguishes effects for one
31 short-term averaging period from another.
March 2008 5-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 • Differences between daily 1-h max and 24-h avg exposures estimates are unlikely to
2 be well characterized by this limited data.
3
4 5.3.2.9 Susceptible and Vulnerable Populations
5 • Based on both short- and long-term studies of an array of respiratory health effects
6 data, persons with preexisting pulmonary conditions are at greater risk from ambient
7 N02 exposures than the general public, with the most extensive evidence available for
8 asthmatics as a potentially susceptible group. In addition, studies suggest that upper
9 respiratory viral infections can trigger susceptibility to the effects of exposure to N02.
10 • There is supporting evidence of age-related differences in susceptibility to N02 health
11 effects such that the elderly population (>65 years of age) appears to be at increased
12 risk of mortality and hospitalizations and that children (< 18 years of age) experience
13 other potentially adverse respiratory health outcomes with increased N02 exposure.
14 • People with occupations that require them to be in or close to traffic or roadways (i.e.,
15 bus and taxi drivers, highway patrol officers) may have enhanced exposure to N02
16 compared to the general population, possibly increasing their vulnerability. A
17 considerable portion of the population resides and/or attends school near major
18 roadways, increasing their exposure to N02 and other traffic pollutants. Otherwise
19 susceptible individuals (schoolchildren, older adults) in this subpopulation, therefore,
20 may be at increased risk.
21 • Recent studies have evaluated the effect of socioeconomic position (SEP) on
22 susceptibility to the effects of N02 exposure; however, to date, these studies are too
23 few in number to draw conclusions.
24 • While data are emerging (Romieu et al., 2006; Islam et al., 2007) and it is believed
25 that a genetic component could be important in characterizing the association
26 between N02 exposure and adverse health effects, currently there are no studies that
27 specifically evaluate this relationship.
28
29
30 5.4 CONCLUSIONS
31 New evidence confirms previous findings in the 1993 Air Quality Criteria Document that
32 short-term nitrogen dioxide (N02) exposure is associated with increased airways responsiveness,
March 2008 5-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 often accompanied by respiratory symptoms, particularly in children and asthmatics.
2 Additionally, the new body of epidemiologic data provides abundant evidence of associations
3 with increased emergency department (ED) visits and hospital admissions for respiratory causes,
4 especially asthma, and short-term ambient exposure to N02. These new findings are based on
5 numerous studies, including panel and field studies, multipollutant studies that control for the
6 effects of other pollutants, and studies conducted in areas where the whole distribution of
7 ambient 24-h average (24-h avg) N02 concentrations was below the current National Ambient
8 Air Quality Standard (NAAQS) level of 53 ppb (see data in Appendix Tables 5A and 5B). These
9 conclusions are supported by evidence from toxicological and controlled human exposure
10 studies. These data sets form a plausible, consistent, and coherent description of a relationship
11 between N02 exposures and an array of adverse health effects that range from the onset of
12 respiratory symptoms to hospital admission. Though an array of studies that examined short-
13 term (24-h avg and 1-h maximum [1-h max]) N02 exposures and respiratory morbidity
14 consistently produced positive associations, it is not possible to discern whether these effects are
15 attributable to average daily (or multiday) concentrations (24-h avg) or high, peak exposures (1-h
16 max).
17 The available evidence on the effects of short-term exposure to N02 for cardiovascular
18 health effects is inadequate to infer the presence or absence of a causal relationship at this time.
19 Though there is no human clinical or animal toxicological evidence, the epidemiologic evidence
20 is suggestive but not sufficient to infer a casual relationship of short-term exposure to N02 with
21 nonaccidental and cardiopulmonary-related mortality.
22 While the evidence supports a direct effect of short-term NO 2 exposure on respiratory
23 morbidity, the available evidence is inadequate to infer the presence or absence of a causal
24 relationship for morbidity and mortality effects related to long-term N02 exposure. Further, the
25 health evidence is found to be inadequate to infer the presence or absence of a causal relationship
26 for carcinogenic, cardiovascular, and reproductive and developmental effects, or for premature
27 mortality, related to long-term N02 exposure.
28 It is difficult to determine from these new studies the extent to which N02 is
29 independently associated with respiratory effects or if N02 is a marker for the effects of another
30 traffic-related pollutant or mix of pollutants (see Chapter 2, Section 5.2.2 for more details on
31 exposure issues). On-road vehicle exhaust emissions are a nearly ubiquitous source of
March 2008 5-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 combustion pollutant mixtures that include N02 and can be an important contributor to N02
2 levels in near-road locations. Although this complicates efforts to disentangle specific N02-
3 related health effects, the evidence summarized in this assessment indicates that N02
4 associations generally remain robust in multipollutant models and supports a direct effect of
5 short-term N02 exposure on respiratory morbidity at ambient concentrations below the current
6 NAAQS. The robustness of epidemiologic findings to adjustment for copollutants, coupled with
7 data from animal and human experimental studies, support a determination that the relationship
8 between N02 and respiratory morbidity is likely causal, while still recognizing the relationship
9 between N02 and other traffic-related pollutants. In addition, an intervention study by Pilotto
10 et al. (2004) found that exposure to N02 from an indoor combustion source is associated with
11 respiratory effects; in this study N02 effects would not be confounded by other motor vehicle
12 emission pollutants, though potential confounding by other pollutants from gas stove emissions,
13 such as ultrafine particles could occur.
14 Identification of a concentration-response relationship is an additional uncertainty that
15 must be considered when describing the association of N02 and adverse health effects. In
16 studies that have examined concentration-response relationships between N02 and health
17 outcomes specifically, there is little evidence of an effect threshold. Because ambient levels of
18 N02 are well below the current NAAQS in many of the epidemiologic study sites, the
19 concentration-response relationship may be shallow, making it difficult to identify any threshold
20 that may exist.
21 Integrating across the epidemiologic, human clinical, and animal toxicological evidence
22 presented above, we find that it is plausible that current N02 exposures can result in adverse
23 impacts to public health at ambient concentrations below the current NAAQS for N02. In
24 particular, a set of coherent and consistent respiratory health outcomes are associated with short-
25 term N02 exposures including exacerbated asthma and other respiratory symptoms, increased
26 airways hyperresponsiveness, inflammation, impaired host defense, aggravated viral infections,
27 and increased emergency department visits and hospital admissions.
March 2008 5-22 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-1. SUMMARY OF EVIDENCE FROM EPIDEMIOLOGICAL,
HUMAN CLINICAL, AND ANIMAL TOXICOLOGICAL STUDIES ON THE
HEALTH EFFECTS ASSOCIATED WITH SHORT- AND LONG-TERM
EXPOSURE TO NO2.
Health Outcome
Conclusion from Previous
NAAQS Review for NOX
Conclusion from 2008 NOX ISA
SHORT-TERM EXPOSURE TO NO2
Respiratory
Morbidity
No Overall Conclusion
"sufficient to infer a likely causal
relationship"
Lung Host Defense
Human clinical studies of host defenses are
rare and their results are equivocal, but
suggestive of the potential for N02 effects;
Animal toxicological studies provide
important evidence indicating that several
defense system components are targets for
inhaled N02, including key elements of host
defenses such as alveolar macrophages
(AMs) and the humoral and cell-mediated
immune systems and further show that N02
exposure can impair the respiratory host
defense system sufficiently so as to result in
the host being more susceptible to
respiratory infection.
Impaired host-defense systems and
increased risk of susceptibility to both viral
and bacterial infections after N02 exposures
have been observed in epidemiologic,
human clinical, and animal toxicologic
studies. Increased susceptibility to cell
injury during ex-vivo viral challenge was
observed following N02 exposures to 0.6
ppm for 3 h in one human clinical study.
Airways
Inflammation
No Studies.
Human clinical studies have reported effects
of N02 on airways inflammation at 1 ppm
for 2 to 3 h exposures in healthy humans.
The animal toxicologic studies and limited
available epidemiologic studies on children
support these findings.
Airways The physiological end point that appears to
Responsiveness be the most sensitive indicator of response
to N02 is a change in airways
responsiveness to bronchoconstrictors in
asthmatics.
In the range of 0.20 and 0.30 ppm, the
increase in responsiveness was attributable
to asthmatics exposed N02 at rest.
Increased responsiveness observed in
healthy individuals exposed to> 1.5 ppm
N02 for 60 min or more.
Human clinical studies of allergen and
nonspecific bronchial challenges in
asthmatics observed increased airways
responsiveness following exposures of 0.2
to O.Sppm N02 for 30 min at rest. Increased
responsiveness to nonspecific challenges
were also observed in animals at higher
N02 levels (1-4 ppm).
Respiratory
Symptoms
Results of a meta-analysis of 9
epidemiologic studies show that children (5-
12 years old) living in homes with gas
stoves are at increased risk for developing
respiratory diseases and illnesses compared
to children living in homes without gas
stoves.
Epidemiologic studies provide consistent
evidence of an association of respiratory
effects with indoor and personal N02
exposures in children. Multicity studies
provide further support for associations
between ambient N02 concentrations and
respiratory symptoms in asthmatic children
at median 24-h avg levels of 18-26 ppb.
March 2008
5-23
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-1 (cont'd). SUMMARY OF EVIDENCE FROM EPIDEMIOLOGICAL,
HUMAN CLINICAL, AND ANIMAL TOXICOLOGICAL STUDIES ON THE
HEALTH EFFECTS ASSOCIATED WITH SHORT- AND LONG-TERM
EXPOSURE TO NO2.
Health Outcome
Conclusion from Previous
NAAQS Review for NOX
Conclusion from 2008 NOX ISA
Lung Function N02 induced lung function changes in
asthmatics have been reported at low (0.2 to
0.5 ppm), but not higher (up to 4 ppm), N02
concentrations. No convincing evidence of
lung function decrements in healthy
individuals at concentrations below 1.0 ppm
N02.
The association between ambient N02
concentrations and lung function in
epidemiologic studies were generally
inconsistent. Recent clinical evidence
generally confirms prior findings.
ED Visits /
Hospital
Admissions
No Studies
Positive and generally robust associations
were observed between ambient N02
concentrations and increased ED visits and
hospital admissions for respiratory causes,
especially asthma. These effects were
observed in studies with mean 24-h avg
concentrations in the range of 15-20 ppb.
Cardiovascular
Morbidity
No Studies
"inadequate to infer the presence or
absence of a causal relationship"
Cardiovascular
Effects
No Studies
Evidence from epidemiologic studies of
heart rate variability, repolarization
changes, and cardiac rhythm disorders
among heart patients with ischemic cardiac
disease are inconsistent.
ED Visits /
Hospital
Admissions
No Studies
Generally positive associations between
ambient N02 concentrations and hospital
admissions or ED visits for cardiovascular
disease have been reported; however, the
effects were not robust to adjustment for
copollutants.
Mortality
No Studies
"suggestive but not sufficient to infer a
casual relationship"
Nonaccidental and No Studies
Cardiopulmonary
Mortality
Large multicity studies and a meta-analysis
study indicated positive and generally
robust associations between ambient N02
concentrations and risk of nonaccidental
and cardiopulmonary mortality.
March 2008
5-24
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-1 (cont'd). SUMMARY OF EVIDENCE FROM EPIDEMIOLOGICAL,
HUMAN CLINICAL, AND ANIMAL TOXICOLOGICAL STUDIES ON THE
HEALTH EFFECTS ASSOCIATED WITH SHORT- AND LONG-TERM
EXPOSURE TO NO2.
Health Outcome
Conclusion from Previous
NAAQS Review for NOX
Conclusion from 2008 NOX ISA
LONG-TERM EXPOSURE TO NO2
Respiratory
Morbidity
No Overall Conclusion.
"suggestive but not sufficient to infer a
casual relationship"
Respiratory Effects
At sufficiently high concentrations of N02
(i.e., >8 ppm) for long periods of exposure,
N02 can cause emphysema (meeting the
human definition criteria) in animals.
A number of epidemiological studies
observed decrements in lung function
growth associated with long-term exposure
to N02. These effects were observed in
studies with mean N02 concentrations in
the range of 21.5 to 34.6 ppb.
Other Morbidity No Studies.
"inadequate to infer the presence or
absence of a causal relationship"
Cancer
Cardiovascular
Effects
No Studies.
No Studies.
While limited epidemiological studies
observed an association between long-term
N02 exposure and incidence of cancer;
animal toxicological studies have not
provided clear evidence that N02 acts as a
carcinogen.
The very limited epidemiological and
toxicological evidence does not suggest that
long-term exposure to N02 has
cardiovascular effects.
Birth Outcomes
No Studies.
The epidemiological evidence for an
association between long-term exposure to
N02 and birth outcomes is generally
inconsistent, with limited support from
animal toxicological studies.
Mortality
No Studies.
"inadequate to infer the presence or
absence of a causal relationship"
Nonaccidental and No Studies.
Cardiopulmonary
Mortality
The results of epidemiological studies
examining the association between long-
term exposure to N02 and mortality were
generally inconsistent.
March 2008
5-25
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-4. LEGEND FOR FIGURE 5.3-1: SUMMARY OF
EPIDEMIOLOGIC STUDIES EXAMINING SHORT-TERM EXPOSURES TO
AMBIENT NO2 AND RESPIRATORY OUTCOMES
RESPIRATORY SYMPTOMS
Ref#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Reference
Schwartz et al. (1994)
Mortimer et al. (2002)
Schildcrout et al.
(2006)
Pino et al. (2004)
Ostro et al. (2001)
Ostro et al. (2001)
Delfmo et al. (2002)
Segalaetal. (1998)
Segalaetal. 1998
Just et al. (2002)
Jalaludin et al. (2004)
Segala et al. (2004)
von Klot et al. (2002)
von Klot et al. (2002)
von Klot et al. (2002)
von Klot et al. (2002)
Ward et al (2002)
Rodriguez et al. (2007)
Boezenetal. (1999)
Outcome Location
Cough Multicity-U.S.
Asthma symptoms Multicity-U.S.
Asthma symptoms Multicity-U.S.
Wheezy bronchitis Chile
Wheeze Southern CA
Cough Southern, CA
Asthma symptoms Southern CA
Asthma symptoms Paris, France
Cough Paris, France
Cough Paris, France
Cough Australia
Cough Paris, France
Wheeze Germany
Phlegm Germany
Cough Germany
Breathing problems Germany
Cough U.K.
Cough Perth, Australia
LRS Netherlands
Age
Children
Children
Children
Infants
Children
Children
Children
Children
Children
Children
Children
Adults
Adults
Adults
Adults
Adults
Children
Children
Children
Avg Time
24-h
4-h
24-h
24-h
1 -h max
1 -h max
8-h
24-h
24-h
24-h
15-h
24-h
24-h
24-h
24-h
24-h
24-h
24-h
24-h
Lag
1-4
1-6
0-2
3
3
3
0
0
3
0
0
0-4
0-4
0-4
0-4
0-4
0
0
0-4
Other
HOSPI'TAL
Ref#
Reference
HA/ED Location
Age
Avg Time
Lag
Other
Respiratory Disease - All Ages
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Tolbert et al. (2007)
Peel et al. (2005)
Luginaah et al. (2005)
Luginaah et al. (2005)
Anderson et al. (2001)
Atkinson etal., (1999a)
Atkinson etal., (1999b)
ED Atlanta
ED Atlanta
HA Windsor, ON
HA Windsor, ON
ED West Midlands, U.K.
HA London
ED London
Ponce de Leon et al. (1996) HA London
Llorca et al. (2005)
Oftedal et al. (2003)
Hagen et al. (2000)
Bedeschi et al. (2007)
Hinwood etal., (2006)
HA Torrelavega, Spain
HA Drammen, Norway
HA Drammen, Norway
HA Reggio Emilia, Italy
HA Perth, Australia
Petroeschevsky etal. (2001) HA Brisbane, Australia
All
All
All
All
All
All
All
All
All
All
All
All
All
All
1-h max
1-hmax
1 -h max
1 -h max
1 -h max
1-hmax
1-hmax
24-h
24-h
24-h
24-h
24-h
24-h
1 -h max
0-2
0-2
0-3
0-3
0-1
1
2
NR
3
0-3
3
1
1
Female
Male
Respiratory Disease - Children
34
35
36
37
38
Yang et al. (2003)
Luginaah et al. (2005)
Luginaah et al. (2005)
Anderson et al. (2001)
Atkinson et al. (1999a)
HA Vancouver, BC
HA Windsor, ON
HA Windsor, ON
HA West Midlands, U.K.
HA London
<3
0-14
0-14
0-14
0-14
24-h
1-h max
1-hmax
1 -h max
1-h max
1
0-3
0-3
0-1
2
Female
Male
March 2008
5-26
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-4 (cont'd). LEGEND FOR FIGURE 5.3-1: SUMMARY OF
EPIDEMIOLOGIC STUDIES EXAMINING SHORT-TERM EXPOSURES TO
AMBIENT NO2 AND RESPIRATORY OUTCOMES
Ref#
11 Ai, M»AIUsH>WED
Reference
'
HA/ED
Location
Age
Avg Time
Lag
Other
Respiratory Disease - Children (cont'd)
39
40
41
42
43
44
45
46
47
48
49
Atkinson et al. (1999b)
Ponce de Leon et al. (1996)
Vigotti et al. (2007)
Petroeschevsky et al. (2001)
Petroeschevsky et al. (2001)
Barnett et al. (2005)
Barnett et al. (2005)
Barnett et al. (2005)
Wongetal. (1999)
Linetal. (1999)
Gouveia and Fletcher (2000)
ED
HA
HA
HA
HA
HA
HA
HA
HA
ED
HA
London
London
Pisa, Italy
Brisbane, Australia
Brisbane, Australia
Multicity- Australia
Multicity- Australia
Multicity- Australia
Hong Kong
Sao Paulo, Brazil
Sao Paulo, Brazil
0-14
0-14
<10
0-4
5-14
0
1-4
5-14
0-4
<13
<5
1-h max
24-h
24-h
1 -h max
1 -h max
24-h
24-h
24-h
24-h
24-h
1-h max
1
2
0-2
3
0
0-1
0-1
0-1
0-3
0-4
0
Respiratory Disease - Adults
50
51
52
53
54
55
56
57
58
59
60
Luginaah et al. (2005)
Luginaah et al. (2005)
Spixetal. (1998)
Anderson et al. (2001)
Atkinson et al. (1999a)
Atkinson et al. (1999b)
Ponce de Leon et al. (1996)
Schouten et al. (1996)
Schouten et al. (1996)
Petroeschevsky et al.
(2001)
Wongetal. (1999)
HA
HA
HA
HA
HA
ED
HA
HA
HA
HA
HA
Windsor, ON
Windsor, ON
Multicity-Europe
West Midlands, U.K.
London
London
London
Amsterdam
Rotterdam
Brisbane, Australia
Hong Kong
15-64
15-64
15-64
15-64
15-64
15-64
15-64
15-64
15-64
15-64
5-64
1-h max
1-h max
24-h
1 -h max
1-h max
1-h max
24-h
24-h
24-h
24-h
24-h
0-3
0-3
1-3
0-2
1
2
1
1
1
0
0-3
Female
Male
Respiratory Disease - Older Adults (65+)
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
Luginaah et al. (2005)
Luginaah et al. (2005)
Fung et al. (2006)
Yang et al. (2003)
Spixetal. (1998)
Anderson et al. (2001)
Atkinson et al. (1999a)
Atkinson et al. (1999b)
Ponce de Leon et al. (1996)
Andersen et al. (2007b)
Andersen et al. (2007a)
Schouten et al. (1996)
Schouten et al. (1996)
Simpson et al. (2005)
Hinwood et al. (2006)
Petroeschevsky et al. (2001)
Wongetal. (1999)
HA
HA
HA
HA
HA
HA
HA
ED
HA
HA
HA
HA
HA
HA
HA
HA
HA
Windsor, ON
Windsor, ON
Vancouver, BC
Vancouver, BC
Multicity-Europe
West Midlands, U.K.
London
London
London
Copenhagen
Copenhagen
Amsterdam
Rotterdam
Multicity- Australia
Perth, Australia
Brisbane, Australia
Hong Kong
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
65+
1-h max
1-h max
24-h
24-h
24-h
1 -h max
1-h max
1-h max
24-h
24-h
24-h
24-h
24-h
1 -h max
24-h
24-h
24-h
0-3
0-3
0-3
1
1-3
0-2
3
0
2
0-4
0-4
2
0
0-1
1
5
0-3
Female
Male
March 2008
5-27
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-4 (cont'd). LEGEND FOR FIGURE 5.3-1: SUMMARY OF
EPIDEMIOLOGIC STUDIES EXAMINING SHORT-TERM EXPOSURES TO
AMBIENT NO2 AND RESPIRATORY OUTCOMES
HOSP
Ref#
Asthma -
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
Asthma
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
11 Ai, M»AIUsH>WED
Reference
- All Ages
Peel et al. (2005)
Itoetal. (2007)*
Burnett et al. (1999)
Anderson et al. (1998)
Atkinson et al. (1999a)
Atkinson et al. (1999b)
Galan et al. (2003)
Chardon et al. (2007)
Schouten et al. (1996)
Migliaretti et al. (2005)
Migliaretti and Cavallo (2004)
Hinwood et al. (2006)
Petroeschevsky et al. (2001)
WongetaL, (1999)
Tsai et al. (2006)
Tsai et al. (2006)
Yang et al. (2007)
Yang et al. (2007)
- Children
Peel et al. (2005)
Tolbert et al. (2000)
Lin et al. (2003)
Lin et al. (2003)
Sunyeretal. (1997)
Anderson et al. (1998)
Atkinson et al. (1999a)
Atkinson et al. (1999b)
Thompson et al. (2001)
Andersen et al. (2007b)
Andersen et al. (2007a)
Migliaretti et al. (2005)
Migliaretti and Cavallo (2004)
Migliaretti and Cavallo (2004)
Barnett et al. (2005)
Barnett et al. (2005)
Hinwood et al. (2006)
Petroeschevsky et al. (2001)
Petroeschevsky et al. (2001)
Morgan et al. (1998)
Ko et al. (2007)
Lee et al. (2006)
Gouveia and Fletcher (2000)
,
HA/ED
ED
ED
HA
HA
HA
ED
ED
HA
HA
ED
HA
HA
HA
HA
HA
HA
HA
HA
ED
ED
HA
HA
ED
HA
HA
ED
ED
HA
HA
ED
HA
HA
HA
HA
HA
HA
HA
HA
HA
HA
HA
(cont'd)
Location
Atlanta
New York, NY
Toronto
London
London
London
Madrid, Spain
Paris, France
Amsterdam
Turin, Italy
Turin, Italy
Perth, Australia
Brisbane, Australia
Hong Kong
Kaohsiung, Taiwan
Kaohsiung, Taiwan
Taipei, Taiwan
Taipei, Taiwan
Atlanta
Atlanta
Toronto
Toronto
Multicity-Europe
London
London
London
Belfast, Ireland
Copenhagen
Copenhagen
Turin, Italy
Turin, Italy
Turin, Italy
Multicity- Australia
Multicity- Australia
Perth, Australia
Brisbane, Australia
Brisbane, Australia
Sydney, Australia
Hong Kong
Hong Kong
Sao Paulo, Brazil
Age
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
2-18
0-16
6-12
6-12
0-14
0-14
0-14
0-14
<18
5-18
5-18
0-14
4-15
<4
1-4
5-14
0-14
0-4
5-14
1-14
0-14
<18
<5
Avg Time
1 -h max
24-h
24-h
24-h
1-h max
1-hmax
24-h
24-h
24-h
24-h
24-h
24-h
1 -h max
24-h
24-h
24-h
24-h
24-h
1 -h max
1-h max
24-h
24-h
24-h
24-h
1-h max
1-hmax
24-h
24-h
24-h
24-h
24-h
24-h
24-h
24-h
24-h
1 -h max
1 -h max
24-h
24-h
24-h
1-hmax
Lag
0-2
0-1
0
0-3
0
0
3
0-3
2
0-3
1-3
0
0-2
0-3
0-2
0-2
0-2
0-2
0-2
1
0-5
0-5
0-3
0-3
3
1
0-3
0-4
0-4
0-3
1-3
1-3
0-1
0-1
0
0
1
0
0-4
3
2
Other
Warm
Cool
Warm
Cool
Male
Female
March 2008
5-28
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-4 (cont'd). LEGEND FOR FIGURE 5.3-1: SUMMARY OF
EPIDEMIOLOGIC STUDIES EXAMINING SHORT-TERM EXPOSURES TO
AMBIENT NO2 AND RESPIRATORY OUTCOMES
••„•:/•' .'
Ref#
Asthma -
119
120
121
Asthma -
122
123
124
125
126
127
128
129
130
131
Asthma -
132
133
134
135
136
Reference
- Children (cont'd)
Jaffe et al. (2003)
Jaffe et al. (2003)
Linn et al. (2000)
Adults
Sunyeretal. (1997)
Anderson et al. (1998)
Atkinson et al. (1999a)
Atkinson et al. (1999b)
Boutin-Forzano et al. (2004)
Teniasetal. (1998)
Castellsague et al. (1995)
Migliaretti et al. (2005)
Morgan et al. (1998)
Ko et al. (2007)
Older Adults (65+)
Anderson etal. (1998)
Atkinson et al. (1999a)
Migliaretti et al. (2005)
Hinwood et al. (2006)
Ko et al. (2007)
HA/ED
ED
ED
HA
ED
HA
HA
ED
ED
ED
ED
ED
HA
HA
HA
HA
ED
HA
HA
Location
Cleveland
Cincinnati
Los Angeles
Multicity, Europe
London
London
London
Marseille, France
Valencia, Spain
Barcelona, Spain
Turin, Italy
Sydney, Australia
Hong Kong
London
London
Turin, Italy
Perth, Australia
Hong Kong
Age
5-34
5-34
>30
15-64
15-64
15-64
15-64
3-49
>14
15-64
15-64
15-64
15-64
65+
65+
65+
65+
65+
Avg Time
24-h
24-h
24-h
24-h
24-h
1-h max
1-hmax
24-h
24-h
24-h
24-h
24-h
24-h
24-h
1-hmax
24-h
24-h
24-h
Lag
1
1
0-1
0-3
0-1
1
1
0
0
0-2
0-3
0
0-4
0-3
3
0-3
0
0-4
Other
'.•,•'' r /.,.:•
Ref#
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
Reference
Ostro et al. (2000)
Location
Age
Coachella Valley, CA
Fairley (1999); (Reanalysis 2003)
Gamble (1998)
Gwynn et al. (2000)
Burnett et al. (2004)
Villeneuve et al. (2003)
Samoli et al. (2006)
Zmirouetal. (1998)
Biggeri et al. (2005)
Anderson et al. (1996)
Bremneretal. (1999)
Anderson et al. (2001)
Le Terte et al. (2002a)
Dab etal. (1996)
Zmirouetal. (1996)
Hoeketal. (2000); (Reanalysis,
Hoeketal. (2000); (Reanalysis,
Hoek (2003)
Hoek (2003)
Santa Clara County,
Dallas, TX
Buffalo, NY
Multicity-Canada
Vancouver, BC
Multicity-Europe
Multicity-Europe
Multicity-Italy
London, U.K.
London, U.K.
West Midlands, U.K
Multi city-France
Paris, France
Lyon, France
The Netherlands
The Netherlands
CA
Avg Time
24-h
24-h
24-h
24-h
24-h
24-h
1-hmax
24-h
24-h
24-h
24-h
1-h max
24-h
24-h
24-h
24-h
24-h
Lag
0
1
4-5
1
0-2
0
0-1
0-3
0-1
1
3
0-1
0-1
0
2
0-6
0-6
March 2008
5-29
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5.3-4 (cont'd). LEGEND FOR FIGURE 5.3-1: SUMMARY OF
EPIDEMIOLOGIC STUDIES EXAMINING SHORT-TERM EXPOSURES TO
AMBIENT NO2 AND RESPIRATORY OUTCOMES
Ref#
Reference
Location
Age
Avg Time
Lag
154 Saez et al. (2002)
155 Garcia-Aymerich et al. (2000)
156 Saez et al. (1999)
157 Sunyeretal. (1996)
158 Borja-Aburto et al. (1998)
159 Gouveia and Fletcher (2000b)
160 Simpson et al. (2005a,b)
161 Simpson et al. (2000)
162 Tsai et al. (2003)
163 Yang et al. (2004b)
164 Wong et al. (2001)
165 Wong et al. (2002)
Multicity-Spain
Barcelona, Spain
Barcelona, Spain
Barcelona, Spain
Mexico City, Mexico
Sao Paulo, Brazil
Multicity- Australia
Brisbane, Australia
Kaohsiumg, Taiwan
Taipei, Taiwan
Hong Kong, China
Hong Kong, China
2-45 yrs
65+
65+
24-h
24-h
24-h
1 -h max
24-h
1-hmax
1-hmax
24-h
24-h
24-h
24-h
24-h
0-3
0-1
0-2
0
1-5
2
0-1
0-1
0-2
0-2
0
0-1
March 2008
5-30
DRAFT-DO NOT QUOTE OR CITE
-------
APPENDIX 5A
March 2008 5A-1 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5A. EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
t-O
o
o
oo
>
I
IN3
O
H
6
O
2
o
m
o
?d
o
K^
H
m
Reference, Study
Location, and Period
Schwartz et al. (1994)
Six cities,
United States
1984-1988
Mortimer et al. (2002)
Eight urban areas,
United States
1993
Schildcrout et al.
(2006)
Eight North American
Cities
1993-1995
Ostroetal. (2001)
Los Angeles and
Pasadena, CA,
United States
Aug-Oct 1993
Study Population
1,844 elementary
school children in 6
U.S. cities
Asthmatic children
(4-9 yrs) from the
National Cooperative
Inner-City Asthma
Study (NCICAS) cohort
990 asthmatic children
(aged 5-13 yrs) enrolled
in Childhood Asthma
Management Program
(CAMP) cohort
138 African- American
asthmatic children
(8- 13 yrs)
Averaging Time,
Mean (SD) NO2
Levels (ppb)
24-havg: 13.3
4-havg: 32
24-havg: 17.8-26.0
L.A.: 1-hmax:
79.5 (43.6)
Pasadena: 1-hmax:
68.1 (31.3)
Statistics for NO2
Air Quality Data (ppb)
Standardized* Percent Excess Risk
98th % 99th % Range (95% CI)
NR NR Max: 44.2 Cough Incidence:
61.3% (8.2, 143.4)
NR NR -7,96 Morning Asthma Symptoms:
48% (2, 116)
NR NR NR Asthma Symptoms:
4.0% (1.0, 7.0)
Rescue Inhaler Use:
3.0% (1.0, 5.0)
NR NR L.A.: 20.0, Shortness of Breath:
220.0 Dayw/symptoms: 4.7% (-0.6,
Pasadena: Onset of symptoms: 8.2% (-0.
30.0, 170.0 wheeze:
Dayw/symptoms: 4.7% (1.2, £
Onset of symptoms: 7.6% (2.4
Cough:
Dayw/symptoms: 1.8% (-1.8,
Onset of symptoms: 7.0% (1.0
10.4)
6, 17.6)
5.7)
, 13.8)
5.3)
, 13.8)
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
Delfino et al. (2002)
Alpine, CA,
United States
Mar-Apr 1996
22 children with asthma
(9-19 yrs old) living in
nonsmoking households
1-h max: 24 (10)
NR
NR
Asthma Symptoms:
N02 Alone: 34.6% (-17.9, 122.1)
On Medication: - 8.9% (- 79.1, 297.6)
Not on Medication: 80.3%
(-10.7,263.7)
With (compared to without) Respiratory
Infection: 299% (-50.6, 1,708)
01
CO
O
H
6
O
2
o
H
m
o
?d
o
t—t
H
m
Delfino et al. (2003a)
East Los Angeles
County, CA,
United States
Nov 1999-Jan 2000
Adamkiewicz et al.
(2004)
Steubenville, OH,
United States
Sept-Dec 2000
Linnetal. (1996)
Los Angeles, CA,
United States
1992-1994
22 Hispanic school
children (ages 10-15)
with asthma
1-h max: 7.2 (2.1)
29 nonsmoking adults
(ages 53+)
24-havg: 10.9
269 school children
(during 4th and 5th
grade school years)
24-h avg: 33 (22)
NR NR 3,14 Asthma Symptoms:
Symptom Scores >1, lag 0:
119.7% (-45.8, 2,038.2)
Symptom Scores >1, lag 1:
197.4% (-36.7, 5,793.5)
Symptom Scores >2, lag 0:
360.6% (-95.8, 3,039,358)
Symptom Scores >2, lag 1:
-75.7% (-205.5, 138,807.3)
NR NR NR Change in Fraction of Exhaled NO:
24-h moving average: 0.53 ppb (-0.35,
1.41)
NR NR 1,96 Total Symptom Score:
Previous 24-h, Morning Score:
-18.2% (-47.3, 27.1)
Current 24-h, Evening Score:
-42.9% (-65.4,-5.9)
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
01
O
H
b
o
2
o
H
Segalaetal. (1998)
Paris, France
1992
84 children 7-15 yrs old
that had at least one
asthma attack in the
past 12 months
24-havg: 29.8 (8.1)
Just et al. (2002)
Paris, France
1996
82 children 7-15 yrs old
that had at least one
asthma attack in the
past 12 months
24-h avg: 28.2 (8.8)
NR NR 12.5,63.8 Mild Asthmatics:
Incident Episodes
Asthma 91% (13,223) lag 0
Cough 76% (21, 156) lag 4
Shortness of Breath 24% (-32, 125) lag 4
Respiratory Infections 88% (4, 243) lag 3
Moderate Asthmatics:
Incident Episodes
Asthma 31% (-16, 106) lag 3
Wheeze 26% (-7, 70) lag 0
Cough 39% (3, 87) lag 2
Shortness of Breath 18% (-6, 47) lag 4
Respiratory Infections 36% (-31, 168)
Iag3
NR NR 12.0,58.1 Incident Episodes
Asthma 82% (-39, 271) lag 0-2
Cough 82% (-11, 292) lag 0-2
Respiratory Infections 675% (4, 5719)
lag 0-2
m
o
?d
o
K^
H
m
-------
t-O
o
o
oo
>
I
en
O
H
6
O
2
o
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period Study Population
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) NO2
Levels (ppb) 98th % 99th % Range
Standardized* Percent Excess Risk
(95% CI)
Ward et al. (2002)
United Kingdom
1997
162 children 9 yrs old 24-h Median:
Winter: 18.0
Summer: 13.3
NR NR Winter:
4,35
Summer:
3,29
Winter
Cough 18% (-14,64) lag 2
Illness 18% (-1,40) lag 0
Shortness of Breath 7% (-13, 32)
Wheeze 12% (-13, 49) lag 3
lagO
Jalaludin et al. (2004)
Sydney, Australia
1994
148 children in 3rd to
5th grade with a history
of wheezing in the
previous 12 months
24-havg: 15 (6)
NR
NR
Max = 47
Summer
Cough 28% (3, 57) lag 0
Illness 3% (-24, 38) lag 0
Shortness of Breath 35% (-3, 85) lag 0
Wheeze-8% (-37, 31) lag 0
Wheeze 7% (-5, 21) lag 2
Dry Cough 7% (-9, 26) lag 0
Wet Cough 13% (0,26) lag 0
m
o
?d
o
K^
H
m
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
01
O
H
b
o
2
o
H
Boezenetal. (1999)
the Netherlands
(Rural: Bodegraven,
Meppel, Nunspeet;
Urban: Rotterdam,
Amsterdam)
1992-1995
632 children 7-1 lyrs
old
24-h avg:
Rural: 15.3
Urban: 25.6
NR
NR
NR
With Bronchial Hyperresponsiveness
(BHR) and High Serum Total IgE
Lower Respiratory Symptoms:
19% (3, 37) lag 0
Upper Respiratory Symptoms:
4% (-5, 13) lag 2
With BHR and Low Serum Total IgE
Lower Respiratory Symptoms:
-27%(-46,-l) lagO
Upper Respiratory Symptoms:
3% (-9, 16) lag 2
Without BHR or Low Serum Total IgE
Lower Respiratory Symptoms:
12% (-8, 37) lagO
Upper Respiratory Symptoms:
8%(-l, 17) lag 2
Without BHR or High Serum Total IgE
Lower Respiratory Symptoms:
4% (-14, 26) lagO
Upper Respiratory Symptoms:
9% (-3, 21) lagO
m
o
?d
o
K^
H
m
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
>
I
-<1
O
H
b
o
2
o
H
Van der Zee et al.
(2000)
the Netherlands
(Rural: Bodegraven,
Meppel, Nunspeet;
Urban: Rotterdam,
Amsterdam)
1992-1995
Harreetal. (1997)
Christchurch, New
Zealand
1994
Segala et al. (2004)
Paris, France
1999-2000
489 adults 50-70 yrs old
40 people >55 with
COPD
46 adult nonsmokers
18-64 yrs old
24-h Median:
Urban: 25.7
Nonurban: 12.3
NR
24-h avg: 30 (8.6)
NR NR NR Symptomatic Adults
Urban
Lower Respiratory Symptoms:
-4% (-13, 7) lag 0
Upper Respiratory Symptoms:
11% (1,22) lagO
Nonsymptomatic Adults
Urban
>10% PEF: 0% (-30, 46) lag 1
Upper Respiratory Symptoms:
0%(-14, 16) lagO
NR NR NR Morning Asthma Symptoms:
-2% (-4,0) lag 1
Evening Asthma Symptoms:
0% (-1,2) lag 1
Chest Symptoms:
140% (-66, 1634) lag 1
Wheeze:
91% (-47, 613) lagl
NR NR 11.5,70.1 Cough: 113% (0,358) lag 0-4
m
o
?d
o
K^
H
m
Higginsetal. (1995)
-------
t-O
o
o
oo
Oi
i
oo
O
H
6
O
2
o
H
m
o
?d
o
t—t
H
m
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period Study Population
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) NO2
Levels (ppb) 98th % 99th % Range
Standardized* Percent Excess Risk
(95% CI)
Desqueyroux et al.
(2002)
Paris, France
1995-1996
Desqueyroux et al.
(2002)
Paris, France
1995-1996
60 severe asthmatic
adults
24-havg: 28.3 (8.1) NR
39 adults with COPD 24-h avg: 28.3 (8.1)
NR
Boezenetal. (1998)
the Netherlands
(Meppel, Amsterdam)
1993-1994
189 adults 48-73 yrs old
24-h avg:
Urban: 24.1
Rural: 13.9
NR
NR 11.0,67.0 Incident Asthma Attacks
16% (-21,70) lagl
29% (-30, 134) lag 0-5
NR 11.0,67.0 Exacerbation of COPD
24-h avg:
8% (-39, 94) lagl
-24% (-73, 120) lag 0-5
1-h max:
12% (-70, 1378) lagl
12% (-78, 2599) lag 0-5
NR Urban: Without Bronchial Hyperresponsiveness
11.6, 39.7 Upper Respiratory Symptoms
Rural: 5% (-5, 16) lag 0
3.4, 28.4 Lower Respiratory Symptoms
1% (-11, 15) lagO
Cough: -2% (-11, 9) lag 0
Phlegm: 1% (-8, 11) lag 0
Hiltermann et al.
(1998)
the Netherlands
1995
60 nonsmoking adults 24-havg: 11.1
with intermittent to
severe persistent asthma
18-55 yrs old
NR NR 3.6,22.1 Shortness of Breath:
25% (0, 54 lag 0
Cough and/or Phlegm:
4% (-11, 25) lagl
Nasal Symptoms:
-14% (-33, 12) lagO
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
01
CD
O
Forsberg et al. (1998)
Landskrona, Sweden
Von Klot et al. (2002)
Erfurt, Germany
1996-1997
38 people with asthma
or asthma-like
symptoms >15 yrs old
53 adult asthmatics
24-havg: 16.0(7.0)
24-havg: 24.1
NR NR 3.0,37.5 Day:
Any Asthma: 17%
Severe Asthma: 127%
Evening:
Any Asthma: 19%
S evere Asthma: 13 4 %
NR NR 4.2,62.3 Wheeze:
!%(-5,7)lagO
8% (1, 15) lag 0-5
Shortness of Breath:
0%(-5,5)lagO
6%(-l, 14) lag 0-5
Phlegm:
5% (-1, 10) lagO
11 (5, 19) lag 0-5
H
b
o
o
H Pino et al. (2004) 504 infants
"O Santiago, Chile
§ 1995-1996
m
0
0
m
Cough:
3% (-3
8% (0,
, 8) lag 0
15) lag 0-5
24-havg: 41.1 (19.2) NR NR NR Wheezing Bronchitis:
14% (4
, 30) lag 6
-------
t-O
o
o
oo
TABLE 5A (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON RESPIRATORY OUTCOMES IN THE
UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized* Percent Excess Risk
(95% CI)
van der Zee et al.
(1999)
633 children 7-11 yrs
old with and without
respiratory symptoms
24-h avg:
Urban: 25.5
NR NR NR With Symptoms:
Lower Respiratory Symptoms:
11% (-7, 30) lag 2
Upper Respiratory Symptoms:
-2% (-11, 8) lag 2
Cough: 3% (-6, 12) lag 2
Without Symptoms:
Upper Respiratory Symptoms:
5% (-8, 19) lag 0
Cough: 1%(-11, 13) lag 2
>
*24-h avg NO? standardized to 20 ppb increment; 1-h max NO? standardized to 30 ppb increment
COPD = Chronic obstructive pulmonary disease.
NR = Not reported.
PEF = Peak expiratory flow.
O
H
6
O
2
o
m
o
?d
o
t—t
H
m
-------
O
to
O
O
oo
TABLE 5B. EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
^
>
H
6
o
o
H
O
O
H
W
O
HH
H
W
Reference, Study
Location, and Period
Emergency Department
Peel et al. (2005)
Study Population
Visits — All Respiratory
484,830 ED visits,
Atlanta, GA, United States all ages from 3 1
Jan 1993-Aug 2000
Stieb et al. (2000)
Saint John, New
Brunswick, Canada
Jul 1992-Mar 1996
Emergency Department
Jaffe et al. (2003) 2 cities,
OH, United States
(Cleveland, Cincinnati)
Jul 91 -Jim 96
NorrisT et al. (1999)
Seattle, WA, United
States, 1995-1996
Lipsettetal(1997)
Santa Clara County, CA,
United States,
1988-1992 (winter only)
hospitals
19,821 ED visits
Visits — Asthma
4,4 16 ED visits for
asthma, age 5-34
900 ED visits for
asthma, <18 yrs
ED visits for asthma
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) NO2 Standardized Percent Excess
Levels (ppb) 98th % 99th % Range Risk (95% CI)
1-h max: 45.9 52 59 Max: 256 1.024(1.009,1.041)
(17.3)
24-havg: 8.9 NR NR 0,82 -14.70%
24-havg: NR NR NR 6.1% (-2.0, 14.0)
Cincinnati:
50(15)
Cleveland:
48(15)
24-havg: NR NR NR 24-havg: -2.0% (-21, 19)
20.2(7.1) 1-havg: 5% (-2, 33)
1-h max:
34.0(11.3)
1-h max: 69(28) NR NR 29,150 48%
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
to
H
6
o
o
H
O
O
H
W
O
HH
H
W
Reference, Study
Location, and Period
Emergency Department
Peel et al. (2005)
Atlanta, GA,
United States
Jan 1993-Aug 2000
Sunyeretal. (1997)
Multi-city, Europe
(Barcelona, Helsinki,
Paris, London)
1986-1992
Atkinson etal. (1999b)
London, United Kingdom
1992-1994
Thompson etal. (2001)
Belfast, Northern Ireland
1993-1995
Boutin-Forzano et al.
(2004)
Marseille, France
1997-1998
Study Population
Visits — Asthma
Asthma ED visits,
all ages and 2-18 yrs
from 3 1 hospitals
ED visits for asthma
forages <15 and
15-64
98,685 all
respiratory and
asthma ED visits for
all ages, 0-14,
15-64, and 65+ from
12 hospitals
1,044 asthma ED
visits for children
549 asthma ED
visits for ages 3-49
Averaging time,
Mean (SD) NO2
Levels (ppb)
1-hmax: 45.9
(17.3)
24-havg: 24.1
1-hmax: 50.3
(17.0)
24-havg: 21.3
24-havg: 18.3
Statistics for NO2
Air Quality Data (ppb)
Standardized Percent Excess
98th % 99th % Range Risk (95% CI)
NR NR NR All Ages: 2.1% (-0.4, 4.5)
2-18 yrs: 4.1% (0.8, 7.6)
NR NR 2.6,181.7 <15 yrs: 3%(0, 5)
15-64 yrs: 3% (1,5)
NR NR NR All ages: 4% (1,6)
0-14 yrs: 7% (4, 11) lag 1
15-64 yrs: 4% (0, 7) lag 2
NR NR NR 25% (6, 44) lag 0-3
NR NR 1.6,44.5 3% (-2, 7) lag 0
-------
o
to
o
o
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
H
6
O
O
H
O
O
H
W
O
HH
H
W
Reference, Study
Location, and Period
Emergency Department
Castellsague et al. (1995)
Barcelona, Spain
1986-1989
Galan et al. (2003)
Madrid, Spain
1995-1998
Teniasetal. (1998)
Valencia, Spain
1993-1995
Migliaretti et al. (2005)
Turin, Italy
1997-1999
Kim et al. (2007)
Seoul, Korea
2002
Tolbert et al. (2000)
Atlanta, GA,
United States,
1993-1995
Study Population
Visits — Asthma (cont'd)
Asthma ED visits
for ages 15-64
4,827 asthma ED
visits for all ages
734 asthma ED
visits forages >14
1,401 asthma ED
visits forages <15,
15-64, and >64 and
201,071 controls
92,535 asthma ED
visits for all ages
5,934 ED visits for
asthma, age 0-16
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) NO2
Levels (ppb) 98th % 99th % Range
24-havg: 26.8 NR NR NR
24-havg: 35.1 NR NR Max: 77.2
(9.4)
24-havg: 30.2 NR NR NR
1-hmax: 52.9
24-havg: 59 NR NR NR
(15.8)
24-havg: 36.0 NR NR 2.3,108.0
(14.7)
1-hmax: 81.7 NR NR 5.35,306
(53.8)
Standardized Percent Excess
Risk (95% CI)
11% (2, 22) lag 0-2
13% (5, 22) lag 3
24-h avg: 33% (8, 62) lag 0
1-hmax: 23% (5, 45) lag 0
All ages; 10% (2, 18) lag 0-3
0-14 yrs: 9%(1, 18) lag 0-3
15-64 yrs: 12% (0, 33) lag 0-3
>65yrs: 33% (1, 72) lag 0-3
0.7% (-0.8, 2.3)
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Emergency Department Visits—Asthma (cont'd)
CassinoT et al. (1999)
New York City, NY,
United States
1989-1993
1,115 ED visits from
11 hospitals
24-havg: 45.0
NR
NR
NR
lagO: -4% (-19, 12)
lagl: 5% (-11, 25)
lag 2: 9% (-8, 28)
>
^
H
6
o
o
H
O
O
H
W
Stiebetal. (1996)
St. John, New Brunswick,
Canada
1984-1992
(summers only)
1,163 ED visits for
asthma, ages 0-15,
15+ from 2 hospitals
1-hmax: 25.2
NR
NR
0, 120
,: -11%
Hospital Admissions — All
GwynnT et al. (2000)
Buffalo, NY,
United States,
1988-1990, Days: 1,090
Burnett etal. (1997a)
16 Canadian Cities,
Canada,
4/1981-12/1991,
Days: 3,927
Respiratory
Respiratory hospital 24-havg: 20.5 NR NR
admissions
All respiratory 1-hmax: 35.5 NR 87
admission from (16.5)
134 hospitals
4.0, 47.5 2.20%
NR Only report results or
multipollutant model adjusted
for
CO, O3, SO2 and CoH: -0.3%
(-2.4%, 1.8%)
O
HH
H
W
-------
O
to
O
O
oo
>
(^
H
6
o
o
H
O
O
H
W
O
HH
H
W
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Hospital Admissions — All
Yang et al. (2003)
Vancouver, BC,
Canada 1986-1998,
Days: 4,748
Fung et al. (2006)
Vancouver, BC,
Study Population
Respiratory (cont'd)
Respiratory hospital
admissions among
young children
(<3 yrs) and elderly
(>65 yrs)
All respiratory
admissions for
Averaging Time,
Mean (SD) NO2
Levels (ppb)
24-h avg:
18.74(5.66)
24-h avg:
16.83 (4.34)
Statistics for
NO2
Air Quality Data (ppb)
98th % 99th %
NR NR
NR NR
Range
NR
7.22,
33.89
Standardized Percent Excess
Risk (95% CI)
<3yrs: 19.1% (7.4, 36.3)
>65yrs: 19.1% (11.2, 36.
9.1% (1.5, 17.2)
3)
Canada
6/1/95-3/31/99
BurnettTetal. (2001)
Toronto, ON,
Canada
1980-1994
Luginaah et al. (2005)
Windsor, ON,
Canada
4/1/95-12/31/00
elderly (65+ yrs)
All respiratory
admissions for
young children
(<2 yrs)
All respiratory
admissions ages
0-14, 15-64, and
65+ from
4 hospitals
1-hmax: 44.1
NR 86
Max =146 18.20%
1-hmax:
38.9(12.3)
NR NR
NR
All ages, female:
6.7% (-5.4, 20.4)
All ages, male:
-10.3% (-20.3, 1.1)
0-14, female: 22.4% (-1.2, 51.5)
0-14, male: -8.3% (-13.7, 0.8)
15-64, female: 23.9% (-4.1, 60.0)
15-64, male: 2.3% (-17.7, 44.3)
65+, female: 3.8% (- 12.8, 23.5)
65+, male: -14.6 (-29.2, 3.0)
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions—All Respiratory (cont'd)
Simpson et al. (2005a)
Multicity study, Australia
(Sydney, Melbourne,
Brisbane, Perth)
1996-1999
All respiratory,
asthma, and
pneumonia with
bronchitis hospital
admissions for ages
15-64 and 65+years
1-h max: 22
NR
NR
NR
>65yrs: 8% (5, 12) lag 0-1
H
6
o
o
H
O
O
H
W
Barnett et al. (2005)
Multicity, Australia/New
Zealand (Auckland,
Brisbane, Canberra,
Christchurch, Melbourne,
Perth, Sydney)
1998-2001
All respiratory,
asthma, and
pneumonia with
bronchitis hospital
admissions for ages
0, 1-4, and 5-14
24-havg: 10
1-h max: 19.1
NR
NR
NR
24-h avg:
Oyrs: 13% (-4, 32) lag 0-1
1-4 yrs: 10% (-3, 24) lag 0-1
5-14yrs: 25% (7, 46) lag 0-1
1-h max:
Oyrs: 8% (-5, 22) lag 0-1
1-4 yrs: 10% (2, 17) lag 0-1
5-14 yrs: 17% (5, 29) lag 0-1
Hinwood et al. (2006)
Perth, Australia
1992-1998
COPD, pneumonia,
and asthma hospital
admissions for all
ages, <15, and 65+
24-havg: 10.3 NR NR NR
(5.0)
All ages: 4% (-4, 8) lag 1
>65yrs: 10% (2, 24) lag 1
O
HH
H
W
-------
O
to
O
O
oo
H
6
o
o
H
O
O
H
W
O
HH
H
W
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Hospital Admissions — All
Petroeschevsky et al.
(2001)
Brisbane, Australia
1987-1994
Schouten et al. (1996)
Multicity, the Netherlands
(Amsterdam, Rotterdam)
Study Population
Respiratory (cont'd)
All respiratory
(3 3, 7 10) and asthma
(13,246) hospital
admissions for all
ages, 0-4, 5-14,
1-S-64 anH 65+
_L ^} \J^ - ClllLl \J^J i
All respiratory,
asthma, and COPD
hospital admissions
Averaging Time,
Mean (SD) NO2
Levels (ppb)
24-havg: 139
1-hmax: 282
24-h avg:
Amsterdam: 26.2
Rotterdam: 28.3
Statistics for NO2
Air Quality Data (ppb)
98th % 99th % Range
NR NR 24-h avg:
12, 497
1-h max:
35, 1558
NR NR NR
Standardized Percent Excess
Risk (95% CI)
24-h avg:
15-64 yrs: 5% (-3, 15) lag 0
>65yrs: - 18% (-28, -8) lag 5
1-h max:
All ages: -3% (-7, 1) lag 1
0-4 yrs: 5%(-l, 11) lag 3
5-14 yrs: -4% (- 14, 6) lag 0
24-hr avg:
Amsterdam:
15-64 yrs: -4% (-9,0) lag 1
1977-1989
for all ages, 15-64,
and 65+
1-h max:
Amsterdam:
39.3
Rotterdam: 42.9
>65yrs: 1% (-4, 6) lag 2
Rotterdam (1985-1989):
15-64 yrs: -l%(-7,4)lag 1
>65yrs: 6% (0, 13) lag 0
1-h max::
Amsterdam:
15-64 yrs: -6% (-11,-2) lag 1
>65yrs: 1% (-5%, 5) lag 2
Rotterdam:
15-64 yrs: 2% (-3, 7) lag 1
>65yrs: 4% (-2, 10) lag 0
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — All Respiratory (cont'd)
Ponce de Leon et al. 19,901 all
(1996) respiratory hospital
London, England admissions for all
1987-1988; ages, 0-14, 15-64,
1991-02/1992 and 65+
24-havg: 37.3 NR NR NR
(13.8)
All ages: 1% (0,2) lag 2
0-14 yrs: I%(0,2)lag2
15-64 yrs: !%(-!, 2) lag 1
>65yrs: 2% (0, 3) lag 2
oo
Atkinson et al. (1999a)
London, England
1992-1994
165,032 all
respiratory, asthma,
asthma + COPD,
lower respiratory
disease hospital
admissions for all
ages, 0-14, 15-64,
and 65+
1-hmax: 50.3
(17.0)
22.0, All ages: 1% (0, 3) lag 1
224.3 0-14 yrs: 2% (0, 4) lag 2
15-64 yrs: !%(-!, 3) lag 1
>65yrs: 2% (0, 4) lag 3
H
6
o
o
H
O
O
H
W
Spixetal. (1998)
Multicity (London,
Amsterdam, Rotterdam,
Paris), Europe
1977 + 1991
All respiratory and
asthma hospital
admissions for ages
15-64 and 65+
24-h avg:
London: 18.3
Amsterdam: 26.2
Rotterdam: 27.7
Paris: 22.0
NR
NR
NR
15-64 yrs: !%(-!, 3)
>65yrs: !%(-!, 5)
O
HH
H
W
-------
O
to
O
O
oo
VO
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Hospital Admissions — All
WongT et al. (2002)
London England and
Hong Kong
London: 1992-1994
Hong Kong: 1995-1997
Anderson et al. (2001)
West Midlands
Study Population
Respiratory (cont'd)
All respiratory and
asthma hospital
admissions for all
ages, 15-64, and
65+
All respiratory,
asthma, and COPD
Averaging Time,
Mean (SD) NO2
Levels (ppb)
24-h avg:
Hong Kong: 29.3
(10.2)
London: 33.7
(10.7)
1-h max: 37.2
(15.1)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th % Range
NR NR Hong
Kong:
15.3,
151.5
London:
23.7,
255.8
NR NR 10.7,
176.1
Standardized Percent
Risk (95% CI)
>65 yrs
Hong Kong:
7% (5, 9) lag 0-1
5% (3, 7) lag 0
London:
0% (-2, 2) lag 0-1
3% (2, 5) lag 3
Excess
All ages: 2% (0, 4) lag 0-1
0-14 yrs: 3% (-1,6) lag
0-1
conurbation, United
Kingdom
1994-1996
hospital admissions
for all ages, 0-14,
15-64, and 65+
15-64 yrs: 0% (-4, 4) lag 0-1
>65yrs: 1% (-2, 5) lag 0-1
H
6
o
o
H
O
O
H
W
O
HH
H
W
Prescott et al. (1998)
Edinburgh, United
Kingdom
1992-1995
All respiratory
hospital admissions
(i.e., Pneumonia and
COPD + asthma)
for ages <65 and
65+
24-h avg: 26.4(7) NR
NR 9,58 >65yrs: 6% (-9,24)
rolling 3-day avg
<65yrs: 0%(-14, 16)
rolling 3-day avg
Hagen et al. (2000)
Drammen, Norway
1994-1997
All respiratory
admissions for all
ages at 1 hospital
24-h avg: 18.9 NR NR NR
(8.4)
14% (-1,31) lag 0-3
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
to
O
H
6
o
o
H
O
O
H
W
Reference, Study
Location, and Period
Hospital Admissions — All
Oftedal et al. (2003)
Drammen, Norway
1994-2000
Andersen et al. (2007a)
Copenhagen, Germany
1999-2004
Andersen et al. (2007b)
Copenhagen, Germany
2001-2004
DabTetal. (1996)
Paris, France
1987-1992
Averaging Time,
Mean (SD) NO2
Study Population Levels (ppb)
Respiratory (cont'd)
All respiratory 24-h avg: 17.7
admissions for all (8.4)
ages
Chronic bronchitis, 24-h avg: 12 (5)
emphysema, COPD,
and asthma hospital
admissions for ages
5-18, and 65+
Chronic bronchitis, 24-h avg: 11(5)
emphysema, COPD,
and asthma hospital
admissions for ages
5-18, and 65+
All respiratory, 24-h avg: 23.6
asthma, and COPD
hospital admissions 1-hmax: 38.6
for all ages at 27
hospitals
Statistics for NO2
Air Quality Data (ppb)
98th % 99th % Range
NR NR NR
NR NR NR
NR NR NR
NR 24-h avg: NR
56.7
1-hmax:
106.1
Standardized Percent Excess
Risk (95% CI)
11% (3, 20) lag 3
>65 yrs: 12% (3, 22) lag 5 day
moving avg
>65yrs: 21% (3, 46) lag 0-4
moving avg
24-h avg: 2% (0, 3) lag 0
1-hmax: 1%(0, 2) lag 0
O
HH
H
W
-------
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
to
o
o
oo
i
to
o
H
6
o
0
H
O
o
w
o
HH
H
W
Reference, Study
Location, and Period
Hospital Admissions — All
Llorca et al. (2005)
Torrelavega, Spain
1992-1995
Farchi et al. (2006)
Rome, Italy
1994-1995
FuscoTetal. (2001)
Rome, Italy
1995-1997
Pantazopoulou et al.
(1995)
Athens, Greece
1988
Gouveia and Fletcher,
(2000)
Sao Paulo, Brazil
1992-1994
Study Population
Respiratory (cont'd)
All respiratory
hospital admissions
for all ages at
1 hospital
2,947 all respiratory
hospital admissions
for ages 6-7
All respiratory,
asthma, COPD, and
respiratory infection
hospital admissions
for all ages and 0-14
15,236 all
respiratory hospital
admissions for all
ages at 14 hospitals
All respiratory,
pneumonia, and
asthma or bronchitis
hospital admissions
for ages <1 and <5
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) JNO2 Standardized Percent Excess
Levels (ppb) 98th % 99th % Range Risk (95% CI)
24-havg: 11.2 NR NR NR 18% (12, 24)
(8.6)
24-havg: 24.6 NR NR 12.6,34.6 157% (-7, 624)
(5.3)
24-havg: 45.4 NR NR NR All ages: 4% (2, 7) lag 0
(8.5) 0-14 yrs: 7%(1, 13) lag 0
24-havg: NR NR NR Winter: 11% (3, 20)
Winter: 49.2 Summer: 3% (-5, 8)
(13.1)
Summer: 58.1
(16.8)
1-hmax: 91.3 NR NR 13.6, <5 yrs: 1% (0,2) lag 0
(53.0) 362.8
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — All Respiratory (cont'd)
BragaTetal. (2001)
Sao Paulo, Brazil
1993-1997
All respiratory
hospital admissions
for ages 0-1 9, <2,
3-5, 6-13, and 14-19
24-havg: 74.0 NR NR 13.1,
(37.3) 341.4
<2yrs: 7% (4, 9) lag 5
3-5 yrs: l%(-5,7)
6-13 yrs: 2% (-4, 7)
14-19 yrs: -2% (-11, 7)
0-1 9 yrs: 5% (2, 7)
to
to
Wong etal. (1999)
Hong Kong, China
1994-1995
All respiratory,
asthma, COPD, and
pneumonia hospital
admissions for all
ages, 0-4, 5-64, and
65+ at 12 hospitals
24-havg: 26.9
NR NR 8.6,64.1 0-4 yrs: 8% (4, 12) lag 0-3
5-64 yrs: 9% (4, 14) lag 0-3
>65yrs: 10% (5, 14) lag 0-3
Hospital Admissions—Asthma
H
6
o
o
H
O
O
H
W
Linn et al. (2000)
Los Angeles, CA,
United States 1992-1995
302,600 COPD and
asthma hospital
admissions
24-h avg:
Winter: 3.4(1.3);
Spring: 2.8(0.9);
Summer: 3.4 (1.0);
Autumn: 4.1 (1.4);
allyr: 3.4(1.3)
NR
NR
NR
2.8% ± 1.0%
O
HH
H
W
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
LinT et al. (2004)
Vancouver, BC, Canada
1987-1991
Asthma hospital 24-h avg:
admissions among 18.65(5.59)
6-12yr olds
NR NR 4.28, Boys, low SES:
45.36 45.3% (12.7, 88.3)
Boys, high SES:
12.7% (-14.6, 49.3)
Girls, low SES:
23.0% (-11. 7, 70.2)
Girls, high SES:
3.1% (-27.6, 45.3)
to
Lin et al. (2003)
Toronto, ON, Canada
1981-1993
Asthma hospital
admissions among
6-12yr olds
24-h avg: 25.24
(9.04)
NR NR 3.0,82.0 Boys: 18.9% (1.8,39.3)
Girls: 17.0% (-5.4, 41.4)
H
6
o
o
H
O
O
H
W
O
HH
H
W
Burnett etal. (1999)
Toronto, ON, Canada
1980-1994
Barnett et al. (2005)
Multicity, Australia/New
Zealand; (Auckland,
Brisbane, Canberra,
Christchurch, Melbourne,
Perth, Sydney)
1998-2001
Asthma hospital
admissions
All respiratory,
asthma, and
pneumonia with
bronchitis hospital
admissions for ages
0, 1-4, and 5-14
24-h avg: 25.2 NR NR NR
(9.1)
24-h avg: 8 NR NR NR
1-hmax: 19.1
2.60%
24-h avg:
1-4 yrs: 11% (-5, 28) lag 0-1
5-14 yrs: 26% (1, 57) lag 0-1
1-h max:
1-4 yrs: 9%(-l, 18) lag 0-1
5-14 yrs: 9% (-7, 28) lag 0-1
-------
O
to
O
O
oo
to
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th % Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
Erbas et al. (2005)
Melbourne, Australia
2000-2001
Hinwood et al. (2006)
Perth, Australia
8, 955 asthma
hospital admissions
among 1-15 yr olds
for 6 hospitals
COPD, pneumonia,
and asthma hospital
24-h avg: 16.80
(8.61)
24-h avg: 10.3
(5.0)
NR NR 2.43,
63.00
NR NR NR
Inner Melbourne:
- 14% (-26, -2) lag 0
Western Melbourne:
10% (2, 18) lag 2
Eastern Melbourne:
8% (-8, 25) lag 0
South/Southeastern Melbourne:
-2% (-23, 21) lag 1
All ages: 2% (-2, 6) lag 0
0-14 yrs: 4% (-4, 8) lag 0
1992-1998
admissions for all
ages, <15, and 65+
>65yrs: -8% (-11, 4) lag 0
H
6
o
o
H
O
O
H
W
Petroeschevsky et al.
(2001)
Brisbane, Australia
1987-1994
All respiratory 1-hmax: 282
(3 3, 7 10) and asthma
(13,246) hospital
admissions for all
ages, 0-4, 5-14,
15-64, and 65+
NR NR 1-hmax:
35, 1558
All ages: - 1 1% (- 18, -3) lag 0-2
0-4 yrs: -7% (-15, 1) lag 0
5-64 yrs: -5% (- 15, 5) lag 1
O
HH
H
W
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
Morgan etal. (1998a)
Sydney, Australia
1990-1994
COPD and asthma
hospital admissions
forages 1-14,
15-64, 65+, all ages
for 27 hospitals
24-havg: 15(6) NR NR
1-hmax: 29(3)
24-h avg:
0,52
1-h max:
0, 139
24-h avg:
1-14 yrs: 4% (-2, 10) lag 0
15-64 yrs: 3% (-3, 9) lag 0
1-h max:
1-14 yrs: 5%(1, 10) lag 0
15-64 yrs: 3% (2, 8) lag 0
to
Sunyeretal. (1997)
Multicity, Europe
(Barcelona, Helsinki,
Paris, London)
1986-1992
Asthma hospital
admissions for ages
<15 and 15-64
24-havg: 24.1
NR
NR
NR
<15 yrs: 3% (0, 5) lag 0-3,
cumulative
15-64 yrs: 3% (1, 5) lag 0-3,
cumulative
H
6
o
o
H
O
O
H
W
Schouten etal. (1996)
Multicity, the Netherlands
(Amsterdam, Rotterdam)
1977-1989
All respiratory,
asthma, and COPD
hospital admissions
for all ages, 15-64,
and 65+
24-h avg:
Amsterdam: 26.2
Rotterdam: 28.3
1-h max:
Amsterdam: 39.3
Rotterdam: 42.9
NR
NR
NR
24-h avg:
Amsterdam:
All ages: 2% (-4, 10) lag 2
O
HH
H
W
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions—Asthma (cont'd)
Atkinson etal. (1999a)
London, England
1992-1994
165,032 all
respiratory, asthma,
asthma + COPD,
lower respiratory
disease hospital
admissions for all
ages, 0-14, 15-64,
and 65+
1-hmax: 50.3
(17.0)
22.0, All ages: 1% (-1,4) lag 0
224.3 0-14 yrs: !%(-!, 5) lag 3
15-64 yrs: 4% (1,8) lag 1
>65yrs: 4% (-2, 10) lag 3
to
H
6
o
o
H
O
O
H
W
O
HH
H
W
WongT et al. (2002)
London England and
Hong Kong
London: 1992-1994
Hong Kong: 1995-1997
Anderson etal. (1998)
London, England
1987-1992
Thompson et al. (2001)
Belfast, Northern Ireland
1993-1995
All respiratory and
asthma hospital
admissions for all
ages, 15-64, and
65+
Asthma hospital
admissions for all
ages, <15, 15-64,
and 65+
1,095 asthma
hospital admissions
forages 0-14
24-h avg:
Hong Kong: 29.3
(10.2)
London: 33.7
(10.7)
NR
NR
24-h avg:
(12.3)
37.2
NR
NR
24-h avg: 21.3
NR
NR
Hong
Kong:
15.3,
151.5
London:
23.7,
255.8
24-h avg:
14, 182
13,28
15-64 yrs:
Hong Kong:
-2% (-8, 4) lag 0-1
-5% (-10,0) lag 1
London:
4% (0,8) lag 0-1
4% (1,8) lag 2
All ages: 4% (2, 6) lag 0-3
0-14 yrs: 4% (1, 6) lag 0-3
15-64 yrs: 2% (-1, 7) lag 0-1
>65yrs: 6% (0, 13) lag 0-3
25% (6, 44) lag 0-3
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th %
Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
P6nka(1991)
Helsinki, Finland
1987-1989
Asthma hospital
admissions for ages
0-14, 15-64, and
65+
24-havg: 20.2
(8.5)
NR
NR
2.1,88.8
Andersen et al. (2007a)
Copenhagen, Germany
1999-2004
to
Chronic bronchitis,
emphysema, COPD,
and asthma hospital
admissions for ages
5-18, and 65+ at 9
hospitals
24-havg: 12(5) NR
NR NR 5-18 yrs: 41% (9, 83) lag 6 day
moving avg
H
6
O
O
H
O
O
H
W
Andersen et al. (2007b)
Copenhagen, Germany
2001-2004
Chronic bronchitis,
emphysema, COPD,
and asthma hospital
admissions for ages
5-18, and 65+ at 9
hospitals
24-havg: 11(5) NR
NR NR 5-18 yrs: 14% (-24, 74) lag 0-5
moving avg
DabTetal. (1996)
Paris, France
1987-1992
All respiratory,
asthma, and COPD
hospital admissions
for all ages at 27
hospitals
24-h avg:
1-h max:
23.6 NR
38.6
24-h avg: NR
56.7
1-h max:
106.1
24-h avg:
1-h max:
: 6% (2, 11) lag 0-1
5% (1, 8) lag 0-1
O
HH
H
W
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
oo
H
6
o
o
H
O
O
H
W
Reference, Study
Location, and Period
Study Population
Statistics for NO2
Averaging Time, Air Quality Data (ppb)
Mean (SD) NO2
Levels (ppb) 98th % 99th % Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
Migliaretti and Cavallo
(2004)
Turin, Italy
1997-1999
FuscoTetal. (2001)
Rome, Italy
1995-1997
Gouveia and Fletcher,
(2000)
Sao Paulo, Brazil
1992-1994
Lee et al. (2006)
Hong Kong, China
1997-2002
734 asthmatics age
matched (<4 or 4-15
yrs) with 25,523
other respiratory
disease controls
All respiratory,
asthma, COPD, and
respiratory infection
hospital admissions
for all ages and 0-14
All respiratory,
pneumonia, and
asthma or bronchitis
hospital admissions
for ages <1 and <5
26,663 asthma
hospital admissions
forages <18
24-havg: 59.3 NR NR NR
24-havg: 45.4 NR NR NR
(8.5)
1-hmax: 91.3 NR NR 13.6,
(53.0) 362.8
24-havg: 33.9 NR NR NR
(10.9)
All ages: 11% (0, 17) lag 1-3
cumulative
<4yrs: 11% (0,21) lag 1-3
cumulative
4-15 yrs: 11% (0,25) lag 1-3
cumulative
All ages: 8%(-l, 18) lag 0
0-14 yrs: 19% (5, 35) lag 1
<5yrs: 2% (-1,5) lag 2
13% (10, 16) lag 3
O
HH
H
W
-------
O
to
O
O
oo
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
H
6
o
o
H
O
O
H
W
Reference, Study
Location, and Period
Study Population
Averaging Time,
Mean (SD) NO2
Levels (ppb)
Statistics for NO2
Air Quality Data (ppb)
98th % 99th % Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
Tsai et al. (2006)
Kaohsiung, Taiwan
1996-2003
LeeT et al. (2002)
Seoul, Korea
1997-1999
Yang et al. (2007)
Taipei, Taiwan
1996-2003
Ko et al. (2007)
Hong Kong, China
2000-2005
Lee et al. (2006)
Hong Kong, China
1997-2002
17,682 asthma
hospital admissions
for all ages
6,436 asthma
hospital admissions
forages <15
25,602 asthma
hospital admissions
for all ages at 47
hospitals
69, 176 asthma
hospital admissions
for all ages at 15
hospitals
26,663 asthma
hospital admissions
forages <18
24-havg: 27.2
(17)
24-havg: 31.5
(10.3)
24-havg: 30.77
24-havg: 27.9
(10.1)
24-havg: 33.9
(10.9)
NR NR 4.83, 63.4
NR NR NR
NR NR 3.84,
77.97
NR NR 6.96, 78.3
NR NR NR
>25°C: 31% (13, 52) lag 0-2
<25°C: 142% (109, 179) lag 0-2
21% (14, 28) lag 0-2
>25°C: 39% (24, 55) lag 0-2
<25°C: 27% (16, 3 9) lag 0-2
All ages: 11% (8, 14) lag 0-4
0-14 yrs: 16% (11, 21) lag 0-4
15-64 yrs: 7% (3, 12) lag 0-4
>65yrs: 9% (5, 13) lag 0-4
13% (10, 16) lag 3
O
HH
H
W
-------
O
to
O
O
oo
H
6
o
o
H
O
O
H
W
TABLE 5B (cont'd). EFFECTS OF SHORT-TERM NO2 EXPOSURE ON EMERGENCY DEPARTMENT VISITS AND
HOSPITAL ADMISSIONS FOR RESPIRATORY OUTCOMES IN THE UNITED STATES AND CANADA
Averaging Time,
Statistics for NO2
Air Quality Data (ppb)
Reference, Study
Location, and Period
Study Population
Mean (SD) NO2
Levels (ppb) 98th % 99th % Range
Standardized Percent Excess
Risk (95% CI)
Hospital Admissions — Asthma (cont'd)
Wong etal. (1999)
Hong Kong, China
1994-1995
Wong etal. (2001)
Hong Kong, China
1993-1994
All respiratory,
asthma, COPD, and
pneumonia hospital
admissions for all
ages, 0-4, 5-64, and
65+ at 12 hospitals
1,217 asthma
hospital admissions
forages < 15 at 1
hospital
24-havg: 26.9 NR NR 8.6,64.1
24-havg: 22.7 NR NR 4.7,55.5
(8.7)
All ages: 10% (4, 17) lag 0-3
34%
*24-h avg NO2 standardized to 20 ppb increment; 1-h max NO2 standardized to 30 ppb increment.
TGAM impacted study.
CoH = Coefficient of haze.
COPD = Chronic obstructive pulmonary disease.
NR = Not reported.
SES = Socioeconomic status.
O
HH
H
W
-------
6. REFERENCES
2
3
4 Abbey, D. E.; Nishino, N.; McDonnell, W. F.; Burchette, R. J.; Knutsen, S. F.; Beeson, W. L.;
5 Yang, J. X. (1999) Long-term inhalable particles and other air pollutants related to
6 mortality in nonsmokers. Am. J. Respir. Crit. Care Med. 159: 373-382.
7 Ackermann-Liebrich, U.; Leuenberger, P.; Schwartz, J.; Schindler, C.; Monn, C.; Bolognini, B.;
8 Bongard, J. P.; Brandli, 0.; Domenighetti, G.; Elsasser, S.; Grize, L.; Karrer, W.; Keller,
9 R.; Keller-Wossidlo, H.; Kiinzli, N.; Martin, B. W.; Medici, T. C.; Perruchoud, A. P.;
10 Schoni, M. H.; Tschopp, J. M.; Villiger, B.; Wiithrich, B.; Zellweger, J. P.; Zemp, E.
11 (1997) Lung function and long term exposure to air pollutants in Switzerland. Am. J.
12 Respir. Crit. Care Med. 155: 122-129.
13 Adamkiewicz, G.; Ebelt, S.; Syring, M.; Slater, J.; Speizer, F. E.; Schwartz, J.; Suh, H.; Gold, D.
14 R. (2004) Association between air pollution exposure and exhaled nitric oxide in an
15 elderly population. Thorax 59: 204-209.
16 Adams, W. C.; Brookes, K. A.; Schelegle, E. S. (1987) Effects of N02 alone and in combination
17 with 03 on young men and women. J. Appl. Physiol. 62: 1698-1704.
18 Adams, P. F.; Hendershot, G. E.; Marano, M. A. (1999) Current estimates from the National
19 Health Interview Survey, 1996. Hyattsville, MD: U.S. Department of Health and Human
20 Services, Public Health Service, National Center for Health Statistics; DHHS publication
21 no. (PHS) 99-1528. (Vital and health statistics: v. 10, data from the National Health
22 Survey, no. 200). Available:
23 http://www.cdc.gov/nchs/products/pubs/pubd/series/srlO/pre-200/pre-200.htm [12
24 March, 2001].
25 Adkins, B., Jr.; Van Stee, E. W.; Simmons, J. E.; Eustis, S. L. (1986) Oncogenic response of
26 strain A/J mice to inhaled chemicals. J. Toxicol. Environ. Health 17: 311-322.
27 Aga, E.; Samoli, E.; Touloumi, G.; Anderson, H. R.; Cadum, E.; Forsberg, B.; Goodman, P.;
28 Goren, A.; Kotesovec, F.; Kriz, B.; Macarol-Hiti, M.; Medina, S.; Paldy, A.; Schindler,
29 C.; Sunyer, J.; Tittanen, P.; Wojtyniak, B.; Zmirou, D.; Schwartz, J.; Katsouyanni, K.
30 (2003) Short-term effects of ambient particles on mortality in the elderly: results from 28
31 cities in the APHEA2 project. Eur. Respir. J. 21 (suppl. 40): 28s-33s.
32 Alcorn, S. H.; Lurmann, F. W. (2004) Southern California children's health study exposure
33 database (final report). Petaluma, CA: Sonoma Technology, Inc., report no. STI-95230-
34 2453-FR3; prepared for University of Southern California School of Medicine, Division
35 of Environmental Health.
36 Aim, S.; Mukala, K.; Pasanen, P.; Tiittanen, P.; Ruuskanen, J.; Tuomisto, J.; Jantunen, M. J.
37 (1998) Personal N02 exposures of preschool children in Helsinki. J. Exposure Anal.
38 Environ. Epidemiol. 8: 79-100.
39 American Academy of Pediatrics, Committee on Environmental Health. (2004) Ambient air
40 pollution: health hazards to children. Pediatrics 114: 1699-1707.
March 2008 6-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 American Thoracic Society. (1985) Guidelines as to what constitutes an adverse respiratory
2 health effect, with special reference to epidemiologic studies of air pollution. Am. Rev.
3 Respir. Dis. 131: 666-668.
4 American Thoracic Society. (2000a) Guidelines for methacholine and exercise challenge testing-
5 1999. Am. J. Respir. Crit. Care Med. 161: 309-329.
6 American Thoracic Society. (2000b) What constitutes an adverse health effect of air pollution?
7 Am. J. Respir. Crit. Care Med. 161: 665-673.
8 Ammann, M.; Kalberer, M.; Jost, D. T.; Tobler, L.; Rossler, E.; Piguet, D.; Gaggeler, H. W.;
9 Baltensperger, U. (1998) Heterogeneous production of nitrous acid on soot in polluted air
10 masses. Nature (London) 395: 157-160.
11 Anderson, H. R.; Spix, C.; Medina, S.; Schouten, J. P.; Castellsague, J.; Rossi, G.; Zmirou, D.;
12 Touloumi, G.; Wojtyniak, B.; Ponka, A.; Bacharova, L.; Schwartz, J.; Katsouyanni, K.
13 (1997) Air pollution and daily admissions for chronic obstructive pulmonary disease in 6
14 European cities: results from the APHEA project. Eur. Respir. J. 10: 1064-1071.
15 Anderson, H. R.; Ponce de Leon, A.; Bland, J. M.; Bower, J. S.; Emberlin, J.; Strachen, D. P.
16 (1998) Air pollution, pollens, and daily admissions for asthma in London 1987-92.
17 Thorax 53: 842-848.
18 Anderson, Z. J.; Wahlin, P.; Raaschou-Nielsen, 0.; Scheike, T.; Loft, S. (2007a) Ambient
19 particle source apportionment and daily hospital admissions among children and elderly
20 in Copenhagen. J. Exposure Sci. Environ. Epidemiol. 17: 625-636.
21 Anderson, Z. J.; Wahlin, P.; Raaschou-Nielsen, 0.; Ketzel, M.; Scheike, T.; Loft, S. (2007b) Size
22 distribution and total number concentration of ultrafine and accumulation mode particles
23 and hospital admissions in children and the elderly in Copenhagen, Denmark. Occup.
24 Environ. Med.: 10.1136/oem.2007.033290.
25 Arbex, M. A.; Martins, L. C.; Carvalho et Oliveira, R.; Pereira, L. A. A.; Arbex, F. F.; Cancado,
26 J. E. D.; Saldiva, P. H. N.; Braga, A. L. F. (2007) Air pollution from biomass burning and
27 asthma hospital admissions in a sugar cane plantation area in Brazil. J. Epidemiol.
28 Community Health 61: 395-400.
29 Arey, J. (1998) Atmospheric reactions of PAHs including formation of nitroarenes. In: Neilson,
30 A. N., vol. ed. Anthropogenic compounds. V. 3, part I. PAHs and related compounds -
31 chemistry. New York, NY: Springer-Verlag; pp. 347-385. (Hutzinger, 0., ed. The
32 handbook of environmental chemistry series).
33 Arey, J.; Zielinska, B.; Atkinson, R.; Winer, A. M.; Ramdahl, T.; Pitts, J. N., Jr. (1986) The
34 formation of nitro-PAH from the gas-phase reactions of fluoranthene and pyrene with the
35 OH radical in the presence of NO*. Atmos. Environ. 20: 2339-2345.
36 Arey, J.; Atkinson, R.; Zielinska, B.; McElroy, P. A. (1989) Diurnal concentrations of volatile
37 polycyclic aromatic hydrocarbons and nitroarenes during a photochemical air pollution
38 episode in Glendora, California. Environ. Sci. Technol. 23: 321-327.
39 Atkinson, R. W.; Anderson, H. R.; Strachan, D. P.; Bland, J. M.; Bremner, S. A.; Ponce de Leon,
40 A. (1999a) Short-term associations between outdoor air pollution and visits to accident
March 2008 6-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 and emergency departments in London for respiratory complaints. Eur. Respir. J. 13:
2 257-265.
3 Atkinson, R. W.; Bremner, S. A.; Anderson, H. R.; Strachan, D. P.; Bland, J. M.; Ponce de Leon,
4 A. (1999b) Short-term associations between emergency hospital admissions for
5 respiratory and cardiovascular disease and outdoor air pollution in London. Arch.
6 Environ. Health 54: 398-411.
7 Atkinson, R. W.; Anderson, H. R.; Sunyer, J.; Ayres, J.; Baccini, M.; Vonk, J. M.; Boumghar,
8 A.; Forastiere, F.; Forsberg, B.; Touloumi, G.; Schwartz, J.; Katsouyanni, K. (2001)
9 Acute effects of particulate air pollution on respiratory admissions: results from APHEA
10 2 project. Am. J. Respir. Crit. Care Med. 164: 1860-1866.
11 Avol, E. L.; Gauderman, W. J.; Tan, S. M.; London, S. J.; Peters, J. M. (2001) Respiratory
12 effects of relocating to areas of differing air pollution levels. Am. J. Respir. Crit. Care
13 Med. 164: 2067-2072.
14 Azadniv, M.; Utell, M. J.; Morrow, P. E.; Gibb, F. R.; Nichols, J.; Roberts, N. J., Jr.; Speers, D.
15 M.; Torres, A.; Tsai, Y.; Abraham, M. K.; Voter, K. Z.; Frampton, M. W. (1998) Effects
16 of nitrogen dioxide exposure on human host defense. Inhalation Toxicol. 10: 585-602.
17 Baccarelli, A.; Zanobetti, A.; Martinelli, I.; Grillo, P.; Hou, L.; Giacomini, S.; Bonzini, M.;
18 Lanzani, G.; Mannucci, P. M.; Bertazzi, P. A.; Schwartz, J. (2007) Effects of exposure to
19 air pollution on blood coagulation. J. Thromb. Haemostasis 5: 252-260.
20 Balabaeva, L.; Tabakova, S. (1985) Lipidnata peroksidatsiya v dvye pokoleniya zhenski beli
21 plukhove, inkhalirani s azoten dvuokis [Lipid peroxidation in two progenies of female
22 albino rats inhaling nitrogen dioxide]. Khig. Zdraveopaz. 28: 41-46.
23 Ballester, F.; Tenias, J. M.; Perez-Hoyos, S. (2001) Air pollution and emergency hospital
24 admissions for cardiovascular diseases in Valencia, Spain. J. Epidemiol. Community
25 Health 55: 57-65.
26 Ballester, F.; Rodriguez, P.; Iniguez, C.; Saez, M.; Daponte, A.; Galan, L; Taracido, M.; Arribas,
27 F.; Bellido, J.; Cirarda, F. B.; Canada, A.; Guillen, J. J.; Guillen-Grima, F.; Lopez, E.;
28 Perez-Hoyos, S.; Lertxundi, A.; Toro, S. (2006) Air pollution and cardiovascular
29 admisisons association in Spain: results within the EMECAS project. J. Epidemiol.
30 Community Health 60: 328-336.
31 Bamford, H. A.; Baker, J. E. (2003) Nitro-polycyclic aromatic hydrocarbon concentrations and
32 sources in urban and suburban atmospheres of the mid-Atlantic region. Atmos. Environ.
33 37: 2077-2091.
34 Bamford, H. A.; Bezabeh, D. Z.; Schantz, M. M.; Wise, S. A.; Baker, J. E. (2003) Determination
35 and comparison of nitrated-polycyclic aromatic hydrocarbons measured in air and diesel
36 particulate reference materials. Chemosphere 50: 575-587.
37 Barck, C.; Sandstrom, T.; Lundahl, J.; Hallden, G.; Svartengren, M.; Strand, V.; Rak, S.; Bylin,
38 G. (2002) Ambient level of NOz augments the inflammatory response to inhaled allergen
39 in asthmatics. Respir. Med. 96: 907-917.
40 Barck, C.; Lundahl, J.; Hallden, G.; Bylin, G. (2005a) Brief exposures to N02 augment the
41 allergic inflammation in asthmatics. Environ. Res. 97: 58-66.
March 2008 6-3 DRAFT-DO NOT QUOTE OR CITE
-------
1 Barck, C.; Lundahl, J.; Holmstrom, M.; Bylin, G. (2005b) Does nitrogen dioxide affect
2 inflammatory markers after nasal allergen challenge? Am. J. Rhinol. 19: 560-566.
3 Barnett, A. G.; Williams, G. M.; Schwartz, J.; Neller, A. H.; Best, T. L.; Petroeschevsky, A. L.;
4 Simpson, R. W. (2005) Air pollution and child respiratory health: a case-crossover study
5 in Australia and New Zealand. Am. J. Respir. Crit. Care Med. 171: 1272-1278.
6 Barnett, A. G.; Williams, G. M.; Schwartz, J.; Best, T. L.; Neller, A. H.; Petroeschevsky, A. L.;
7 Simpson, R. W. (2006) The effects of air pollution on hospitalization for cardiovascular
8 disease in elderly people in Australian and New Zealand cities. Environ. Health Perspect.
9 114:1018-1023.
10 Earth, P. J.; Miiller, B.; Wagner, U.; Bittinger, A. (1994) Assessment of proliferative activity in
11 type II pneumocytes after inhalation of NOz by agnor-analysis. Exp. Toxicol. Pathol. 46:
12 335-342.
13 Bates, D. V.; Baker-Anderson, M.; Sizto, R. (1990) Asthma attack periodicity: a study of
14 hospital emergency visits in Vancouver. Environ. Res. 51: 51-70.
15 Bauer, M. A.; Utell, M. J.; Morrow, P. E.; Speers, D. M.; Gibb, F. R. (1986) Inhalation of 0.30
16 ppm nitrogen dioxide potentiates exercise-induced bronchospasm in asthmatics. Am.
17 Rev. Respir. Dis. 134: 1203-1208.
18 Baxter, L. K.; Clougherty, J. E.; Laden, F.; Levy, J. I. (2007a) Predictors of concentrations of
19 nitrogen dioxide, fine particulate matter, and particle constituents inside of lower
20 socioeconomic status urban homes. J. Exposure Sci. Environ. Epidemiol. 17: 433-444.
21 Baxter, L. K.; Clougherty, J. E.; Paciorek, C. J.; Wright, R. J.; Levy, J. I. (2007b) Predicting
22 residential indoor concentrations of nitrogen dioxide, fine particulate matter, and
23 elemental carbon using questionnaire and geographic information system based data.
24 Atmos. Environ. 41: 6561-6571.
25 Bedeschi, E.; Campari, C.; Candela, S.; Collini, G.; Caranci, N.; Frasca, G.; Galassi, C.;
26 Francesca, G.; Vigotti, M. A. (2007) Urban air pollution and respiratory emergency visits
27 at pediatric unit, Reggio Emilia, Italy. J. Toxicol. Environ. Health Part A 70: 261-265.
28 Beil, M.; Ulmer, W. T. (1976) Wirkung von NOz im MAK-Bereich auf Atemmechanik und
29 bronchiale Acetylcholinempfindlichkeit bei Normalpersonen [Effect of NOz in workroom
30 concentrations on respiratory mechanics and bronchial susceptibility to acetylcholine in
31 normal persons]. Int. Arch. Occup. Environ. Health 38: 31-44.
32 Belanger, K.; Gent, J. F.; Triche, E. W.; Bracken, M. B.; Leaderer, B. P. (2006) Association of
33 indoor nitrogen dioxide exposure with respiratory symptoms in children with asthma.
34 Am. J. Respir. Crit. Care Med. 173: 297-303.
35 Bell, M. L.; Ebisu, K.; Belanger, K. (2007) Ambient air pollution and low birth weight in
36 Connecticut and Massachusetts. Environ. Health Perspect. 115: 1118-1125.
37 Bermudez, E. (2001) Detection of poly(ADP-ribose) synthetase activity in alveolar macrophages
38 of rats exposed to nitrogen dioxide and ozone. Inhalation Toxicol. 13: 69-84.
39 Bernard, N.; Saintot, M.; Astre, C.; Gerber, M. (1998) Personal exposure to nitrogen dioxide
40 pollution and effect on plasma antioxidants. Arch. Environ. Health 53: 122-128.
March 2008 6-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 Bezabeh, D. Z.; Bamford, H. A.; Schantz, M. M.; Wise, S. A. (2003) Determination of nitrated
2 polycyclic aromatic hydrocarbons in diesel particulate-related standard reference
3 materials by using gas chromatography/mass spectrometry with negative ion chemical
4 ionization. Anal. Bioanal. Chem. 375: 381-388.
5 Biggeri, A.; Baccini, M.; Bellini, P.; Terracini, B. (2005) Meta-analysis of the Italian studies of
6 short-term effects of air pollution (MISA), 1990-1999. Int. J. Occup. Environ. Health 11:
7 107-122.
8 Bignal, K. L.; Ashmore, M. R.; Headley, A. D.; Stewart, K.; Weigert, K. (2007) Ecological
9 impacts of air pollution from road transport on local vegetation. Appl. Geochem. 22:
10 1265-1271.
11 Bils, R. F. (1976) The connective tissues and alveolar walls in the lungs of normal and oxidant-
12 exposed squirrel monkeys. J. Cell Biol. 70: 318.
13 Blomberg, A.; Krishna, M. T.; Bocchino, V.; Biscione, G. L.; Shute, J. K.; Kelly, F. J.; Frew, A.
14 J.; Holgate, S. T.; Sandstrom, T. (1997) The inflammatory effects of 2 ppm N02 on the
15 airways of healthy subjects. Am. J. Respir. Crit. Care Med. 156: 418-424.
16 Blomberg, A.; Krishna, M. T.; Helleday, R.; Soderberg, M.; Ledin, M.-C.; Kelly, F. J.; Frew, A.
17 J.; Holgate, S. T.; Sandstrom, T. (1999) Persistent airway inflammation but accomodated
18 antioxidant and lung function responses after repeated daily exposure to nitrogen dioxide.
19 Am. J. Respir. Crit. Care Med. 159: 536-543.
20 Bobak, M. (2000) Outdoor air pollution, low birth weight, and prematurity. Environ. Health
21 Perspect. 108: 173-176.
22 Boezen, M.; Schouten, J.; Rijcken, B.; Vonk, J.; Gerritsen, J.; Van Der Zee, S.; Hoek, G.;
23 Brunekreef, B.; Postma, D. (1998) Peak expiratory flow variability, bronchial
24 responsiveness, and susceptibility to ambient air pollution in adults. Am. J. Respir. Crit.
25 Care Med. 158: 1848-1854.
26 Boezen, H. M.; Van Der Zee, S. C.; Postma, D. S.; Vonk, J. M.; Gerritsen, J.; Hoek, G.;
27 Brunekreef, B.; Rijcken, B.; Schouten, J. P. (1999) Effects of ambient air pollution on
28 upper and lower respiratory symptoms and peak expiratory flow in children. Lancet 353:
29 874-878.
30 Boutin-Forzano, S.; Adel, N.; Gratecos, L.; Jullian, H.; Gamier, J. M.; Ramadour, M.;
31 Lanteaume, A.; Hamon, M.; Lafay, V.; Charpin, D. (2004) Visits to the emergency room
32 for asthma attacks and short-term variations in air pollution. A case-crossover study.
33 Respiration 71: 134-137.
34 Braga, A. L. F.; Saldiva, P. H. N.; Pereira, L. A. A.; Menezes, J. J. C.; Conceigao, G. M. C.; Lin,
35 C. A.; Zanobetti, A.; Schwartz, J.; Dockery, D. W. (2001) Health effects of air pollution
36 exposure on children and adolescents in Sao Paulo, Brazil. Pediatr. Pulmonol. 31: 106-
37 113.
38 Brauer, M.; Ryan, P. B.; Suh, H. H.; Koutrakis, P.; Spengler, J. D.; Leslie, N. P.; Billick, I. H.
39 (1990) Measurements of nitrous acid inside two research houses. Environ. Sci. Technol.
40 24: 1521-1527.
March 2008 6-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 Brauer, M.; Rasmussen, T. R.; Kjaergaard, S. K.; Spengler, J. D. (1993) Nitrous acid formation in
2 an experimental exposure chamber. Indoor Air 3: 94-105.
3 Brauer, M.; Hoek, G.; Van Vliet, P.; Meliefste, K.; Fischer, P. H.; Wijga, A.; Koopman, L. P.;
4 Neijens, H. J.; Gerritsen, J.; Kerkhof, M.; Heinrich, J.; Bellander, T.; Brunekreef, B.
5 (2002) Air pollution from traffic and the development of respiratory infections and
6 asthmatic and allergic symptoms in children. Am. J. Respir. Crit. Care Med. 166: 1092-
7 1098.
8 Brauer, M.; Hoek, G.; Smit, H. A.; De Jongste, J. C.; Gerritsen, J.; Postma, D. S.; Kerkhof, M.;
9 Brunekreef, B. (2007) Air pollution and development of asthma, allergy and infections in
10 a birth cohort. Eur. Respir. J. 29: 879-888.
11 Brook, R. D.; Franklin, B.; Cascio, W.; Hong, Y.; Howard, G.; Lipsett, M.; Luepker, R.;
12 Mittleman, M.; Samet, J.; Smith, S. C., Jr.; Tager, I. (2004) Air pollution and
13 cardiovascular disease. A statement for healthcare professionals from the Expert Panel on
14 Population and Prevention Science of the American Heart Association. Circulation 109:
15 2655-2671.
16 Brook, J. R.; Burnett, R. T.; Dann, T. F.; Cakmak, S.; Goldberg, M. S.; Fan, X.; Wheeler, A. J.
17 (2007) Further interpretation of the acute effect of nitrogen dioxide observed in Canadian
18 time-series studies. J. Exposure Sci. Environ. Epidemiol. 17: S36-S44.
19 Brown, S. K.; Mahoney, K. J.; Cheng, M. (2004) Room chamber assessment of the pollutant
20 emission properties of (nominally) low-emission unflued gas heaters. Indoor Air
21 14(suppl. 8): 84-91.
22 Buchdahl, R.; Parker, A.; Stebbings, T.; Babiker, A. (1996) Association between air pollution
23 and acute childhood wheezy episodes: prospective observational study. Br. Med. J. 312:
24 661-664.
25 Burnett, R. T.; Cakmak, S.; Brook, J. R.; Krewski, D. (1997a) The role of particulate size and
26 chemistry in the association between summertime ambient air pollution and
27 hospitalization for cardiorespiratory diseases. Environ. Health Perspect. 105: 614-620.
28 Burnett, R. T.; Brook, J. R.; Yung, W. T.; Dales, R. E.; Krewski, D. (1997b) Association
29 between ozone and hospitalization for respiratory diseases in 16 Canadian cities. Environ.
30 Res. 72:24-31.
31 Burnett, R. T.; Cakmak, S.; Brook, J. R. (1998) The effect of the urban ambient air pollution mix
32 on daily mortality rates in 11 Canadian cities. Can. J. Public Health 89: 152-156.
33 Burnett, R. T.; Smith-Doiron, M.; Stieb, D.; Cakmak, S.; Brook, J. R. (1999) Effects of
34 particulate and gaseous air pollution on cardiorespiratory hospitalizations. Arch. Environ.
35 Health 54: 130-139.
36 Burnett, R. T.; Brook, J.; Dann, T.; Delocla, C.; Philips, 0.; Cakmak, S.; Vincent, R.; Goldberg,
37 M. S.; Krewski, D. (2000) Association between particulate- and gas-phase components of
38 urban air pollution and daily mortality in eight Canadian cities. In: Grant, L. D., ed.
39 PM2000: particulate matter and health. Inhalation Toxicol. 12(suppl. 4): 15-39.
40 Burnett, R. T.; Smith-Doiron, M.; Stieb, D.; Raizenne, M. E.; Brook, J. R.; Dales, R. E.; Leech,
41 J. A.; Cakmak, S.; Krewski, D. (2001) Association between ozone and hospitalization for
March 2008 6-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 acute respiratory diseases in children less than 2 years of age. Am. J. Epidemiol. 153:
2 444-452.
3 Burnett, R. T.; Stieb, D.; Brook, J. R.; Cakmak, S.; Dales, R.; Raizenne, M.; Vincent, R.; Dann,
4 T. (2004) Associations between short-term changes in nitrogen dioxide and mortality in
5 Canadian cities. Arch. Environ. Health 59: 228-236.
6 Bush, T.; Smith, S.; Stevenson, K.; Moorcroft, S. (2001) Validation of nitrogen dioxide diffusion
7 tube methodology in the UK. Atmos. Environ. 35: 289-296.
8 Bylin, G.; Lindvall, T.; Rehn, T.; Sundin, B. (1985) Effects of short-term exposure to ambient
9 nitrogen dioxide concentrations on human bronchial reactivity and lung function. Eur. J.
10 Respir. Dis. 66: 205-217.
11 Cabral-Anderson, L. J.; Evans, M. J.; Freeman, G. (1977) Effects of NOz on the lungs of aging
12 rats: I. morphology. Exp. Mol. Pathol. 27: 353-365.
13 California Air Resources Board. (2007a) Review of the California ambient air quality standard
14 for nitrogen dioxide. Staff report: initial statement of reasons for proposed rulemaking.
15 Sacramento, CA: California Environmental Protection Agency, Air Resources Board.
16 Available: http://www.arb.ca.gov/research/aaqs/no2-rs/no2-doc.htm [23 July, 2007].
17 California Air Resources Board. (2007b) Review of the California ambient air quality standard
18 for nitrogen dioxide. Technical support document. Sacramento, CA: California
19 Environmental Protection Agency, Air Resources Board. Available:
20 http://www.arb.ca.gov/research/aaqs/no2-rs/no2-doc.htm [23 July, 2007].
21 Campbell, G. W.; Stedman, J. R.; Stevenson, K. (1994) A survey of nitrogen dioxide
22 concentrations in the United Kingdom using diffusion tubes, July-December 1991.
23 Atmos. Environ. 28: 477-486.
24 Cape, J. N. T.; Tang, Y. S.; van Dijk, N.; Love, L.; Sutton, M. A.; Palmer, S. C. F. (2004)
25 Concentrations of ammonia and nitrogen dioxide at roadside verges, and their
26 contribution to nitrogen deposition. Environ. Pollut. 132: 469-478.
27 Carslaw, N. (2007) A new detailed chemical model for indoor air pollution. Atmos. Environ. 41:
28 1164-1179.
29 Carson, J. L.; Collier, A. M.; Hu, S. -C.; Delvin, R. B. (1993) Effect of nitrogen dioxide on
30 human nasal epithelium. Am. J. Respir. Cell. Mol. Biol. 9: 264-270.
31 Castellsague, J.; Sunyer, J.; Saez, M.; Anto, J. M. (1995) Short-term association between air
32 pollution and emergency room visits for asthma in Barcelona. Thorax 50: 1051-1056.
33 Centers for Disease Control and Prevention. (2004) The health consequences of smoking: a
34 report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human
35 Services, National Center for Chronic Disease Prevention and Health Promotion, Office
36 on Smoking and Health. Available:
37 http://www.cdc.gov/tobacco/data_statistics/sgr/sgr_2004/chapters.htm (21 February,
38 2008)
39 Chan, L. Y.; Chan, C. Y.; Qin, Y. (1999) The effect of commuting microenvironment on
40 commuter exposures to vehicular emission in Hong Kong. Atmos. Environ. 33: 1777-
41 1787.
March 2008 6-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 Chan, C.-C.; Chuang, K.-J.; Su, T.-C.; Lin, L.-Y. (2005) Association between nitrogen dioxide
2 and heart rate variability in a susceptible population. Eur. J. Cardiovasc. Prev. Rehabil.
3 12: 580-586.
4 Chan, C.-C.; Chuang, K.-J.; Chien, L.-C.; Chen, W.-J.; Chang, W.-T. (2006) Urban air pollution
5 and emergency admissions for cerebrovascular diseases in Taipei, Taiwan. Eur. Heart J.
6 27: 1238-1244.
7 Chang, C.-C.; Tsai, S.-S.; Ho, S.-C.; Yang, C.-Y. (2005) Air pollution and hospital admissions
8 for cardiovascular disease in Taipei, Taiwan. Environ. Res. 98: 114-119.
9 Chardon, B.; Lefranc, A.; Granados, D.; Gremy, I. (2007) Air pollution and doctors' house calls
10 for respiratory diseases in the greater Paris area (2000-3). Occup. Environ. Med. 64: 320-
11 324.
12 Chauhan, A. J.; Krishna, M. T.; Frew, A. J.; Holgate, S. T. (1998) Exposure to nitrogen dioxide
13 (N02) and respiratory disease risk. Rev. Environ. Health 13: 73-90.
14 Chauhan, A. J.; Inskip, H. M.; Linaker, C. H.; Smith, S.; Schreiber, J. ; Johnston, S. L.; Holgate,
15 S. T. (2003) Personal exposure to nitrogen dioxide (N02) and the severity of virus-
16 induced asthma in children. Lancet 361: 1939-1944.
17 Chen, Y.; Yang, Q.; Krewski, D.; Burnett, R. T.; Shi, Y.; McGrail, K. M. (2005) The effect of
18 coarse ambient particulate matter on first, second, and overall hospital admissions for
19 respiratory disease among the elderly. Inhalation Toxicol. 17: 649-655.
20 Chew, F. T.; Goh, D. Y. T.; Ooi, B. C.; Saharom, R.; Hui, J. K. S.; Lee, B. W. (1999)
21 Association of ambient air-pollution levels with acute asthma exacerbation among
22 children in Singapore. Allergy (Copenhagen) 54: 320-329.
23 Clougherty, J. E.; Levy, J. I.; Kubzansky, L. D.; Ryan, P. B.; Suglia, S. F.; Canner, M. J.;
24 Wright, R. J. (2007) Synergistic effects of traffic-related air pollution and exposure to
25 violence on urban asthma etiology. Environ. Health Perspect. 115: 1140-1146.
26 Cocheo, V.; Boaretto, C.; Sacco, P. (1996) High uptake rate radial diffusive sampler suitable for
27 both solvent and thermal desorption. Am. Ind. Hyg. Assoc. J. 57: 897-904.
28 Cockcroft, D. W.; Davis, B. E. (2006) Airway hyperresponsiveness as a determinant of the early
29 asthmatic response to inhaled allergen. J. Asthma 43: 175-178.
30 Cockcroft, D. W.; Davis, B. E.; Todd, D. C.; Smycniuk, A. J. (2005) Methacholine challenge:
31 comparison of two methods. Chest 127: 839-844.
32 Code of Federal Regulations. (2002) Ambient air quality surveillance; appendix E - probe and
33 monitoring path citing criteria for ambient air quality monitoring. C. F. R. 40: §58.
34 Coffin, D. L.; Gardner, D. E. (1972) Interaction of biological agents and chemical air pollutants.
35 Ann. Occup. Hyg. 15: 219-234.
36 Connell, D. P.; Withum, J. A.; Winter, S. E.; Statnick, R. M.; Bilonick,, R. A. (2005) The
37 Steubenville Comprehensive Air Monitoring Program (SCAMP): associations among
38 fine particulate matter, co-pollutants, and meteorological conditions. J. Air Waste
39 Manage. Assoc. 55: 481-496.
March 2008 6-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 Cotterill, A.; Kingham, S. (1997) Nitrogen dioxide in the home: cooking, double glazing, or
2 outdoor air? Indoor Built Environ. 6: 344-349.
3 Cox, R. M. (2003) The use of passive sampling to monitor forest exposure to 03, N02, and S02:
4 a review and some case studies. Environ. Pollut. 126: 301-311.
5 Crosley, D. R. (1996) NO^ blue ribbon panel. J. Geophys. Res. [Atmos.] 101: 2049-2052.
6 Cross, C. E.; Valacchi, G.; Schock, B.; Wilson, M.; Weber, S.; Eiserich, J.; Van der Vliet, A.
7 (2002) Environmental oxidant pollutant effects on biologic systems. Am. J. Respir. Crit.
8 Care Med. 166: 544-550.
9 Cyrys, J.; Heinrich, J.; Richter, K.; Wolke, G.; Wichmann, H. E. (2000) Sources and
10 concentrations of indoor nitrogen dioxide in Hamburg (west Germany) and Erfurt (east
11 Germany). Sci. Total Environ. 250: 51-62.
12 Cyrys, J.; Stolzel, M.; Heinrich, J.; Kreyling, W. G.; Menzel, N.; Wittmaack, K.; Tuch, T.;
13 Wichmann, H.-E. (2003) Elemental composition and sources of fine and ultrafine
14 ambient particles in Erfurt, Germany. Sci. Total Environ. 305: 143-156.
15 Dab, W.; Medina, S.; Quenel, P.; Le Moullec, Y.; Le Tertre, A.; Thelot, B.; Monteil, C.;
16 Lameloise, P.; Pirard, P.; Momas, I.; Ferry, R.; Festy, B. (1996) Short term respiratory
17 health effects of ambient air pollution: results of the APHEA project in Paris. In: St
18 Leger, S., ed. The APHEA project. Short term effects of air pollution on health: a
19 European approach using epidemiological time series data. J. Epidemiol. Commun.
20 Health 50(suppl. 1): S42-S46.
21 Dales, R.; Burnett, R. T.; Smith-Doiron, M.; Stieb, D. M.; Brook, J. R. (2004) Air pollution and
22 sudden infant death syndrome. Pediatrics 113: 628-631.
23 Dales, R. E.; Cakmak, S.; Doiron, M. S. (2006) Gaseous air pollutants and hospitalization for
24 respiratory disease in the neonatal period. Environ. Health Perspect. 114: 1751-1754.
25 DeFrances, C. J.; Hall, M. J.; Podgornik, M. N. (2005) 2003 National Hospital Discharge
26 Survey. Hyattsville, MD: National Center for Health Statistics; DHHS publication no.
27 (PHS) 2004-1250. (Advance data from vital and health statistics; no. 359). Available:
28 http://www.cdc.gov/nchs/data/ad/ad359.pdf [3 August, 2005].
29 Delfino, R. J.; Zeiger, R. S.; Seltzer, J. M.; Street, D. H.; McLaren, C. E. (2002) Association of
30 asthma symptoms with peak particulate air pollution and effect modification by anti-
31 inflammatory medication use. Environ. Health Perspect. 110: A607-A617.
32 Delfino, R. J.; Gone, H.; Linn, W. S.; Pellizzari, E. D.; Hu, Y. (2003) Asthma symptoms in
33 Hispanic children and daily ambient exposures to toxic and criteria air pollutants.
34 Environ. Health Perspect. Ill: 647-656.
35 Delfino, R. J.; Staimer, N.; Gillen, D.; Tjoa, T.; Sioutas, C.; Fung, K.; George, S. C.; Kleinman,
36 M. T. (2006) Personal and ambient air pollution is associated with increased exhaled
37 nitric oxide in children with asthma. Environ. Health Perspect. 114: 1736-1743.
38 Dennekamp, M.; Howarth, S.; Dick, C. A. J.; Cherrie, J. W.; Donaldson, K.; Seaton, A. (2001)
39 Ultrafine particles and nitrogen oxides generated by gas and electric cooking. Occup.
40 Environ. Med. 58: 511-516.
March 2008 6-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 Dentener, F.; Stevenson, D.; Ellingsen, K.; Van Noije, T.; Schultz, M.; Amann, M.; Atherton, C.;
2 Bell, N.; Bergmann, D.; Bey, I.; Bouwman, L.; Butler, T.; Cofala, J.; Collins, B.; Drevet,
3 J.; Doherty, R.; Eickhout, B.; Eskes, H.; Fiore, A.; Gauss, M.; Hauglustaine, D.;
4 Horowitz, L.; Isaksen, I. S. A.; Josse, B.; Lawrence, M.; Krol, M.; Lamarque, J. F.;
5 Montanaro, V.; Muller, J. F.; Peuch, V. H.; Pitari, G.; Pyle, J.; Rast, S.; Rodriguez, J.;
6 Sanderson, M.; Savage, N. H.; Shindell, D.; Strahan, S.; et al. (2006a) The global
7 atmospheric environment for the next generation. Environ. Sci. Technol. 40: 3586-3594.
8 Dentener, F.; Drevet, J.; Lamarque, J. F.; Bey, L; Eickhout, B.; Fiore, A. M.; Hauglustaine, D.;
9 Horowitz, L. W.; Krol, M.; Kulshrestha, U. C.; Lawrence, M.; Galy-Lacaux, C.; Rast, S.;
10 Shindell, D.; Stevenson, D.; Van Noije, T.; Atherton, C.; Bell, N.; Bergman, D.; Butler,
11 T.; Cofala, J.; Collins, B.; Doherty, R.; Ellingsen, K.; Galloway, J.; Gauss, M.;
12 Montanaro, V.; Muller, J. F.; Pitari, G.; Rodriguez, J.; Sanderson, M.; Solman, F.;
13 Strahan, S.; Schultz, M.; Sudo, K.; Szopa, S.; Wild, 0. (2006b) Nitrogen and sulfur
14 deposition on regional and global scales: a multi-model evaluation. Global Biogeochem.
15 Cycles: 20(GB4003): 10.1029/2005GB002672.
16 Desqueyroux, H.; Pujet, J.-C.; Prosper, M.; Le Moullec, Y.; Momas, I. (2002) Effects of air
17 pollution on adults with chronic obstructive pulmonary disease. Arch. Environ. Health
18 57: 554-560.
19 Devalia, J. L.; Campbell, A. M.; Sapsford, R. J.; Rusznak, C.; Quint, D.; Godard, P.; Bousquet, J.
20 Davies, R. J. (1993a) Effect of nitrogen dioxide on synthesis of inflammatory cytokines
21 expressed by human bronchial epithelial cells in vitro. Am. J. Respir. Cell. Mol. Biol. 9:
22 271-278.
23 Devalia, J. L.; Sapsford, R. J.; Cundell, D. R.; Rusznak, C.; Campbell, A. M.; Davies, R. J.
24 (1993b) Human bronchial epithelial cell dysfunction following in vitro exposure to
25 nitrogen dioxide. Eur. Respir. J. 6: 1308-1316.
26 Devalia, J. L.; Rusznak, C.; Herdman, M. J.; Trigg, C. J.; Tarraf, H.; Davies, R. J. (1994) Effect
27 of nitrogen dioxide and sulphur dioxide on airway response of mild asthmatic patients to
28 allergen inhalation. Lancet 344: 1668-1671.
29 Devlin, R.; Horstman, D.; Becker, S.; Gerrity, T.; Madden, M.; Koren, H. (1992) Inflammatory
30 response in humans exposed to 2.0 ppm N02. Am. Rev. Respir. Dis. 145: A456.
31 Devlin, R. B.; Horstman, D. P.; Gerrity, T. R.; Becker, S.; Madden, M. C. (1999) Inflammatory
32 response in humans exposed to 2.0 PPM nitrogen dioxide. Inhalation Toxicol. 11: 89-
33 109.
34 Dewanji, A.; Moolgavkar, S. H. (2000) A Poisson process approach for recurrent event data with
35 environmental covariates. Environmetrics 11: 665-673.
36 Dietert, R. R.; Etzel, R. A.; Chen, D.; Halonen, M.; Holladay, S. D.; Jarabek, A. M.; Landreth,
37 K.; Peden, D. B.; Pinkerton, K.; Smialowicz, R. J.; Zoetis, T. (2000) Workshop to
38 identify critical window of exposure for children's health: immune and respiratory
39 systems work group summary. Environ. Health Perspect. Suppl. 108(3): 483-490.
40 Di Giovanni, V.; Cagiano, R.; Carratu, M. R.; De Salvia, M. A.; Giustino, A.; Cuomo, V. (1994)
41 Alterations in the ontogeny of rat pup ultrasonic vocalization produced by prenatal
42 exposure to nitrogen dioxide. Psychopharmacology 116: 423-427.
March 2008 6-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 Dimitroulopoulou, C.; Ashmore, M. R.; Byrne, M. A.; Kinnersley, R. P. (2001) Modelling of
2 indoor exposure to nitrogen dioxide in the UK. Atmos. Environ. 35: 269-279.
3 D'Ippoliti, D.; Forastiere, F.; Ancona, C.; Agabiti, N.; Fusco, D.; Michelozzi, P.; Perucci, C. A.
4 (2003) Air pollution and myocardial infarction in Rome: a case-crossover analysis.
5 Epidemiology 14: 528-535.
6 Dockery, D. W.; Brunekreef, B. (1996) Longitudinal studies of air pollution effects on lung
7 function. Am. J. Respir. Crit. Care Med. 154(suppl.): S250-S256.
8 Dockery, D. W.; Pope, C. A., Ill; Xu, X.; Spengler, J. D.; Ware, J. H.; Fay, M. E.; Ferris, B. G.,
9 Jr.; Speizer, F. E. (1993) An association between air pollution and mortality in six U.S.
10 cities. N. Engl. J. Med. 329: 1753-1759.
11 Dockery, D. W.; Luttmann-Gibson, H.; Rich, D. Q.; Link, M. S.; Schwartz, J. D.; Gold, D. R.;
12 Koutrakis, P.; Verrier, R. L.; Mittleman, M. A. (2005) Particulate air pollution and
13 nonfatal cardiac events. Part II. Association of air pollution with confirmed arrhythmias
14 recorded by implanted defibrillators. Boston, MA: Health Effects Institute; research
15 report no. 124; pp. 83-126; discussion; pp. 127-148. Available:
16 http://pubs.healtheffects.org/ [7 June, 2007].
17 Dominici, F.; McDermott, A.; Zeger, S. L.; Samet, J. M. (2002) On the use of generalized
18 additive models in time-series studies of air pollution and health. Am. J. Epidemiol. 156:
19 193-203.
20 Dominici, F.; McDermott, A.; Daniels, M.; Zeger, S. L.; Samet, J. M. (2003) Mortality among
21 residents of 90 cities. In: Revised analyses of time-series studies of air pollution and
22 health. Special report. Boston, MA: Health Effects Institute; pp. 9-24. Available:
23 http://www.healtheffects.org/Pubs/TimeSeries.pdf [12 May, 2004].
24 Douglas, G. J.; Price, J. F.; Page, C. P. (1994) A method for the long-term exposure of rabbits to
25 environmental pollutant gases. Eur. Respir. J. 7: 1516-1526.
26 Downs, S. H.; Schindler, C.; Liu, L.-J. S.; Keidel, D.; Bayer-Oglesby, L.; Brutsche, M. H.;
27 Gerbase, M. W.; Keller, R.; Kiinzli, N.; Leuenberger, P.; Probst-Hensch, N. M.; Tschopp,
28 J.-M.; Zellweger, J.-P.; Rochat, T.; Schwartz, J.; Ackermann-Liebrich, U.; the
29 SALPADIA Team. (2007) Reduced exposure to PMio and attenuated age-related decline
30 in lung function. N. Engl. J. Med. 357: 2338-2347.
31 Drechsler-Parks, D. M. (1995) Cardiac output effects of 03 and N02 exposure in healthy older
32 adults. Toxicol. Ind. Health 11: 99-109.
33 Dunlea, E. J.; Herndon, S. C.; Nelson, D. D.; Volkamer, R. M.; San Martini, F.; Sheehy, P. M.;
34 Zahniser, M. S.; Shorter, J. H.; Wormhoudt, J. C.; Lamb, B. K.; Allwine, E. J.; Gaffney,
35 J. S.; Marley, N. A.; Grutter, M.; Marquez, C.; Blanco, S.; Cardenas, B.; Retama, A.;
36 Ramon Villegas, C. R.; Kolb, C. E.; Molina, L. T.; Molina, M. J. (2007) Evaluation of
37 nitrogen dioxide chemiluminescence monitors in a polluted urban environment. Atmos.
38 Chem. Phys. 7: 2691-2704.
39 Durant, J. L.; Busby, W. F., Jr.; Lafleur, A. L.; Penman, B. W.; Crespi, C. L. (1996) Human cell
40 mutagenicity of oxygenated, nitrated and unsubstituted polycyclic aromatic hydrocarbons
41 associated with urban aerosols. Mutat. Res. 371: 123-157.
March 2008 6-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 Button, S. J.; Hannigan, M. P.; Miller, S. L. (2001) Indoor pollutant levels from the use of
2 unvented natural gas fireplaces in Boulder, Colorado. J. Air Waste Manage. Assoc. 51:
3 1654-1661.
4 Eggleston, H. S. (1993) Uncertainties in the estimates of emissions of VOCs from motor cars. In:
5 Proceedings of the TNO/EURASAP workshop on the reliability of VOC emission
6 databases; June; The Netherlands: Delft, IMW-TNO Publication, P 93/040, pp. 59-73.
7 Ehrlich, R. (1966) Effect of nitrogen dioxide on resistance to respiratory infection. Bacteriol.
8 Rev. 30: 604-614.
9 Ehrlich, R.; Findlay, J. C.; Gardner, D. E. (1979) Effects of repeated exposures to peak
10 concentrations of nitrogen dioxide and ozone on resistance to streptococcal pneumonia. J.
11 Toxicol. Environ. Health 5: 631-642.
12 Ekberg, L. E. (1996) Relationships between indoor and outdoor contaminants in mechanically
13 ventilated buildings. Indoor Air 6: 41-47.
14 Elsom, D. M. (1992) Atmospheric pollution: a global problem. 2nd ed. Oxford, United
15 Kingdom: Blackwell Publishers.
16 Erbas, B.; Kelly, A.-M.; Physick, B.; Code, C.; Edwards, M. (2005) Air pollution and childhood
17 asthma emergency hospital admissions: estimating intra-city regional variations. Int. J.
18 Environ. Health Res. 15: 11-20.
19 Evans, M. J.; Stephens, R. J.; Cabral, L. J.; Freeman, G. (1972) Cell renewal in the lungs of rats
20 exposed to low levels of N02. Arch. Environ. Health 24: 180-188.
21 Evans, M. J.; Cabral, L. J.; Stephens, R. J.; Freeman, G. (1973a) Cell division of alveolar
22 macrophages in rat lung following exposure to N02. Am. J. Pathol. 70: 199-208.
23 Evans, M. J.; Cabral, L. J.; Stephens, R. J.; Freeman, G. (1973b) Renewal of alveolar epithelium
24 in the rat following exposure to N02. Am. J. Pathol. 70: 175-190.
25 Evans, M. J.; Cabral, L. C.; Stephens, R. J.; Freeman, G. (1974) Acute kinetic response and
26 renewal of the alveolar epithelium following injury by nitrogen dioxide. Chest 65 (suppl.):
27 62S-65S.
28 Evans, H. L.; Laties, V. G.; Weiss, B. (1975) Behavioral effects of mercury and methylmercury.
29 Fed. Proc. 34: 1858-1867.
30 Evans, M. J.; Johnson, L. V.; Stephens, R. J.; Freeman, G. (1976) Renewal of the terminal
31 bronchiolar epithelium in the rat following exposure to NOz or Os. Lab. Invest. 35: 246-
32 257.
33 Evans, M. J.; Cabral-Anderson, L. J.; Freeman, G. (1977) Effects of N02 on the lungs of aging
34 rats: II. cell proliferation. Exp. Mol. Pathol. 27: 366-376.
35 Fairley, D. (1999) Daily mortality and air pollution in Santa Clara County, California: 1989-
36 1996. Environ. Health Perspect. 107: 637-641.
37 Fairley, D. (2003) Mortality and air pollution for Santa Clara County, California, 1989-1996. In:
38 Revised analyses of time-series studies of air pollution and health. Special report. Boston,
39 MA: Health Effects Institute; pp. 97-106. Available:
40 http://pubs.healtheffects.org/view.php?id=4 [12 February, 2008].
March 2008 6-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 Fan, Z.; Chen, D.; Birla, P.; Kamens, R. M. (1995) Modeling of nitro-polycyclic aromatic
2 hydrocarbon formation and decay in the atmosphere. Atmos. Environ. 29: 1171-1181.
3 Fan, Z.; Kamens, R. M.; Hu, J.; Zhang, J.; McDow, S. (1996) Photostability of nitro-polycyclic
4 aromatic hydrocarbons on combustion soot particles in sunlight. Environ. Sci. Technol.
5 30: 1358-1364.
6 Farchi, S.; Forastiere, F.; Cesaroni, G.; Perucci, C. A. (2006) Environmental exposures and
7 hospitalisation for respiratory conditions in children: a five year follow up study in Rome,
8 Italy. Occup. Environ. Med. 63: 573-576.
9 Farrow, A.; Greenwood, R.; Preece, S.; Golding, J. (1997) Nitrogen dioxide, the oxides of
10 nitrogen, and infants' health symptoms. Arch. Environ. Health 52: 189-194.
11 Febo, A.; Perrino, C. (1991) Prediction and experimental evidence for high air concentration of
12 nitrous acid in indoor environments. Atmos. Environ. Part A 25: 1055-1061.
13 Federal Register. (1971) National primary and secondary ambient air quality standards. F. R.
14 (April 30) 36: 8186-8201.
15 Federal Register. (1984) Proposed reaffirmation of the National Ambient Air Quality Standards
16 for Nitrogen Dioxide. F. R. 49 (February 23): 6866-6879.
17 Federal Register. (1985) Retention of the national ambient air quality standards for nitrogen
18 dioxide: final rule. F. R. (June 19) 50: 25,532-25,545.
19 Federal Register. (1987) Air quality criteria for carbon monoxide; air quality criteria for oxides
20 of nitrogen. F. R. 52 (July 22): 27580.
21 Federal Register. (1991) Draft criteria document for oxides of nitrogen. F. R. 56 (November 25):
22 59285.
23 Federal Register. (1995) National ambient air quality standards for nitrogen dioxide: proposed
24 decision. F. R. 60 (October 11): 52874-52889.
25 Federal Register. (2005) Air quality criteria for oxides of nitrogen. F. R. 70 (December 9):
26 73236-73237.
27 Fehsenfeld, F. C.; Dickerson, R. R.; Hiibler, G.; Luke, W. T.; Nunnermacker, L. J.; Williams, E.
28 J.; Roberts, J. M.; Calvert, J. G.; Curran, C. M.; Delany, A. C.; Eubank, C. S.; Fahey, D.
29 W.; Fried, A.; Gandrud, B. W.; Langford, A. 0.; Murphy, P. C.; Norton, R. B.; Pickering,
30 K. E.; Ridley, B. A. (1987) A ground-based intercomparison of NO, NOX, and N0y
31 measurement techniques. J. Geophys. Res. [Atmos.] 92: 14,710-14,722.
32 Feilberg, A.; Nielsen, T. (2001) Photodegradation of nitro-PAHs in viscous organic media used
33 as models of organic aerosols. Environ. Sci. Technol. 35: 108-113.
34 Feilberg, A.; Kamens, R. M.; Strommen, M. R.; Nielsen, T. (1999) Modeling the formation,
35 decay, and partitioning of semivolatile nitro-polycyclic aromatic hydrocarbons
36 (nitronaphthalenes) in the atmosphere. Atmos. Environ. 33: 1231-1243.
37 Fenters, J. D.; Findlay, J. C.; Port, C. D.; Ehrlich, R.; Coffin, D. L. (1973) Chronic exposure to
38 nitrogen dioxide: immunologic, physiologic, and pathologic effects in virus-challenged
39 squirrel monkeys. Arch. Environ. Health 27: 85-89.
March 2008 6-13 DRAFT-DO NOT QUOTE OR CITE
-------
1 Filleul, L.; Rondeau, V.; Vandentorren, S.; Le Moual, N.; Cantagrel, A.; Annesi-Maesano, I.;
2 Charpin, D.; Declercq, C.; Neukirch, F.; Paris, C.; Vervloet, D.; Brochard, P.; Tessier, J.
3 F.; Kauffmann, F.; Baldi, I. (2005) Twenty five year mortality and air pollution: results
4 from the French PAARC survey. Occup. Environ. Med. 62: 453-460.
5 Finkelstein, M. M.; Jerrett, M.; DeLuca, P.; Finkelstein, N.; Verma, D. K.; Chapman, K.; Sears,
6 M. R. (2003) Relation between income, air pollution and mortality: a cohort study. Can.
7 Med. Assoc. J. 169: 397-402.
8 Finkelstein, M. M.; Jerrett, M.; Sears, M. R. (2004) Traffic air pollution and mortality rate
9 advancement periods. Am. J. Epidemiol. 160: 173-177.
10 Finkelstein, M. M.; Jerrett, M.; Sears, M. R. (2005) Environmental inequality and circulatory
11 disease mortality gradients. J. Epidemiol. Community Health 59: 481-487.
12 Finlayson-Pitts, B. J.; Pitts, J. N., Jr. (2000) Chemistry of the upper and lower atmosphere:
13 theory, experiments and applications. San Diego, CA: Academic Press.
14 Fisher, R. A. (1925) The correlation coefficient. In: Statistical methods for research workers.
15 Edinburgh, UK: Oliver and Boyd. Available:
16 http://psychclassics.yorku.ca/Fisher/Methods/chap6.htm (23 February, 2008).
17 Folinsbee, L. J. (1992) Does nitrogen dioxide exposure increase airways responsiveness?
18 Toxicol. Ind. Health 8: 273-283.
19 Forsberg, B.; Stjernberg, N.; Linne, R.; Segerstedt, B.; Wall, S. (1998) Daily air pollution levels
20 and acute asthma in southern Sweden. Eur. Respir. J. 12: 900-905.
21 Fortmann, R.; Kariher, P.; Clayton, R. (2001) Indoor air quality: residential cooking exposures.
22 Final report. Sacramento, CA: State of California Air Resources Board; ARE Contract
23 No. 97-330. Available: http://arb.ca.gov/research/abstracts/97-330.htm [22 May, 2007].
24 Frampton, M. W.; Morrow, P. E.; Cox, C.; Gibb, F. R.; Speers, D. M.; Utell, M. J. (1991) Effects
25 of nitrogen dioxide exposure on pulmonary function and airway reactivity in normal
26 humans. Am. Rev. Respir. Dis. 143: 522-527.
27 Frampton, M. W.; Boscia, J.; Roberts, N. J., Jr.; Azadniv, M.; Torres, A.; Cox, C.; Morrow, P.
28 E.; Nichols, J.; Chalupa, D.; Frasier, L. M.; Gibb, F. R.; Speers, D. M.; Tsai, Y.; Utell, M.
29 J. (2002) Nitrogen dioxide exposure: effects on airway and blood cells. Am. J. Physiol.
30 282:L155-L165.
31 Freeman, B. A.; Mudd, J. B. (1981) Reaction of ozone with sulfhydryls of human erythrocytes.
32 Arch. Biochem. Biophys. 208: 212-220.
33 Freeman, G.; Furiosi, N. J.; Haydon, G. B. (1966) Effects of continuous exposure of 0.8 ppm
34 NOz on respiration of rats. Arch. Environ. Health 13: 454-456.
35 Freeman, G.; Stephens, R. J.; Crane, S. C.; Furiosi, N. J. (1968) Lesion of the lung in rats
36 continuously exposed to two parts per million of nitrogen dioxide. Arch. Environ. Health
37 17: 181-192.
38 Freeman, G.; Crane, S. C.; Furiosi, N. J.; Stephens, R. J.; Evans, M. J.; Moore, W. D. (1972)
39 Covert reduction in ventilatory surface in rats during prolonged exposure to subacute
40 nitrogen dioxide. Am. Rev. Respir. Dis. 106: 563-579.
March 2008 6-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 Fung, K. Y.; Luginaah, I.; Gorey, K. M.; Webster, G. (2005) Air pollution and daily hospital
2 admissions for cardiovascular diseases in Windsor, Ontario. Can. J. Public Health 96: 29-
3 33.
4 Fung, K. Y.; Khan, S.; Krewski, D.; Chen, Y. (2006) Association between air pollution and
5 multiple respiratory hospitalizations among the elderly in Vancouver, Canada. Inhalation
6 Toxicol. 18: 1005-1011.
7 Fung, K. Y.; Luginaah, I. N.; Gorey, K. M. (2007) Impact of air pollution on hospital admissions
8 in Southwestern Ontario, Canada: generating hypotheses in sentinel high-exposure
9 places. Environ. Health. 6: 18.
10 Fusco, D.; Forastiere, F.; Michelozzi, P.; Spadea, T.; Ostro, B.; Area, M.; Perucci, C. A. (2001)
11 Air pollution and hospital admissions for respiratory conditions in Rome, Italy. Eur.
12 Respir.J. 17: 1143-1150.
13 Gair, A. J.; Penkett, S. A. (1995) The effects of wind speed and turbulence on the performance of
14 diffusion tube samplers. Atmos. Environ. 29: 2529-2533.
15 Gair, A. J.; Penkett, S. A.; Oyola, P. (1991) Development of a simple passive technique for the
16 determination of nitrogen-dioxide in remote continental locations. Atmos. Environ. Part
17 A 25: 1927-1939.
18 Galan, I.; Tobias, A.; Banegas, J. R.; Aranguez, E. (2003) Short-term effects of air pollution on
19 daily asthma emergency room admissions. Eur. Respir. J. 22: 802-808.
20 Gamble, J. L. (1998) Effects of ambient air pollution on daily mortality: a time series analysis of
21 Dallas, Texas, 1990-1994. Presented at: 91st annual meeting and exhibition of the Air &
22 Waste Management Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste
23 Management Association; paper no. 98-MP26.03.
24 Garcia-Algar, 0.; Zapater, M.; Figueroa, C.; Vail, 0.; Basagana, X.; Sunyer, J.; Freixa, A.;
25 Guardino, X.; Pichini, S. (2003) Sources and concentrations of indoor nitrogen dioxide in
26 Barcelona, Spain. J. Air Waste Manage. Assoc. 53: 1312-1317.
27 Garcia Algar, 0.; Pichini, S.; Basagana, X.; Puig, C.; Vail, 0.; Torrent, M.; Harris, J.; Sunyer, J.;
28 Cullinan, P. (2004) Concentrations and determinants of NOz in homes of Ashford, UK
29 and Barcelona and Menorca, Spain. Indoor Air 14: 298-304.
30 Garcia-Aymerich, J.; Tobias, A.; Anto, J. M.; Sunyer, J. (2000) Air pollution and mortality in a
31 cohort of patients with chronic obstructive pulmonary disease: a time series analysis. J.
32 Epidemiol. Community Health 54: 73-74.
33 Gardner, D. E. (1982) Use of experimental airborne infections for monitoring altered host
34 defenses. Environ. Health Perspect. 43: 99-107.
35 Gardner, D. E.; Miller, F. J.; Blommer, E. J.; Coffin, D. L. (1979) Influence of exposure mode on
36 the toxicity of N02. Environ. Health Perspect. 30: 23-29.
37 Garrett, M. H.; Hooper, M. A.; Hooper, B. M.; Abramson, M. J. (1998) Respiratory symptoms in
38 children and indoor exposure to nitrogen dioxide and gas stoves. Am. J. Respir. Crit.
39 Care Med. 158:891-895.
March 2008 6-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 Garrett, M. H.; Hooper, M. A.; Hooper, B. M. (1999) Nitrogen dioxide in Australian homes:
2 levels and sources. J. Air Waste Manage. Assoc. 49: 76-81.
3 Garty, B. Z.; Kosman, E.; Ganor, E.; Berger, V.; Garty, L.; Wietzen, T.; Waisman, Y.; Mimouni,
4 M.; Waisel, Y. (1998) Emergency room visits of asthmatic children, relation to air
5 pollution, weather, and airborne allergens. Ann. Allergy Asthma Immunol. 81: 563-570.
6 Gauderman, W. J.; Avol, E.; Gilliland, F.; Vora, H.; Thomas, D.; Berhane, K.; McConnell, R.;
7 Kuenzli, N.; Lurmann, F.; Rappaport, E.; Margolis, H.; Bates, D.; Peters, J. (2004) The
8 effect of air pollution on lung development from 10 to 18 years of age. N. Engl. J. Med.
9 351: 1057-1067.
10 Gauderman, W. J.; Avol, E.; Lurmann, F.; Kuenzli, N.; Gilliland, F.; Peters, J.; McConnell, R.
11 (2005) Childhood asthma and exposure to traffic and nitrogen dioxide. Epidemiology 16:
12 737-743.
13 Gauderman, W. J.; Vora, H.; McConnell, R.; Berhane, K.; Gilliland, F.; Thomas, D.; Lurmann,
14 F.; Avol, E.; Kunzli, N. (2007) Effect of exposure to traffic on lung development from 10
15 to 18 years of age: a cohort study. Lancet 369: 571-577.
16 Gauvin, S.; Le Moullec, Y.; Bremont, F.; Momas, I.; Balducci, F.; Ciognard, F.; Poilve, M.-P.;
17 Zmirou, D.; VESTA Investigators. (2001) Relationships between nitrogen dioxide
18 personal exposure and ambient air monitoring measurements among children in three
19 French metropolitan areas: VESTA study. Arch. Environ. Health 56: 336-341.
20 Gavras, J. B.; Frampton, M. W.; Ryan, D. H.; Levy, P. C.; Looney, R. J.; Cox, C.; Morrow, P.
21 E.; Utell, M. J. (1994) Expression of membrane antigens on human alveolar macrophages
22 after exposure to nitrogen dioxide. Inhalation Toxicol. 6: 633-646.
23 Gee, G. C.; Payne-Sturges, D. C. (2004) Environmental health disparities: a framework
24 integrating psychosocial and environmental concepts. Environ. Health Perspect. 112:
25 1645-1653.
26 Gehring, U.; Cyrys, J.; Sedlmeir, G.; Brunekreef, B.; Bellander, T.; Fischer, P.; Bauer, C. P.;
27 Reinhardt, D.; Wichmann, H. E.; Heinrich, J. (2002) Traffic-related air pollution and
28 respiratory health during the first 2 yrs. of life. Eur. Respir. J. 19: 690-698.
29 Gehring, U.; Heinrich, J.; Kramer, U.; Grote, V.; Hochadel, M.; Sugiri, D.; Kraft, M.; Rauchfuss,
30 K.; Eberwein, H. G.; Wichmann, H.-E. (2006) Long-term exposure to ambient air
31 pollution and cardiopulmonary mortality in women. Epidemiology 17: 545-551.
32 Gianetti, J.; Bevilacqua, S.; De Caterina, R. (2002) Inhaled nitric oxide: more than a selective
33 pulmonary vasodilator. Eur. J. Clin. Invest. 32: 628-635.
34 Gibson, T. L. (1983) Sources of direct-acting nitroarene mutagens in airborne particulate matter.
35 Mutat. Res. 122: 115-121.
36 Gilbert, N. L.; Goldberg, M. S.; Beckerman, B.; Brook, J. R.; Jerrett, M. (2005) Assessing spatial
37 variability of ambient nitrogen dioxide in Montreal, Canada, with a land-use regression
38 model. J. Air Waste Manage. Assoc. 55: 1059-1063.
39 Gilbert, N. L.; Gauvin, D.; Guay, M.; Heroux, M.-E.; Dupuis, G.; Legris, M.; Chan, C. C.; Dietz,
40 R. N.; Levesque, B. (2006) Housing characteristics and indoor concentrations of nitrogen
41 dioxide and formaldehyde in Quebec City, Canada. Environ. Res. 102: 1-8.
March 2008 6-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 Gilbert, N. L.; Goldberg, M. S.; Brook, J. R.; Jerrett, M. (2007) The influence of highway traffic
2 on ambient nitrogen dioxide concentration in the immediate vicinity of highways. Atmos.
3 Environ. 41: 2670-2673.
4 Gilliland, F. D.; McConnell, R.; Peters, J.; Gong, H, Jr. (1999) A theoretical basis for
5 investigation ambient air pollution and children's respiratory health. Environ. Health
6 Perspect. 107(suppl. 3): 403-407.
7 Gilliland, F. D.; Rappaport, E. B.; Berhane, K.; Islam, T.; Dubeau, L.; Gauderman, W. J.;
8 McConnell, R. (2002) Effects of glutathione S-Transferase PI, Ml, and Tl on acute
9 respiratory illness in school children. Am. J. Respir. Crit. Care Med. 166: 346-351.
10 Gilmour, M. L; Park, P.; Belgrade, M. K. (1996) Increased immune and inflammatory responses
11 to dust mite antigen in rats exposed to 5 ppm N02. Fundam. Appl. Toxicol. 31: 65-70.
12 Girman, J. R.; Apte, M. G.; Traynor, G. W.; Allen, J. R.; Hollowell, C. D. (1982) Pollutant
13 emission rates from indoor combustion appliances and sidestream cigarette smoke.
14 Environ. Int. 8: 213-221.
15 Goings, S. A. J.; Kulle, T. J.; Bascom, R.; Sauder, L. R.; Green, D. J.; Hebel, J. R.; Clements, M.
16 L. (1989) Effect of nitrogen dioxide exposure on susceptibility to influenza A virus
17 infection in healthy adults. Am. Rev. Respir. Dis. 139: 1075-1081.
18 Gong, H., Jr., Linn, W. S.; Clark, K. W.; Anderson, K. R.; Geller, M. D.; Sioutas, C. (2005)
19 Respiratory responses to exposures with fine particulates and nitrogen dioxide in the
20 elderly with and without COPD. Inhalation Toxicol. 17: 123-132.
21 Gonzales, M.; Quails, C.; Hudgens, E.; Neas, L. (2005) Characterization of a spatial gradient of
22 nitrogen dioxide across a United States-Mexico border city during winter. Sci. Total
23 Environ. 337: 163-173.
24 Gooch, P. C.; Luippold, H. E.; Creasia, D. A.; Brewen, J. G. (1977) Observations on mouse
25 chromosomes following nitrogen dioxide inhalation. Mutat. Res. 48: 117-119.
26 Goss, C. H.; Newsom, S. A.; Schildcrout, J. S.; Sheppard, L.; Kaufman, J. D. (2004) Effect of
27 ambient air pollution on pulmonary exacerbations and lung function in cystic fibrosis.
28 Am. J. Respir. Crit. Care Med. 169: 816-821.
29 Gouveia, N.; Fletcher, T. (2000) Respiratory diseases in children and outdoor air pollution in Sao
30 Paulo, Brazil: a time series analysis. Occup. Environ. Med. 57: 477-483.
31 Green, R. S.; Smorodinsky, S.; Kim, J. J.; McLaughlin, R.; Ostro, B. (2004) Proximity of
32 California public schools to busy roads. Environ. Health Perspect. 112: 61-66.
33 Grosovsky, A. J.; Sasaki, J. C.; Arey, J.; Eastmond, D. A.; Parks, K. K.; Atkinson, R. (1999)
34 Evaluation of the potential health effects of the atmospheric reaction products of
35 polycyclic aromatic hydrocarbons. Cambridge, MA: Health Effects Institute; research
36 report no. 84.
37 Gwynn, R. C.; Burnett, R. T.; Thurston, G. D. (2000) A time-series analysis of acidic particulate
38 matter and daily mortality and morbidity in the Buffalo, New York, region. Environ.
39 Health Perspect. 108: 125-133.
March 2008 6-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hackney, J. D.; Thiede, F. C.; Linn, W. S.; Pedersen, E. E.; Spier, C. E.; Law, D. C.; Fischer, D.
2 A. (1978) Experimental studies on human health effects of air pollutants. IV. Short-term
3 physiological and clinical effects of nitrogen dioxide exposure. Arch. Environ. Health 33:
4 176-181.
5 Hackney, J. D.; Linn, W. S.; Avol, E. L.; Shamoo, D. A.; Anderson, K. R.; Solomon, J. C.;
6 Little, D. E.; Peng, R.-C. (1992) Exposures of older adults with chronic respiratory illness
7 to nitrogen dioxide: a combined laboratory and field study. Am. Rev. Respir. Dis. 146:
8 1480-1486.
9 Hagen, J. A.; Nafstad, P.; Skrondal, A.; Bjorkly, S.; Magnus, P. (2000) Associations between
10 outdoor air pollutants and hospitalization for respiratory diseases. Epidemiology 11: 136-
11 140.
12 Hajat, S.; Haines, A.; Goubet, S. A.; Atkinson, R. W.; Anderson, H. R. (1999) Association of air
13 pollution with daily GP consultations for asthma and other lower respiratory conditions in
14 London. Thorax 54: 597-605.
15 Hansen, C.; Neller, A.; Williams, G.; Simpson, R. (2006) Maternal exposure to low levels of
16 ambient air pollution and preterm birth in Brisbane, Australia. BJOG 113: 935-941.
17 Harre, E. S. M.; Price, P. D.; Ayrey, R. B.; Toop, L. J.; Martin, I. R.; Town, G. I. (1997)
18 Respiratory effects of air pollution in chronic obstructive pulmonary disease: a three
19 month prospective study. Thorax 52: 1040-1044.
20 Hatch, G. E.; Slade, R.; Selgrade, M. K.; Stead, A. G. (1986) Nitrogen dioxide exposure and
21 lung antioxidants in ascorbic acid-deficient guinea pigs. Toxicol. Appl. Pharmacol. 82:
22 351-359.
23 Hayden, K. L.; Anlauf, K. G.; Hastie, D. R.; Bottenheim, J. W. (2003) Partitioning of reactive
24 atmospheric nitrogen oxides at an elevated site in southern Quebec, Canada. J. Geophys.
25 Res. [Atmos.] 108(D19): 10.1029/2002JD003188.
26 Hazenkamp-von Arx, M. E.; Gotschi, T.; Ackermann-Liebrich, U.; Bono, R.; Burney, P.; Cyrys,
27 J.; Jarvis, D.; Lillienberg, L.; Luczynska, C.; Maldonado, J. A.; Jaen, A.; de Marco, R.;
28 Mi, Y.; Modig, L.; Bayer-Oglesby, L.; Payo, F.; Soon, A.; Sunyer, J.; Villani, S.; Weyler,
29 J.; Kiinzli, N. (2004) PM2.5 and N02 assessment in 21 European study centres of ECRHS
30 II: annual means and seasonal differences. Atmos. Environ. 38: 1943-1953.
31 Hazucha, M. J.; Folinsbee, L. J.; Seal, E.; Bromberg, P. A. (1994) Lung function response of
32 healthy women after sequential exposures to N02 and Os. Am. J. Respir. Crit. Care Med.
33 150:642-647.
34 Heal, M. R.; O'Donoghue, M. A.; Cape, J. N. (1999) Overestimation of urban nitrogen dioxide
35 by passive diffusion tubes: a comparative exposure and model study. Atmos. Environ. 33:
36 513-524.
37 Helleday, R.; Huberman, D.; Blomberg, A.; Stjernberg, N.; Sandstrom, T. (1995) Nitrogen
38 dioxide exposure impairs the frequency of the mucociliary activity in healthy subjects.
39 Eur. Respir. J. 8: 1664-1668.
March 2008 6-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 Henneberger, A.; Zareba, W.; Ibald-Mulli, A.; Riickerl, R.; Cyrys, J.; Couderc, J.-P.; Mykins, B.;
2 Woelke, G.; Wichmann, H.-E.; Peters, A. (2005) Repolarization changes induced by air
3 pollution in ischemic heart disease patients. Environ. Health Perspect. 113: 440-446.
4 Henry, M. C.; Findlay, J.; Spangler, J.; Ehrlich, R. (1970) Chronic toxicity of N02 in squirrel
5 monkeys: III. effect on resistance to bacterial and viral infection. Arch. Environ. Health
6 20: 566-570.
7 Higgins, B. G.; Francis, H. C.; Yates, C. J.; Warburton, C. J.; Fletcher, A. M.; Reid, J. A.;
8 Pickering, C. A. C.; Woodcock, A. A. (1995) Effects of air pollution on symptoms and
9 peak expiratory flow measurements in subjects with obstructive airways disease. Thorax
10 50: 149-155.
11 Higgins, B. G.; Francis, H. C.; Yates, C.; Warburton, C. J.; Fletcher, A. M.; Pickering, C. A. C.;
12 Woodcock, A. A. (2000) Environmental exposure to air pollution and allergens and peak
13 flow changes. Eur. Respir. J. 16: 61- 66.
14 Hill, A. B. (1965) The environment and disease: association or causation? Proc. R. Soc. Med. 58:
15 295-300.
16 Hiltermann, T. J. N.; Stolk, J.; Van der Zee, S. C.; Brunekreef, B.; De Bruijne, C. R.; Fischer, P.
17 H.; Ameling, C. B.; Sterk, P. J.; Hiemstra, P. S.; Van Bree, L. (1998) Asthma severity
18 and susceptibility to air pollution. Eur. Respir. J. 11: 686-693.
19 Hinwood, A. L.; De Klerk, N.; Rodriguez, C.; Jacoby, P.; Runnion, T.; Rye, P.; Landau, L.;
20 Murray, F.; Feldwick, M.; Spickett, J. (2006) The relationship between changes in daily
21 air pollution and hospitalizations in Perth, Australia 1992-1998: a case-crossover study.
22 Int. J. Environ. Health Res. 16: 27-46.
23 Hirsch, T.; Weiland, S. K.; Von Mutius, E.; Safeca, A. F.; Grafe, H.; Csaplovics, E.; Duhme, H.;
24 Keil, U.; Leupold, W. (1999) Inner city air pollution and respiratory health and atopy in
25 children. Eur. Respir. J. 14: 669-677.
26 Hochadel, M.; Heinrich, J.; Gehring, U.; Morgenstern, V.; Kuhlbusch, T.; Link, E.; Wichmann,
27 H.-E.; Kramer, U. (2006) Predicting long-term average concentrations of traffic-related
28 air pollutants using CIS-based information. Atmos. Environ. 40: 542-553.
29 Hoek, G. (2003) Daily mortality and air pollution in The Netherlands. In: Revised analyses of
30 time-series studies of air pollution and health. Special report. Boston, MA: Health Effects
31 Institute; pp. 133-141. Available: http://www.healtheffects.org/Pubs/TimeSeries.pdf [12
32 May, 2004].
33 Hoek, G.; Brunekreef, B. (1994) Effects of low-level winter air pollution concentrations on
34 respiratory health of Dutch children. Environ. Res. 64: 136-150.
35 Hoek, G.; Brunekreef, B.; Verhoeff, A.; Van Wijnen, J.; Fischer, P. (2000) Daily mortality and
36 air pollution in the Netherlands. J. Air Waste Manage. Assoc. 50: 1380-1389.
37 Hoek, G.; Brunekreef, B.; Fischer, P.; Van Wijnen, J. (2001) The association between air
38 pollution and heart failure, arrhythmia, embolism, thrombosis, and other cardiovascular
39 causes of death in a time series study. Epidemiology 12: 355-357.
March 2008 6-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hoek, G.; Brunekreef, B.; Goldbohm, S.; Fischer, P.; Van den Brandt, P. A. (2002) Association
2 between mortality and indicators of traffic-related air pollution in the Netherlands: a
3 cohort study. Lancet 360: 1203-1209.
4 Holguin, F.; Tellez-Rojo, M. M.; Hernandez, M.; Cortez, M.; Chow, J. C.; Watson, J. G.;
5 Mannino, D.; Romieu, I. (2003) Air pollution and heart rate variability among the elderly
6 in Mexico City. Epidemiology 14: 521-527.
7 Horowitz, L. W.; Walters, S.; Mauzerall, D. L.; Emmons, L. K.; Rasch, P. J.; Granier, C.; Tie,
8 X.; Lamarque, J.-F.; Schultz, M. G.; Tyndall, G. S.; Orlando, J. J.; Brasseur, G. P. (2003)
9 A global simulation of tropospheric ozone and related tracers: description and evaluation
10 of MOZART, version 2. J. Geophys. Res. [Atmos.] 108(D24): 10.1029/2002JD002853.
11 Horowitz, L. W.; Fiore, A. M.; Milly, G. P.; Cohen, R. C.; Perring, A.; Wooldridge, P. J.; Hess,
12 P. G.; Emmons, L. K.; Lamarque, J. F. (2007) Observational constraints on the chemistry
13 of isoprene nitrates over the eastern United States. J. Geophys. Res. (Atmos.)
14 112(D12S08): 10.1029/2006JD007747.
15 Hubbard, A. K.; Symanowicz, P. T.; Thibodeau, M.; Thrall, R. S.; Schramm, C. M.; Cloutier, M.
16 M; Morris, J. B. (2002) Effect of nitrogen dioxide on ovalbumin-induced allergic airway
17 disease in a murine model. J. Toxicol. Environ. Health A 65: 1999-2005.
18 Hwang, J.-S.; Chan, C.-C. (2002) Effects of air pollution on daily clinic visits for lower
19 respiratory tract illness. Am. J. Epidemiol. 155: 1-10.
20 Ichinose, T.; Sagai, M. (1992) Combined exposure to N02, 03 and H2S04-aerosol and lung
21 tumor formation in rats. Toxicology 74: 173-184.
22 Ichinose, T.; Fujii, K.; Sagai, M. (1991) Experimental studies on tumor promotion by nitrogen
23 dioxide. Toxicology 67: 211-225.
24 Ilabaca, M.; Olaeta, I.; Campos, E.; Villaire, J.; Tellez-Rojo, M. M.; Romieu, I. (1999)
25 Association between levels of fine particulate and emergency visits for pneumonia and
26 other respiratory illnesses among children in Santiago, Chile. J. Air Waste Manage.
27 Assoc. 49: 154-163.
28 Inoue, H.; Fukunaga, A.; Okubo, S. (1981) Mutagenic effects of nitrogen dioxide combined with
29 methylurea and ethylurea in Drosophila melanogaster. Mutat. Res. 88: 281-290.
30 Institute of Medicine (IOM). (2007) Improving the presumptive disability decision-making
31 process for veterans. Washington, DC: The National Academy of Sciences. Available:
32 http://www.nap.edu/catalog.php?record_id=l 1908 (11 February, 2008).
33 International Agency for Research on Cancer (IARC). (2006) IARC monographs on the
34 evaluation of carcinogenic risks to humans: Preamble. Lyon, France: International
35 Agency for Research on Cancer. (IARC monographs on the evaluation of carcinogenic
36 risks to humans). Available:
37 http://monographs.iarc.fr/ENG/Preamble/CurrentPreamble.pdf (21 February, 2008).
38 Iqbal, Z. M. (1984) In-vivo nitrosation of amines in mice by inhaled nitrogen dioxide and
39 inhibition of biosynthesis of 7V-nitrosamines. In: O'Neill, I. K.; Von Borstel, R. C.; Miller,
40 C. T.; Long, J.; Bartsch, H., eds. N-nitroso compounds: occurrence, biological effects and
41 relevance to human cancer: proceedings of the VHIth international symposium on N-
March 2008 6-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 nitroso compounds; September 1983; Banff, Canada. Lyon, France: International Agency
2 for Research on Cancer; pp. 291-300. (IARC scientific publications no. 57).
3 Iqbal, Z. M.; Dahl, K.; Epstein, S. S. (1980) Role of nitrogen dioxide in the biosynthesis of
4 nitrosamines in mice. Science (Washington, DC) 207: 1475-1477.
5 Iqbal, Z. M.; Dahl, K.; Epstein, S. S. (1981) Biosynthesis of dimethylnitrosamine in
6 dimethylamine-treated mice after exposure to nitrogen dioxide. JNCIJ. Natl. Cancer Inst.
7 67: 137-141.
8 Islam, T.; Gauderman, W. J.; Berhane, K.; McConnell, R.; Avol, E.; Peters, J. M.; Gilliland, F.
9 D. (2007) The relationship between air pollution, lung function and asthma in
10 adolescents. Thorax 62: 957-963.
11 Isomura, K.; Chikahira, M.; Teranishi, K.; Hamada, K. (1984) Induction of mutations and
12 chromosome aberrations in lung cells following in vivo exposure of rats to nitrogen
13 oxides. Mutat. Res. 136: 119-125.
14 Ito, K. (2004) Revised ozone risk estimates for daily mortality and hospitalizations in Detroit,
15 Michigan [personal communication with attachments to Jee Young Kim]. New York,
16 NY: New York University School of Medicine, Nelson Institute of Environmental
17 Medicine; October 31.
18 Ito, K. (2007) Association between doarse particles and asthma emergency department (ED)
19 visits in New York City. Presented at: American Thoracic Society international
20 conference; San Francisco, CA.
21 Jacob, D. J. (1999) Introduction to atmospheric chemistry. Princeton, NJ: Princeton University
22 Press.
23 Jacobson, M. Z. (2002) Atmospheric pollution: history, science, and regulation. New York, NY:
24 Cambridge University Press.
25 Jaffe, D. H.; Singer, M. E.; Rimm, A. A. (2003) Air pollution and emergency department visits
26 for asthma among Ohio Medicaid recipients, 1991-1996. Environ. Res. 91: 21-28.
27 Jalaludin, B. B.; O'Toole, B. I.; Leeder, S. R. (2004) Acute effects of urban ambient air pollution
28 on respiratory symptoms, asthma medication use, and doctor visits for asthma in a cohort
29 of Australian children. Environ Res. 95: 32-42.
30 Jalaludin, B.; Morgan, G.; Lincoln, D.; Sheppeard, V.; Simpson, R.; Corbett, S. (2006)
31 Associations between ambient air pollution and daily emergency department attendances
32 for cardiovascular disease in the elderly (65+ years), Sydney, Australia. J. Exposure Sci.
33 Environ. Epidemiol. 16: 225-237.
34 Jarvis, D. L.; Leaderer, B. P.; Chinn, S.; Burney, P. G. (2005) Indoor nitrous acid and respiratory
35 symptoms and lung function in adults. Thorax 60: 474-479.
36 Jenkins, H. S.; Devalia, J. L.; Mister, R. L.; Bevan, A. M.; Rusznak, C.; Davies, R. J. (1999) The
37 effect of exposure to ozone and nitrogen dioxide on the airway response of atopic
38 asthmatics to inhaled allergen: dose- and time-dependent effects. Am. J. Respir. Crit.
39 Care Med. 160:33-39.
March 2008 6-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 Jerrett, M. (2007) Does traffic-related air pollution contribute to respiratory disease formation in
2 children? Eur. Respir. J. 29: 825-826.
3 Jerrett, M.; Finkelstein, M. (2005) Geographies of risk in studies linking chronic air pollution
4 exposure to health outcomes. J. Toxicol. Environ. Health Part A 68: 1207-1242.
5 Jerrett, M.; Burnett, R. T.; Brook, J.; Kanaroglou, P.; Giovis, C.; Finkelstein, N.; Hutchison, B.
6 (2004) Do socioeconomic characteristics modify the short term association between air
7 pollution and mortality? Evidence from a zonal time series in Hamilton, Canada. J.
8 Epidemiol. Community Health 58: 31-40.
9 Jet Propulsion Laboratory. (2006) Chemical kinetics and photochemical data for use in
10 atmospheric studies. Evaluation number 15. Pasadena, CA: California Institute of
11 Technology. JPL publication 06-2.
12 Jorres, R.; Magnussen, H. (1990) Airways response of asthmatics after a 30 min exposure, at
13 resting ventilation, to 0.25 ppm N02 or 0.5 ppm S02. Eur. Respir. J. 3: 132-137.
14 Jorres, R.; Magnussen, H. (1991) Effect of 0.25 ppm nitrogen dioxide on the airway response to
15 methacholine in asymptomatic asthmatic patients. Lung 169: 77-85.
16 Jorres, R.; Nowak, D.; Grimminger, F.; Seeger, W.; Oldigs, M.; Magnussen, H. (1995) The
17 effect of 1 ppm nitrogen dioxide on bronchoalveolar lavage cells and inflammatory
18 mediators in normal and asthmatic subjects. Eur. Respir. J. 8: 416-424.
19 Just, J.; Segala, C.; Sahraoui, F.; Priol, G.; Grimfeld, A.; Neukirch, F. (2002) Short-term health
20 effects of particulate and photochemical air pollution in asthmatic children. Eur. Respir.
21 J. 20: 899-906.
22 Karr, C.; Lumley, T.; Shepherd, K.; Davis, R.; Larson, T.; Ritz, B.; Kaufman, J. (2006) A case-
23 crossover study of wintertime ambient air pollution and infant bronchiolitis. Environ.
24 Health Perspect. 114: 277-281.
25 Katsouyanni, K.; Schwartz, J.; Spix, C.; Touloumi, G.; Zmirou, D.; Zanobetti, A.; Wojtyniak, B.;
26 Vonk, J. M.; Tobias, A.; Ponka, A.; Medina, S.; Bacharova, L.; Andersen, H. R. (1996)
27 Short term effects of air pollution on health: a European approach using epidemiology
28 time series data: the APHEA protocol. In: St Leger, S., ed. The APHEA project. Short
29 term effects of air pollution on health: a European approach using epidemiological time
30 series data. J. Epidemiol. Community Health 50(suppl. 1): S12-S18.
31 Katsouyanni, K.; Touloumi, G.; Samoli, E.; Gryparis, A.; Le Tertre, A.; Monopolis, Y.; Rossi,
32 G.; Zmirou, D.; Ballester, F.; Boumghar, A.; Anderson, H. R.; Wojtyniak, B.; Paldy, A.;
33 Braunstein, R.; Pekkanen, J.; Schindler, C.; Schwartz, J. (2001) Confounding and effect
34 modification in the short-term effects of ambient particles on total mortality: results from
35 29 European cities within the APHEA2 project. Epidemiology 12: 521-531.
36 Katsouyanni, K.; Touloumi, G.; Samoli, E.; Petasakis, Y.; Analitis, A.; Le Tertre, A.; Rossi, G.;
37 Zmirou, D.; Ballester, F.; Boumghar, A.; Anderson, H. R.; Wojtyniak, B.; Paldy, A.;
38 Braunstein, R.; Pekkanen, J.; Schindler, C.; Schwartz, J. (2003) Sensitivity analysis of
39 various models of short-term effects of ambient particles on total mortality in 29 cities in
40 APHEA2. In: Revised analyses of time-series studies of air pollution and health. Special
41 report. Boston, MA: Health Effects Institute; pp. 157-164. Available:
42 http://www.healtheffects.org/Pubs/TimeSeries.pdf [18 October, 2004].
March 2008 6-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 Rattan, M.; Gergen, P. J.; Eggleston, P.; Visness, C. M.; Mitchell, H. E. (2007) Health effects of
2 indoor nitrogen dioxide and passive smoking on urban asthmatic children. J. Allergy
3 Clin. Immunol. 120: 618-624.
4 Kauffmann, F.; Post Genome Respiratory Epidemiology Group. (2004) Post-genome respiratory
5 epidemiology: a multidisciplinary challenge. Eur. Respir. J. 24: 471-480.
6 Kawamoto, T.; Matsuno, K.; Arashidani, K.; Yoshikawa, M.; Kayama, F.; Kodama, Y. (1993)
7 Personal exposure to nitrogen dioxide from indoor heaters and cooking stoves. Arch.
8 Environ. Contam. Toxicol. 25: 534-538.
9 Kean, A. J.; Grosjean, E.; Grosjean, D.; Harley, R. A. (2001) On-road measurement of carbonyls
10 in California light-duty vehicle emissions. Environ. Sci. Technol. 35: 4198-4204.
11 Kelly, F. J.; Tetley, T. D. (1997) Nitrogen dioxide depletes uric acid and ascorbic acid but not
12 glutathione from lung lining fluid. Biochem. J. 325: 95-99.
13 Kelly, F. J.; Blomberg, A.; Frew, A.; Holgate, S. T.; Sandstrom, T. (1996) Antioxidant kinetics
14 in lung lavage fluid following exposure of humans to nitrogen dioxide. Am. J. Respir.
15 Crit. Care Med. 154: 1700-1705.
16 Kelly, F. J.; Dunster, C.; Mudway, I. (2003) Air pollution and the elderly: oxidant/antioxidant
17 issues worth consideration. Eur. Respir. J. Suppl. 40: 70S-75S.
18 Kerr, H. D.; Kulle, T. J.; Mcllhany, M. L.; Swidersky, P. (1979) Effects of nitrogen dioxide on
19 pulmonary function in human subjects: an environmental chamber study. Environ. Res.
20 19: 392-404.
21 Kesten, S.; Szalai, J.; Dzyngel, B. (1995) Air quality and the frequency of emergency room visits
22 for asthma. Ann. Allergy Asthma Immunol. 74: 269-273.
23 Kim, S. U.; Koenig, J. Q.; Pierson, W. E.; Hanley, Q. S. (1991) Acute pulmonary effects of
24 nitrogen dioxide exposure during exercise in competitive athletes. Chest 99: 815-819.
25 Kim, J. J.; Smorodinsky, S.; Lipsett, M.; Singer, B. C.; Hodgson, A. T.; Ostro, B. (2004a)
26 Traffic-related air pollution near busy roads: the East Bay children's Respiratory Health
27 Study. Am. J. Respir. Crit. Care Med. 170: 520-526.
28 Kim, S.-Y.; Lee, J.-T.; Hong, Y.-C.; Ahn, K.-J.; Kim, H. (2004b) Determining the threshold
29 effect of ozone on daily mortality: an analysis of ozone and mortality in Seoul, Korea,
30 1995-1999. Environ. Res. 94: 113-119.
31 Kim, E.; Hopke, P. K.; Pinto, J. P.; Wilson, W. E. (2005) Spatial variability of fine particle mass,
32 components, and source contributions during the regional air pollution study in St. Louis.
33 Environ. Sci. Technol. 39: 4172-4179.
34 Kim, D.; Sass-Kortsak, A.; Purdham, J. T.; Dales, R. E.; Brook, J. R. (2006) Associations
35 between personal exposures and fixed-site ambient measurements of fine particulate
36 matter, nitrogen dioxide, and carbon monoxide in Toronto, Canada. J. Exposure Sci.
37 Environ. Epidemiol. 16: 172-183.
38 Kim, S-Y.; O'Neill, M.; Lee, J-T.; Cho, Y.; Kim, J.; Kim, H. (2007) Air pollution,
39 socioeconomic position, and emergency hospital visits for asthma in Seoul, Korea. Int.
40 Arch. Occup. Environ. Health 80: 701-710.
March 2008 6-23 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kinney, P. L.; Ozkaynak, H. (1991) Associations of daily mortality and air pollution in Los
2 Angeles County. Environ. Res. 54: 99-120.
3 Kinsella, J. P.; Cutter, G. R.; Walsh, W. F.; Gerstmann, D. R.; Bose, C. L.; Hart, C.; Sekar, K.
4 C.; Auten, R. L.; Bhutani, V. K.; Gerdes, J. S.; George, T. N.; Southgate, W. M.;
5 Carriedo, H.; Couser, R. J.; Mammel, M. C.; Hall, D. C.; Pappagallo, M.; Sardesai, S.;
6 Strain, J. D.; Baier, M.; Abman, S. H. (2006) Early inhaled nitric oxide therapy in
7 premature newborns with respiratory failure. N. Engl. J. Med. 355: 354-364.
8 Kirby, C.; Fox, M.; Waterhouse, J.; Drye, T. (2001) Influences of environmental parameters on
9 the accuracy of nitrogen dioxide passive diffusion tubes for ambient measurement. J.
10 Environ. Monit. 3: 150-158.
11 Kitabatake, M.; Yamamoto, H.; Yuan, P. F.; Manjurul, H.; Murase, S.; Yamauchi, T. (1995)
12 Effects of exposure to N02 or S02 on bronchopulmonary reaction induced by Candida
13 albicans in guinea pigs. J. Toxicol. Environ. Health 45: 75-82.
14 Kleeberger, S. R. (2005) Genetic aspects of pulmonary responses to inhaled pollutants. Exp.
15 Toxicol. Pathol. 57(suppl. 1): 147-153.
16 Kleinman, M. T.; Mautz, W. J. (1991) The effects of exercise on dose and dose distribution of
17 inhaled automotive pollutants. Cambridge, MA: Health Effects Institute; research report
18 no. 45.
19 Kleinman, M. T.; Bailey, R. M.; Linn, W. S.; Anderson, K. R.; Whynot, J. D.; Shamoo, D. A.;
20 Hackney, J. D. (1983) Effects of 0.2 ppm nitrogen dioxide on pulmonary function and
21 response to bronchoprovocation in asthmatics. J. Toxicol. Environ. Health 12: 815-826.
22 Klepeis, N. E.; Nelson, W. C.; Ott. W. R.; Robinson, J. P. Tsang, A. M.; Switzer, P.; Behar, J.
23 V.; Hern, S. C.; Engelmann, W. H. (2001) The National Human Activity Pattern Survey
24 (NHAPS): a resource for assessing exposure to environmental pollutants. J. Exposure
25 Anal. Environ. Epidemiol. 11: 231-252.
26 Ko, F. W. S.; Tarn, W.; Wong, T. W.; Lai, C. K. W.; Wong, G. W. K.; Leung, T.-F.; Ng, S. S. S.;
27 Hui, D. S. C. (2007) Effects of air pollution on asthma hospitalization rates in different
28 age groups in Hong Kong. Clin. Exp. Allergy 37: 1312-1319.
29 Kobayashi, T.; Miura, T. (1995) Concentration- and time-dependent increase in specific airway
30 resistance after induction of airway hyperresponsiveness by subchronic exposure of
31 guinea pigs to nitrogen dioxide. Fundam. Appl. Toxicol. 25: 154-158.
32 Kochanek, K. D.; Murphy, S. L.; Anderson, R. N.; Scott, C. (2004) Deaths: final data for 2002.
33 Hyattsville, MD: U.S. Department of Health & Human Services, National Center for
34 Health Statistics; DHHS publication no. (PHS) 2005-1120. (National vital statistics
35 reports: v. 53, no. 5). Available:
36 http://www.cdc.gov/nchs/data/nvsr/nvsr53/nvsr53_05.pdf [3 August, 2005].
37 Kodama, Y.; Arashidani, K.; Tokui, N.; Kawamoto, T.; Matsuno, K.; Kunugita, N.; Minakawa,
38 N. (2002) Environmental NOz concentration and exposure in daily life along main roads
39 in Tokyo. Environ. Res. A 89: 236-244.
40 Koenig, J. Q.; Covert, D. S.; Pierson, W. E.; Hanley, Q. S.; Rebolledo, V.; Dumler, K.;
41 McKinney, S. E. (1994) Oxidant and acid aerosol exposure in healthy subjects and
March 2008 6-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 subjects with asthma. Part I: effects of oxidants, combined with sulfuric or nitric acid, on
2 the pulmonary function of adolescents with asthma. Cambridge, MA: Health Effects
3 Institute; pp. 1-36; research report no. 70.
4 Koutrakis, P.; Suh, H. H.; Sarnat, J. A.; Brown, K. W.; Coull, B. A.; Schwartz, J. (2005)
5 Characterization of particulate and gas exposures of sensitive populations living in
6 Baltimore and Boston. Boston, MA: Health Effects Institute; research report no. 131.
7 Kramer, U.; Koch, T.; Ranft, U.; Ring, J.; Behrendt, H. (2000) Traffic-related air pollution is
8 associated with atopy in children living in urban areas. Epidemiology 11: 64-70.
9 Krewski, D.; Burnett, R. T.; Goldberg, M. S.; Hoover, K.; Siemiatycki, J.; Jerrett, M.;
10 Abrahamowicz, M.; White, W. H. (2000) Reanalysis of the Harvard Six Cities study and
11 the American Cancer Society study of particulate air pollution and mortality: a special
12 report of the Institute's Particle Epidemiology Reanalysis Project. Cambridge, MA:
13 Health Effects Institute. Available: http://pubs.healtheffects.org/view.php?id=6 [6 March,
14 2007].
15 Kripke, B. J.; Sherwin, R. P. (1984) Nitrogen dioxide exposure - influence on rat testes. Anesth.
16 Analg. (NY) 63: 526-528.
17 Krupa, S. V.; Legge, A. H. (2000) Passive sampling of ambient, gaseous air pollutants: an
18 assessment from an ecological perspective. Environ. Pollut. 107: 31-45.
19 Kulkarni, M. M.; Patil, R. S. (2002) An empirical model to predict indoor NOz concentrations.
20 Atmos. Environ. 36: 4777-4785.
21 Kumae, T.; Arakawa, H. (2006) Comparison of effects of in vivo nitrogen dioxide exposure
22 starting from different periods on alveolar macrophage activity, assessed by a
23 chemiluminescence technique in Brown-Norway rats. Luminescence 21: 226-232.
24 Kunimoto, M.; Mochitate, K.; Kaya, K.; Miura, T.; Kubota, K. (1984) Effects of nitrogen
25 dioxide on red blood cells of rats: alterations of cell membrane components and
26 populational changes of red blood cells during in vivo exposure to N02. Environ. Res. 33:
27 361-369.
28 Kiinzli, N. (2005) Unifying susceptibility, exposure, and time: discussion of unifying analytic
29 approaches and future directions. J. Toxicol. Environ. Health A 68: 1263-1271.
30 Lagorio, S.; Forastiere, F.; Pistelli, R.; lavarone, I.; Michelozzi, P.; Fano, V.; Marconi, A.;
31 Ziemacki, G.; Ostro, B. D. (2006) Air pollution and lung function among susceptible
32 adult subjects: a panel study. Environ. Health: Global Access Sci. Source 5: 11.
33 Available: http://www.ehjournal.net/content/5/l/ll [16 January, 2006].
34 Lai, H. K.; Kendall, M.; Ferrier, H.; Lindup, I.; Aim, S.; Hanninen, 0.; Jantunen, M.; Mathys, P.;
35 Colvile, R.; Ashmore, M. R.; Cullinan, P.; Nieuwenhuijsen, M. J. (2004) Personal
36 exposures and microenvironment concentrations of PM2.5, VOC, NOz and CO in Oxford,
37 UK. Atmos. Environ. 38: 6399-6410.
38 Lai, S.; Patil, R. S. (2001) Monitoring of atmospheric behaviour of NOx from vehicular traffic.
39 Environ. Monit. Assess. 68: 37-50.
40 Lamsal, L. N.; Martin, R. V.; Van Donkelaar, A.; Steinbacher, M.; Celarier, E. A.; Bucsela, E.;
41 Dunlea, E. J.; Pinto, J. P. (2008) Ground-level nitrogen dioxide concentrations inferred
March 2008 6-25 DRAFT-DO NOT QUOTE OR CITE
-------
1 from the satellite-borne ozone monitoring instrument. J. Geophys. Res. (Atmos.): in
2 press.
3 La Rovere, M. T.; Pinna, G. D.; Maestri, R.; Mortara, A.; Capomolla, S.; Febo, 0.; Ferrari, R.;
4 Franchini, M.; Gnemmi, M.; Opasich, C.; Riccardi, P. G.; Traversi, E.; Cobelli, F. (2003)
5 Short-term heart rate variability strongly predicts sudden cardiac death in chronic heart
6 failure patients. Circulation 107: 565-570.
7 Last, J. A.; Gerriets, J. E.; Hyde, D. M. (1983) Synergistic effects on rat lungs of mixtures of
8 oxidant air pollutants (ozone or nitrogen dioxide) and respirable aerosols. Am. Rev.
9 Respir. Dis. 128: 539-544.
10 Leaderer, B. P.; Zagraniski, R. T.; Berwick, M.; Stolwijk, J. A. J. (1986) Assessment of exposure
11 to indoor air contaminants from combustion sources: methodology and application. Am.
12 J. Epidemiol. 124:275-289.
13 Lee, K.; Yanagisawa, Y.; Spengler, J. D.; Billick, I. H. (1995) Classification of house
14 characteristics based on indoor nitrogen dioxide concentrations. Environ. Int. 21: 277-
15 282.
16 Lee, K.; Yanagisawa, Y.; Spengler, J. D.; Fukumura, Y.; Billick, I. H. (1996) Classification of
17 house characteristics in a Boston residential nitrogen dioxide characterization study.
18 Indoor Air 6: 211-216.
19 Lee, K.; Levy, J. I.; Yanagisawa, Y.; Spengler, J. D.; Billick, I. H. (1998) The Boston Residential
20 Nitrogen Dioxide Characterization Study: classification and prediction of indoor N02
21 exposure. J. Air Waste Manage. Assoc. 48: 736-742.
22 Lee, S. C.; Chan, L. Y.; Chiu, M. Y. (1999) Indoor and outdoor air quality investigation at 14
23 public places in Hong Kong. Environ. Int. 25: 443-450.
24 Lee, K.; Yang, W.; Bofinger, N. D. (2000) Impact of microenvironmental nitrogen dioxide
25 concentrations on personal exposures in Australia. J. Air Waste Manage. Assoc. 50:
26 1739-1744.
27 Lee, K.; Xue, J.; Geyh, A. S.; Ozkaynak, H.; Leaderer, B. P.; Weschler, C. J.; Spengler, J. D.
28 (2002) Nitrous acid, nitrogen dioxide, and ozone concentrations in residential
29 environments. Environ. Health Perspect. 110: 145-150.
30 Lee, B. E.; Ha, E. H.; Park, H. S.; Kim, Y. J.; Hong, Y. C.; Kim, H.; Lee, J. T. (2003) Exposure
31 to air pollution during different gestational phases contributes to risks of low birth
32 weight. Hum. Reprod. 18: 638-643.
33 Lee, Y.-L.; Lin, Y.-C.; Lee, Y.-C.; Wang, J.-Y.; Hsiue, T.-R.; Guo, Y. L. (2004) Glutathione S-
34 transferase PI gene polymorphism and air pollution as interactive risk factors for
35 childhood asthma. Clin. Exp. Allergy 34: 1707-1713.
36 Lee, S. L.; Wong, W. H. S.; Lau, Y. L. (2006) Association between air pollution and asthma
37 admission among children in Hong Kong. Clin. Exp. Allergy 36: 1138-1146.
38 Lee, I.-M.; Tsai, S.-S.; Chang, C.-C.; Ho, C.-K.; Yang, C.-Y. (2007) Air pollution and hospital
39 admissions for chronic obstructive pulmonary disease in a tropical city: Kaohsiung,
40 Taiwan. Inhalation Toxicol. 19: 393-398.
March 2008 6-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 Leem, J.-H.; Kaplan, B. M.; Shim, Y. K.; Pohl, H. R.; Gotway, C. A.; Bullard, S. M.; Rogers, J.
2 F.; Smith, M. M.; Tylenda, C. A. (2006) Exposures to air pollutants during pregnancy
3 and preterm delivery. Environ. Health Perspect. 114: 905-910.
4 Le Tertre, A.; Quenel, P.; Eilstein, D.; Medina, S.; Prouvost, H.; Pascal, L.; Boumghar, A.;
5 Saviuc, P.; Zeghnoun, A.; Filleul, L.; Declercq, C.; Cassadou, S.; Le Coaster, C. (2002)
6 Short-term effects of air pollution on mortality in nine French cities: a quantitative
7 summary. Arch. Environ. Health 57: 311-319.
8 Levesque, B.; Allaire, S.; Gauvin, D.; Koutrakis, P.; Gingras, S.; Rhainds, M.; Prud'Homme, H.;
9 Duchesne, J.-F. (2001) Wood-burning appliances and indoor air quality. Sci. Total
10 Environ. 281: 47-62.
11 Levy, J. L; Lee, K.; Yanagisawa, Y.; Hutchinson, P.; Spengler, J. D. (1998a) Determinants of
12 nitrogen dioxide concentrations in indoor ice skating rinks. Am. J. Public Health 88:
13 1781-1786.
14 Levy, J. L; Lee, K.; Spengler, J. D.; Yanagisawa, Y. (1998b) Impact of residential nitrogen
15 dioxide exposure on personal exposure: an international study. J. Air Waste Manage.
16 Assoc. 48: 553-560.
17 Lewne, M.; Nise, G.; Lind, M. L.; Gustavsson, P. (2006) Exposure to particles and nitrogen
18 dioxide among taxi, bus and lorry drivers. Int. Arch. Occup. Environ. Health 79: 220-226.
19 Liao, D.; Duan, Y.; Whitsel, E. A.; Zheng, Z.-J.; Heiss, G.; Chinchilli, V. M.; Lin, H.-M. (2004)
20 Association of higher levels of ambient criteria pollutants with impaired cardiac
21 autonomic control: a population-based study. Am. J. Epidemiol. 159: 768-777.
22 Liao, D.; Heiss, G.; Chinchilli, V. M.; Duan, Y.; Folsom, A. R.; Lin, H. M.; Salomaa, V. (2005)
23 Association of criteria pollutants with plasma hemostatic/inflammatory markers: a
24 population-based study. J. Exposure Anal. Environ. Epidemiol. 15: 319-328.
25 Liard, R.; Zureik, M.; Le Moullec, Y.; Soussan, D.; Glorian, M.; Grimfeld, A.; Neukirch, F.
26 (1999) Use of personal passive samplers for measurement of N02, NO, and 03 levels in
27 panel studies. Environ. Res. 81: 339-348.
28 Lin, C. A.; Martins, M. A.; Farhat, S. C. L.; Pope, C. A., Ill; Conceigao, G. M. S.; Anastacio, V.
29 M.; Hatanaka, M.; Andrade, W. C.; Hamaue, W. R.; Bohm, G. M.; Saldiva, P. H. N.
30 (1999) Air pollution and respiratory illness of children in Sao Paulo, Brazil. Paediatr.
31 Perinat. Epidemiol. 13: 475-488.
32 Lin, M.; Chen, Y.; Burnett, R. T.; Villeneuve, P. J.; Krewski, D. (2003) Effect of short-term
33 exposure to gaseous pollution on asthma hospitalisation in children: a bi-directional case-
34 crossover analysis. J. Epidemiol. Community Health 57: 50-55.
35 Lin, M.; Chen, Y.; Villeneuve, P. J.; Burnett, R. T.; Lemyre, L.; Hertzman, C.; McGrail, K. M.;
36 Krewski, D. (2004) Gaseous air pollutants and asthma hospitalization of children with
37 low household income in Vancouver, British Columbia, Canada. Am. J. Epidemiol. 159:
38 294-303.
39 Lin, M.; Stieb, D. M.; Chen, Y. (2005) Coarse particulate matter and hospitalization for
40 respiratory infections in children younger than 15 years in Toronto: a case-crossover
41 analysis. Pediatrics 116: 235-240.
March 2008 6-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 Linaker, C. H.; Chauhan, A. J.; Inskip, H.; Frew, A. J.; Sillence, A.; Coggon, D.; Holgate, S. T.
2 (1996) Distribution and determinants of personal exposure to nitrogen dioxide in school
3 children. Occup. Environ. Med. 53: 200-203.
4 Linaker, C. H.; Chauhan, A. J.; Inskip, H. M.; Holgate, S. T.; Coggon, D. (2000) Personal
5 exposures of children to nitrogen dioxide relative to concentrations in outdoor air. Occup.
6 Environ. Med. 57: 472-476.
7 Linares, C.; Diaz, J.; Tobias, A.; Miguel, J. M. De.; Otero, A. (2006) Impact of urban air
8 pollutants and noise levels over daily hospital admissions in children in Madrid: a time
9 series analysis. Int. Arch. Occup. Environ. Health 79: 143-152.
10 Linn, W. S.; Solomon, J. C.; Trim, S. C.; Spier, C. E.; Shamoo, D. A.; Venet, T. G.; Avol, E. L.;
11 Hackney, J. D. (1985a) Effects of exposure to 4 ppm nitrogen dioxide in healthy and
12 asthmatic volunteers. Arch. Environ. Health 40: 234-239.
13 Linn, W. S.; Shamoo, D. A.; Spier, C. E.; Valencia, L. M.; Anzar, U. T.; Venet, T. G.; Avol, E.
14 L.; Hackney, J. D. (1985b) Controlled exposure of volunteers with chronic obstructive
15 pulmonary disease to nitrogen dioxide. Arch. Environ. Health 40: 313-317.
16 Linn, W. S.; Shamoo, D. A.; Anderson, K. R.; Peng, R.-C.; Avol, E. L.; Hackney, J. D.; Gong,
17 H., Jr. (1996) Short-term air pollution exposures and responses in Los Angeles area
18 schoolchildren. J. Exposure Anal. Environ. Epidemiol. 6: 449-472.
19 Linn, W. S.; Szlachcic, Y.; Gong, H., Jr.; Kinney, P. L.; Berhane, K. T. (2000) Air pollution and
20 daily hospital admissions in metropolitan Los Angeles. Environ. Health Perspect. 108:
21 427-434.
22 Lipfert, F. W.; Perry, H. M., Jr.; Miller, J. P.; Baty, J. D.; Wyzga, R. E.; Carmody, S. E. (2000)
23 The Washington University-EPRI veterans' cohort mortality study: preliminary results.
24 In: Grant, L. D., ed. PM2000: particulate matter and health. Inhalation Toxicol. 12(suppl.
25 4): 41-73.
26 Lipfert, F. W.; Perry, H. M., Jr.; Miller, J. P.; Baty, J. D.; Wyzga, R. E.; Carmody, S. E. (2003)
27 Air pollution, blood pressure, and their long-term associations with mortality. Inhalation
28 Toxicol. 15:493-512.
29 Lipfert, F. W.; Wyzga, R. E.; Baty, J. D.; Miller, J. P. (2006a) Traffic density as a surrogate
30 measure of environmental exposures in studies of air pollution health effects: long-term
31 mortality in a cohort of US veterans. Atmos. Environ. 40: 154-169.
32 Lipfert, F. W.; Baty, J. D.; Miller, J. P.; Wyzga, R. E. (2006b) PM2.5 constituents and related air
33 quality variables as predictors of survival in a cohort of U.S. military veterans. Inhalation
34 Toxicol. 18: 645-657.
35 Lipsett, M.; Hurley, S.; Ostro, B. (1997) Air pollution and emergency room visits for asthma in
36 Santa Clara County, California. Environ. Health Perspect. 105: 216-222.
37 Liu, S.; Krewski, D.; Shi, Y.; Chen, Y.; Burnett, R. T. (2003) Association between gaseous
38 ambient air pollutants and adverse pregnancy outcomes in Vancouver, Canada. Environ.
39 Health Perspect. Ill: 1773-1778.
March 2008 6-28 DRAFT-DO NOT QUOTE OR CITE
-------
1 Llorca, J.; Salas, A.; Prieto-Salceda, D.; Chinchon-Bengoechea, V.; Delgado-Rodriguez, M.
2 (2005) Nitrogen dioxide increases cardiorespiratory admissions in Torrelavega (Spain). J.
3 Environ. Health 68: 30-35.
4 Loupa, G.; Charpantidou, E.; Kioutsioukis, I.; Rapsomanikis, S. (2006) Indoor microclimate,
5 ozone and nitrogen oxides in two medieval churches in Cyprus. Atmos. Environ. 40:
6 7457-7466.
7 Love, G. J.; Lan, S.-P.; Shy, C. M.; Riggan, W. B. (1982) Acute respiratory illness in families
8 exposed to nitrogen dioxide ambient air pollution in Chattanooga, Tennessee. Arch.
9 Environ. Health 37: 75-80.
10 Luginaah, I. N.; Fung, K. Y.; Gorey, K. M.; Webster, G.; Wills, C. (2005) Association of
11 ambient air pollution with respiratory hospitalization in a government designated "area of
12 concern": the case of Windsor, Ontario. Environ. Health Perspect. 113: 290-296.
13 Luttmann-Gibson, H.; Suh, H. H.; Coull, B. A.; Dockery, D. W.; Sarnet, S. E.; Schwartz, J.;
14 Stone, P. H.; Gold, D. R. (2006) Short-term effects of air pollution on heart rate
15 variability in senior adults in Steubenville, Ohio. J. Occup. Environ. Med. 48: 780-788.
16 Magas, 0. K.; Gunter, J. T.; Regens, J. L. (2007) Ambient air pollution and daily pediatric
17 hospitalizations for asthma. Environ. Sci. Pollut. Res. 14: 19-23.
18 Mage, D.; Wilson, W.; Hasselblad, V.; Grant, L. (1999) Assessment of human exposure to
19 ambient particulate matter. J. Air Waste Manage. Assoc. 49: 1280-1291.
20 Makri, A.; Stilianakis, N. I. (2008) Vulnerability to air pollution health effects. Int. J. Hyg.
21 Environ. Health: in press.
22 Mann, J. K.; Tager, I. B.; Lurmann, F.; Segal, M.; Quesenberry, C. P., Jr.; Lugg, M. M.; Shan, J.;
23 Van den Eeden, S. K. (2002) Air pollution and hospital admissions for ischemic heart
24 disease in persons with congestive heart failure or arrhythmia. Environ. Health Perspect.
25 110:1247-1252.
26 Mannes, T.; Jalaludin, B.; Morgan, G.; Lincoln, D.; Sheppeard, V.; Corbett, S. (2005) Impact of
27 ambient air pollution on birth weight in Sydney, Australia. Occup. Environ. Med. 62:
28 524-530.
29 Maroziene, L.; Grazuleviciene, R. (2002) Maternal exposure to low-level air pollution and
30 pregnancy outcomes: a population-based study. Environ. Health: Global Access Sci.
31 Source 1: 6. Available: http://www.ehjournal.net/content/l/l/6 [10 October, 2006].
32 Martin, R. V.; Jacob, D. J.; Chance, K. V.; Kurosu, T. P.; Palmer, P. L; Evans, M. J. (2003)
33 Global inventory of nitrogen oxide emissions constrained by space-based observations of
34 N02 columns. J. Geophys. Res. [Atmos.] 108(D17): 10.1029/2003JD003453.
35 Martins, M. C. H.; Fatigati, F. L.; Vespoli, T. C.; Martins, L. C.; Martins, M. A.; Saldiva, P. H.
36 N.; Braga, A. L. F. (2004) Influence of socioeconomic conditions on air pollution effects
37 in elderly people an analysis of six regions in Sao Paolo, Brazil. J. Epidemiol. Comm.
38 Health 58: 41-46.
39 Maruo, Y. Y.; Ogawa, S.; Ichino, T.; Murao, N.; Uchiyama, M. (2003) Measurement of local
40 variations in atmospheric nitrogen dioxide levels in Sapporo, Japan, using a new method
41 with high spatial and high temporal resolution. Atmos. Environ. 37: 1065-1074.
March 2008 6-29 DRAFT-DO NOT QUOTE OR CITE
-------
1 McCaig, L. R; Burt, C. W. (2005) National Hospital Ambulatory Medical Care Survey: 2003
2 Emergency Department Summary. Hyattsville, MD: National Center for Health
3 Statistics; DHHS publication no. (PHS) 2005-1250. (Advance data from vital and health
4 statistics; no. 358). Available: http://www.cdc.gov/nchs/data/ad/ad358.pdf [3 August,
5 2005].
6 McClenny, W. A.; Williams, E. J.; Cohen, R. C.; Stutz, J. (2002) Preparing to measure the
7 effects of the NOX SIP Call—methods for ambient air monitoring of NO, NOz, N0y, and
8 individual NOZ species. J. Air Waste Manage. Assoc. 52: 542-562.
9 McConnell, R.; Berhane, K.; Gilliland, R; Molitor, J.; Thomas, D.; Lurmann, R; Avol, E.;
10 Gauderman, W. J.; Peters, J. M. (2003) Prospective study of air pollution and bronchitic
11 symptoms in children with asthma. Am. J. Respir. Crit. Care Med. 168: 790-797.
12 McConnell, R.; Berhane, K.; Yao, L.; Jerrett, M.; Lurmann, R; Gilliland, R; Kunzli, N.;
13 Gauderman, J.; Avol, E.; Thomas, D.; Peters, J. (2006) Traffic, susceptibility, and
14 childhood asthma. Environ. Health Perspect. 114: 766-772.
15 McCreanor, J.; Cullinan, P.; Nieuwenhuijsen, M. J.; Stewart-Evans, J.; Malliarou, E.; Jarup, L.;
16 Harrington, R.; Svartengren, M.; Han, I-K.; Ohman-Strickland, P.; Chung, K. R; Zhang,
17 J. (2007) Respiratory effects of exposure to diesel traffic in persons with asthma. N. Engl.
18 J. Med. 357: 2348-2358.
19 McCurdy, T.; Glen, G.; Smith, L.; Lakkadi, Y. (2000) The National Exposure Research
20 Laboratory's Consolidated Human Activity Database. J. Exposure Anal. Environ.
21 Epidemiol. 10: 566-578.
22 Mercer, R. R.; Costa, D. L.; Crapo, J. D. (1995) Effects of prolonged exposure to low doses of
23 nitric oxide or nitrogen dioxide on the alveolar septa of the adult rat lung. Lab. Invest. 73:
24 20-28.
25 Mersch, J.; Dyce, B. J.; Haverback, B. J.; Sherwin, R. P. (1973) Diphosphoglycerate content of
26 red blood cells: measurements in guinea pigs exposed to 0.4 ppm nitrogen dioxide. Arch.
27 Environ. Health 27: 94-95.
28 Metzger, K. B.; Tolbert, P. E.; Klein, M.; Peel, J. L.; Flanders, W. D.; Todd, K. H.; Mulholland,
29 J. A.; Ryan, P. B.; Frumkin , H. (2004) Ambient air pollution and cardiovascular
30 emergency department visits. Epidemiology 15: 46-56.
31 Metzger, K. B.; Klein, M.; Flanders, W. D.; Peel, J. L.; Mulholland, J. A.; Langberg, J. J.;
32 Tolbert, P. E. (2007) Ambient air pollution and cardiac arrhythmias in patients with
33 implantable defibrillators. Epidemiology 18: 585-592.
34 Migliaretti, G.; Cavallo, F. (2004) Urban air pollution and asthma in children. Pediatr. Pulmonol.
35 38: 198-203.
36 Migliaretti, G.; Cadum, E.; Migliore, E.; Cavallo, F. (2005) Traffic air pollution and hospital
37 admission for asthma: a case-control approach in a Turin (Italy) population. Int. Arch.
38 Occup. Environ. Health. 78: 164-169.
39 Miller, F. J.; Graham, J. A.; Raub, J. A.; Illing, J. W.; Menache, M. G.; House, D. E.; Gardner,
40 D. E. (1987) Evaluating the toxicity of urban patterns of oxidant gases. II. Effects in mice
41 from chronic exposure to nitrogen dioxide. J. Toxicol. Environ. Health 21: 99-112.
March 2008 6-30 DRAFT-DO NOT QUOTE OR CITE
-------
1 Miller, K. A.; Siscovick, D. S.; Sheppard, L.; Shepherd, K.; Sullivan J. H.; Anderson, G. L.;
2 Kaufman, J. D. (2007) Long-term exposure to air pollution and incidence of
3 cardiovascular events in women. N. Engl. J. Med. 356: 447-458.
4 Millstein, J.; Gilliland, F.; Berhane, K.; Gauderman, W. J.; McConnell, R.; Avol, E.; Rappaport,
5 E. B.; Peters, J. M. (2004) Effects of ambient air pollutants on asthma medication use and
6 wheezing among fourth-grade school children from 12 Southern California communities
7 enrolled in The Children's Health Study. Arch. Environ. Health 59: 505-514.
8 Mirvish, S. S.; Issenberg, P.; Sams, J. P. (1981) 7V-nitrosomorpholine synthesis in rodents
9 exposed to nitrogen dioxide and morpholine. In: Scanlan, R. A.; Tannenbaum, S. R., eds.
10 7V-nitroso compounds: based on a symposium cosponsored by the Divisions of
11 Agricultural and Food Chemistry and Pesticide Chemistry at the 181st meeting of the
12 American Chemical Society; March-April; Atlanta, GA. Washington, DC: American
13 Chemical Society; pp. 181-191. (ACS symposium series 174).
14 Mirvish, S. S.; Sams, J. P.; Issenberg, P. (1983) The nitrosating agent in mice exposed to
15 nitrogen dioxide: improved extraction method and localization in the skin. Cancer Res.
16 43: 2550-2554.
17 Mirvish, S. S.; Babcook, D. M.; Deshpande, A. D.; Nagel, D. L. (1986) Identification of
18 cholesterol as a mouse skin lipid that reacts with nitrogen dioxide to yield a nitrosating
19 agent, and of cholesteryl nitrite as the nitrosating agent produced in a chemical system
20 from cholesterol. Cancer Lett. (Shannon, Irel.) 31: 97-104.
21 Mirvish, S. S.; Ramm, M. D.; Sams, J. P.; Babcook, D. M. (1988) Nitrosamine formation from
22 amines applied to the skin of mice after and before exposure to nitrogen dioxide. Cancer
23 Res. 48: 1095-1099.
24 Mochitate, K.; Miura, T. (1984) In vivo effect of nitrogen dioxide on the activities of glycolytic
25 enzymes in red blood cells of rats. Toxicol. Lett. 22: 315-321.
26 Modig, L.; Sunesson, A.-L.; Levin, J.-O.; Sundgren, M.; Hagenbjork-Gustafsson, A.; Forsberg,
27 B. (2004) Can N02 be used to indicate ambient and personal levels of benzene and 1,3-
28 butadiene in air? J. Environ. Monit. 6: 957-962.
29 Mohsenin, V. (1987a) Effect of vitamin C on NCVinduced airway hyperresponsiveness in
30 normal subjects: a randomized double-blind experiment. Am. Rev. Respir. Dis. 136:
31 1408-1411.
32 Mohsenin, V. (1987b) Airway responses to nitrogen dioxide in asthmatic subjects. J. Toxicol.
33 Environ. Health 22: 371-380.
34 Mohsenin, V. (1988) Airway responses to 2.0 ppm nitrogen dioxide in normal subjects. Arch.
35 Environ. Health 43: 242-246.
36 Molitor, J.; Jerrett, M.; Chang, C.-C.; Molitor, N.-T.; Gauderman, J.; Berhane, K.; McConnell,
37 R.; Lurmann, F.; Wu, J.; Winer, A.; Thomas, D. (2007) Assessing uncertainty in spatial
38 exposure models for air pollution health effects assessment. Environ. Health Perspect.
39 115:1147-1153.
March 2008 6-31 DRAFT-DO NOT QUOTE OR CITE
-------
1 Monn, C. (2001) Exposure assessment of air pollutants: a review on spatial heterogeneity and
2 indoor/outdoor/personal exposure to suspended particulate matter, nitrogen dioxide and
3 ozone. Atmos. Environ. 35: 1-32.
4 Monn, C.; Fuchs, A.; Hogger, D.; Junker, M.; Kogelschatz, D.; Roth, N.; Wanner, H.-U. (1997)
5 Particulate matter less than 10 pm (PMio) and fine particles less than 2.5 pm (PM2.s):
6 relationships between indoor, outdoor and personal concentrations. Sci. Total Environ.
7 208: 15-21.
8 Monn, C.; Brandli, 0.; Schindler, C.; Ackermann-Liebrich, U.; Leuenberger, P.; SAPALDIA
9 team. (1998) Personal exposure to nitrogen dioxide in Switzerland. Sci. Total Environ.
10 215:243-251.
11 Moolgavkar, S. H. (2000) Air pollution and hospital admissions for diseases of the circulatory
12 system in three U.S. metropolitan areas. J. Air Waste Manage Assoc. 50: 1199-1206.
13 Moolgavkar, S. H. (2003) Air pollution and daily deaths and hospital admissions in Los Angeles
14 and Cook counties. In: Revised analyses of time-series studies of air pollution and health.
15 Special report. Boston, MA: Health Effects Institute; pp. 183-198. Available:
16 http://www.healtheffects.org/news.htm [16 May, 2003].
17 Morello-Frosch, R.; Shenassa, E. D. (2006) The environmental "riskscape" and social inequality:
18 implications for explaining maternal and child health disparities. Environ. Health
19 Perspect. 114: 1150-1153.
20 Morgan, G.; Corbett, S.; Wlodarczyk, J. (1998a) Air pollution and hospital admissions in
21 Sydney, Australia, 1990 to 1994. Am. J. Public Health 88: 1761-1766.
22 Morgan, G.; Corbett, S.; Wlodarczyk, J.; Lewis, P. (1998b) Air pollution and daily mortality in
23 Sydney, Australia, 1989 through 1993. Am. J. Public Health 88: 759-764.
24 Morris, R. D.; Naumova, E. N.; Munasinghe, R. L. (1995) Ambient air pollution and
25 hospitalization for congestive heart failure among elderly people in seven large US cities.
26 Am. J. Public Health 85: 1361-1365.
27 Morrow, P. E.; Utell, M. J.; Bauer, M. A.; Smeglin, A. M.; Frampton, M. W.; Cox, C.; Speers,
28 D. M.; Gibb, F. R. (1992) Pulmonary performance of elderly normal subjects and
29 subjects with chronic obstructive pulmonary disease exposed to 0.3 ppm nitrogen
30 dioxide. Am. Rev. Respir. Dis. 145: 291-300.
31 Mortimer, K. M.; Neas, L. M.; Dockery, D. W.; Redline, S.; Tager, I. B. (2002) The effect of air
32 pollution on inner-city children with asthma. Eur. Respir. J. 19: 699-705.
33 Moseler, M.; Hendel-Kramer, A.; Karmaus, W.; Forster, J.; Weiss, K.; Urbanek, R.; Kuehr, J.
34 (1994) Effect of moderate NOz air pollution on the lung function of children with
35 asthmatic symptoms. Environ. Res. 67: 109-124.
36 Moshammer, H.; Hutter, H.-P.; Hauck, H.; Neuberger, M. (2006) Low levels of air pollution
37 induce changes of lung function in a panel of schoolchildren. Eur. Respir. J. 27: 1138-
38 1143.
39 Mosqueron, L.; Momas, L; Le Moullec, Y. (2002) Personal exposure of Paris office workers to
40 nitrogen dioxide and fine particles. Occup. Environ. Med. 59: 550-555.
March 2008 6-32 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mukala, K.; Pekkanen, J.; Tiittanen, P.; Aim, S.; Salonen, R. 0.; Tuomisto, J. (1999) Personally
2 measured weekly exposure to N02 and respiratory health among preschool children. Eur.
3 Respir.J. 13: 1411-1417.
4 Mukala, K.; Aim, S.; Tiittanen, P.; Salonen, R. 0.; Jantunen, M.; Pekkanen, J. (2000) Nitrogen
5 dioxide exposure assessment and cough among preschool children. Arch. Environ.
6 Health. 55: 431-438.
7 Mukerjee, S.; Smith, L. A.; Norris, G. A.; Morandi, M. T.; Gonzales, M.; Noble, C. A.; Neas, L.
8 M.; Ozkaynak, A. H. (2004) Field method comparison between passive air sampling and
9 continuous monitors for VOCs and NOz in El Paso, Texas. J. Air Waste Manage. Assoc.
10 54:307-319.
11 Miiller, B.; Schafer, H.; Earth, P.; Von Wichert, P. (1994) Lung surfactant components in
12 bronchoalveolar lavage after inhalation of N02 as markers of altered surfactant
13 metabolism. Lung 172: 61-72.
14 Naess, 0.; Nafstad, P.; Aamodt, G.; Claussen, B.; Rosland, P. (2007) Relation between
15 concentration of air pollution and cause-specific mortality: four-year exposures to
16 nitrogen dioxide and particulate matter pollutants in 470 neighborhoods in Oslo, Norway.
17 Am. J. Epidemiol. 165: 435-443.
18 Nafstad, P.; Haheim, L. L.; Oftedal, B.; Gram, F.; Holme, L; Hjermann, I.; Leren, P. (2003)
19 Lung cancer and air pollution: a 27 year follow up of 16,209 Norwegian men. Thorax 58:
20 1071-1076.
21 Nafstad, P.; Haheim, L. L.; Wisloff, T.; Gram, F.; Oftedal, B.; Holme, L; Hjermann, I.; Leren, P.
22 (2004) Urban air pollution and mortality in a cohort of Norwegian men. Environ. Health
23 Perspect. 112: 610-605.
24 Nakai, S.; Nitta, H.; Maeda, K. (1995) Respiratory health associated with exposure to automobile
25 exhaust II. Personal NOz exposure levels according to distance from the roadside. J.
26 Exposure Anal. Environ. Epidemiol. 5: 125-136.
27 Nakajima, T.; Kusumoto, S. (1968) [Effect of nitrogen dioxide exposure on the contents of
28 reduced glutathione in mouse lung]. Osaka-Furitsu Koshu Eisei Kenkyusho Kenkyu
29 Hokoku Rodo Eisei Hen 6: 17-21.
30 National Acid Precipitation Program (NAPAP). (1991) The experience and legacy of NAPAP.
31 Report of the Oversight Review Board. Washington, DC: National Acid Precipitation
32 Program.
33 National Center for Health Statistics (NCHS). (2006a) Summary health statistics for U.S. adults:
34 National Health Interview Survey, 2004. Hyattsville, MD: U.S. Department of Health &
35 Human Services, Centers for Disease Control and Prevention. (Vital and Health
36 Statistics, Series 10, no. 228). Available:
37 http://www.cdc.gov/nchs/data/series/sr_10/srlO_228.pdf (14 September, 2007).
38 National Center for Health Statistics (NCHS). (2006b) Summary health statistics for U.S.
39 children: National Health Interview Survey, 2005. Hyattsville, MD: U.S. Department of
40 Health & Human Services, Centers for Disease Control and Prevention. (Vital and Health
41 Statistics, Series 10, no. 231). Available:
42 http://www.cdc.gov/nchs/data/series/sr_10/srlO_231.pdf (14 September, 2007).
March 2008 6-33 DRAFT-DO NOT QUOTE OR CITE
-------
1 National Center for Health Statistics (NCHS). (2006c) Asthma prevalence, health care use and
2 mortality: United States, 2003-05. Hyattsville, MD: U.S. Department of Health & Human
3 Services, Centers for Disease Control and Prevention. Available:
4 http://www.cdc.gov/nchs/products/pubs/pubd/hestats/ashtma03-05/asthma03-05.htm (26
5 February, 2008).
6 National Heart, Lung, and Blood Institute (NHLBI). (2007) Expert panel report 3: guidelines for
7 the diagnosis and management of asthma. Bethesda, MD: U.S. Department of Health and
8 Human Services, National Institutes of Health, National Asthma Education and
9 Prevention Program; NIH Publication No. 07-4051.
10 National Research Council. (2001) Research priorities for airborne particulate matter. III. Early
11 research progress. Washington, DC: National Academy Press. Available:
12 http://www.nap.edu/books/0309073375/html/ (4 June 2003).
13 National Research Council. (2004) Research priorities for airborne particulate matter. IV.
14 Continuing research progress. Washington, DC: National Academies Press. Available:
15 http://www.nap.edu/catalog.php?record_id=10957 [1 August, 2007].
16 National Toxicology Program. (2005) Report on carcinogens. 11th ed. Washington, DC:
17 National Toxicology Program.
18 Nazaroff, W. W.; Cass, G. R. (1986) Mathematical modeling of chemically reactive pollutants in
19 indoor air. Environ. Sci. Technol. 20: 924-934.
20 Nazaroff, W. W.; Weschler, C. J. (2004) Cleaning products and air fresheners: exposure to
21 primary and secondary air pollutants. Atmos. Environ. 38: 2841-2865.
22 Neidell, M. J. (2004) Air pollution, health, and socio-economic status: the effect of outdoor air
23 quality on childhood asthma. J. Health Econ. 23: 1209-1236.
24 Nerriere, E.; Zmirou-Navier, D.; Blanchard, 0.; Momas, I.; Ladner, J.; Le Moullec, Y.;
25 Personnaz, M.-B.; Lameloise, P.; Delmas, V.; Target, A.; Desqueyroux, H. (2005) Can
26 we use fixed ambient air monitors to estimate population long-term exposure to air
27 pollutants? The case of spatial variability in the Genotox ER study. Environ. Res. 97: 32-
28 42.
29 Ng, T. P.; Seet, C. S. R.; Tan, W. C.; Foo, S. C. (2001) Nitrogen dioxide exposure from domestic
30 gas cooking and airway response in asthmatic women. Thorax 56: 596-601.
31 Nicolai, T.; Carr, D.; Weiland, S. K.; Duhme, H.; Von Ehrenstein, 0.; Wagner, C.; Von Mutius,
32 E. (2003) Urban traffic and pollutant exposure related to respiratory outcomes and atopy
33 in a large sample of children. Eur. Respir. J. 21: 956-963.
34 Nieding, G. von; Wagner, H. M. (1977) Experimental studies on the short-term effect of air
35 pollutants on pulmonary function in man: two-hour exposure to N02, 03 and S02 alone
36 and in combination. In: Kasuga, S.; Suzuki, N.; Yamada, T.; Kimura, G.; Inagaki, K.;
37 Onoe, K., eds. Proceedings of the fourth international clean air congress; May; Tokyo,
38 Japan. Tokyo, Japan: Japanese Union of Air Pollution Prevention Associations; pp. 5-8.
39 Nieding, G. von; Wagner, H. M.; Krekeler, H.; Loellgen, H.; Fries, W.; Beuthan, A. (1979)
40 Controlled studies of human exposure to single and combined action of N02, 03, and
41 S02. Int. Arch. Occup. Environ. Health 43: 195-210.
March 2008 6-34 DRAFT-DO NOT QUOTE OR CITE
-------
1 Nieding, G. von; Wagner, H. M.; Casper, H.; Beuthan, A.; Smidt, U. (1980) Effect of
2 experimental and occupational exposure to N02 in sensitive and normal subjects. In: Lee,
3 S. D., ed. Nitrogen oxides and their effects on health. Ann Arbor, MI: Ann Arbor Science
4 Publishers, Inc.; pp. 315-331.
5 Nitschke, M.; Pilotto, L. S.; Attewell, R. G.; Smith, B. J.; Pisaniello, D.; Martin, J.; Ruffin, R. E.;
6 Killer, J. E. (2006) A cohort study of indoor nitrogen dioxide and house dust mite
7 exposure in asthmatic children. J. Occup. Environ. Med. 48: 462-469.
8 Norris, G.; Young-Pong, S. N.; Koenig, J. Q.; Larson, T. V.; Sheppard, L.; Stout, J. W. (1999)
9 An association between fine particles and asthma emergency department visits for
10 children in Seattle. Environ. Health Perspect. 107: 489-493.
11 Nunnermacker, L. J.; Imre, D.; Daum, P. H.; Kleinman, L.; Lee, Y.-N.; Lee, J. H.; Springston,
12 S. R.; Newman, L.; Weinstein-Lloyd, J.; Luke, W. T.; Banta, R.; Alvarez, R.; Senff, C.;
13 Sillman, S.; Holdren, M.; Keigley, G. W.; Zhou, X. (1998) Characterization of the
14 Nashville urban plume on July 3 and July 18, 1995. J. Geophys. Res. [Atmos.] 103:
15 28,129-28,148.
16 Nyberg, F.; Gustavsson, P.; Jarup, L.; Bellander, T.; Berglind, N.; Jakobsson, R.; Pershagen, G.
17 (2000) Urban air pollution and lung cancer in Stockholm. Epidemiology 11: 487-495.
18 Oftedal, B.; Nafstad, P.; Magnus, P.; Bj0rkly, S.; Skrondal, A. (2003) Traffic related air pollution
19 and acute hospital admission for respiratory diseases in Drammen, Norway 1995-2000.
20 Eur. J. Epidemiol. 18: 671-675.
21 Oftedal, B.; Brunekreef, B.; Nystad, W.; Madsen, C.; Walker, S.-E.; Nafstad, P. (2008)
22 Residential outdoor air pollution and lung function in schoolchildren. Epidemiology 19:
23 129-137.
24 Ogawa & Company. (2007) Ambient air passive sampler for NO-N02, NOx, S02, 03, NH3.
25 Pompano Beach, FL. Available: http://www.ogawausa.com/passive.html [18 July, 2007].
26 Ohashi, Y.; Nakai, Y.; Sugiura, Y.; Ohno, Y.; Okamoto, H.; Tanaka, A.; Kakinoki, Y.; Hayashi,
27 M. (1994) Nitrogen dioxide-induced eosinophilia and mucosal injury in the nose of the
28 guinea pig. Acta Oto Laryngol. 114: 547-551.
29 Ohyama, K.; Ito, T.; Kanisawa, M. (1999) The roles of diesel exhaust particle extracts and the
30 promotive effects of N02 and/or S02 exposure on rat lung tumorigenesis. Cancer Lett.
31 139: 189-197.
32 O'Neill, M. S.; Jerrett, M.; Kawachi, L; Levy, J. I.; Cohen, A. J.; Gouveia, N.; Wilkinson, P.;
33 Fletcher, T.; Cifuentes, L.; Schwartz, J.; with input from participants of the Workshop on
34 Air Pollution and Socioeconomic Conditions. (2003) Health, wealth, and air pollution:
35 advancing theory and methods. Environ. Health Perspect. Ill: 1861-1870.
36 Ostro, B.; Lipsett, M.; Mann, J.; Braxton-Owens, H.; White, M. (2001) Air pollution and
37 exacerbation of asthma in African-American children in Los Angeles. Epidemiology 12:
38 200-208.
39 Ott, W.; Wallace, L.; Mage, D. (2000) Predicting particulate (PMio) personal exposure
40 distributions using a random component superposition statistical model. J. Air Waste
41 Manage. Assoc. 50: 1390-1406.
March 2008 6-35 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pagani, P.; Romano, M.; Erroi, A.; Ferro, M.; Salmona, M. (1994) Biochemical effects of acute
2 and subacute nitrogen dioxide exposure in rat lung and bronchoalveolar lavage fluid.
3 Arch. Environ. Contam. Toxicol. 27: 426-430.
4 Palmes, E. D.; Gunnison, A. F.; DiMattio, J.; Tomczyk, C. (1976) Personal sampler for nitrogen
5 dioxide. Am. Ind. Hyg. Assoc. J. 37: 570-577.
6 Pantazopoulou, A.; Katsouyanni, K.; Kourea-Kremastinou, J.; Trichopoulos, D. (1995) Short-
7 term effects of air pollution on hospital emergency outpatient visits and admissions in the
8 greater Athens, Greece area. Environ. Res. 69: 31-36.
9 Park, J.-H.; Spengler, J. D.; Yoon, D.-W.; Dumyahn, T.; Lee, K.; OOzkaynak, H. (1998)
10 Measurement of air exchange rate of stationary vehicles and estimation of in-vehicle
11 exposure. J. Exposure Anal. Environ. Epidemiol. 8: 65-78.
12 Park, J. W.; Lim, Y. H.; Kyung, S. Y.; An, C. H.; Lee, S. P.; Jeong, S. H.; Ju, S.-Y. (2005a)
13 Effects of ambient particulate matter on peak expiratory flow rates and respiratory
14 symptoms of asthmatics during Asian dust periods in Korea. Respirology 10: 470-476.
15 Park, S. K.; O'Neill, M. S.; Vokonas, P. S.; Sparrow, D.; Schwartz, J. (2005b) Effects of air
16 pollution on heart rate variability: the VA normative aging study. Environ. Health
17 Perspect. 113: 304-309.
18 Parrish, D. D.; Fehsenfeld, F. C. (2000) Methods for gas-phase measurements of ozone, ozone
19 precursors and aerosol precursors. Atmos. Environ. 34: 1921-1957.
20 Pathmanathan, S.; Krishna, M. T.; Blomberg, A.; Helleday, R.; Kelly, F. J.; Sandstrom, T.;
21 Holgate, S. T.; Wilson, S. J.; Frew, A. J. (2003) Repeated daily exposure to 2 ppm
22 nitrogen dioxide upregulates the expression of IL-5, IL-10, IL-13, and ICAM-1 in the
23 bronchial epithelium of healthy human airways. Occup. Environ. Med. 60: 892-896.
24 Pattemore, P. K.; Asher, M. I.; Harrison, A. C.; Mitchell, E. A.; Rea, H. H.; Stewart, A. W.
25 (1990) The interrelationship among bronchial hyperresponsiveness, the diagnosis of
26 asthma, and asthma symptoms. Am. Rev. Respir. Dis. 142: 549-554.
27 Peacock, J. L.; Symonds, P.; Jackson, P.; Bremner, S. A.; Scarlett, J. F.; Strachan, D. P.;
28 Anderson, H. R. (2003) Acute effects of winter air pollution on respiratory function in
29 schoolchildren in southern England. Occup. Environ. Med. 60: 82-89.
30 Peel, J. L.; Tolbert, P. E.; Klein, M.; Metzger, K. B.; Flanders, W. D.; Knox, T.; Mulholland, J.
31 A.; Ryan, P. B.; Frumkin, H. (2005) Ambient air pollution and respiratory emergency
32 department visits. Epidemiology 16: 164-174.
33 Peel, J. L.; Metzger, K. B.; Klein, M.; Flanders, W. D.; Mulholland, J. A.; Tolbert, P. E. (2006)
34 Ambient air pollution and cardiovascular emergency department visits in potentially
35 sensitive groups. Am. J. Epidemiol. 165: 625-633.
36 Peel, J. L.; Metzger, K. B.; Klein, M.; Flanders, W. D.; Mulholland, J. A.; Tolbert, P. E. (2007)
37 Ambient air pollution and cardiovascular emergency department visits in potentially
38 sensitive groups. Am. J. Epidemiol. 165: 625-633.
39 Pekkanen, J.; Brunner, E. J.; Anderson, H. R.; Tiittanen, P.; Atkinson, R. W. (2000) Daily
40 concentrations of air pollution and plasma fibrinogen in London. Occup. Environ. Med.
41 57:818-822.
March 2008 6-36 DRAFT-DO NOT QUOTE OR CITE
-------
1 Perez-Padilla, R.; Regalado, J.; Rojas, M.; Catalan, M.; Mendoza, L.; Rojas, R.; Chapela, R.;
2 Villalba, J.; Torres, V.; Borja, V.; et al. (2003a) Spirometric function in children of
3 Mexico City compared to Mexican-American children. Pediatr. Pulmonol. 35: 177-183.
4 Perez-Padilla, R.; Regalado-Pineda, J.; Mendoza, L.; Rojas, R.; Torres, V.; Borja-Aburto, V.;
5 Olaiz, G.; EMPECE Study Group. (2003b) Spirometric variability in a longitudinal study
6 of school-age children. Chest 123: 1090-1095.
7 Perrini, G.; Tomasello, M.; Librando, V.; Minniti, Z. (2005) Nitrated polycyclic aromatic
8 hydrocarbons in the environment: formation, occurrences and analysis. Ann. Chim. 95:
9 567-577.
10 Peters, J. M.; Avol, E.; Navidi, W.; London, S. J.; Gauderman, W. J.; Lurmann, F.; Linn, W. S.;
11 Margolis, H.; Rappaport, E.; Gong, H., Jr.; Thomas, D. C. (1999) A study of twelve
12 southern California communities with differing levels and types of air pollution. I.
13 Prevalence of respiratory morbidity. Am. J. Respir. Crit. Care Med. 159: 760-767.
14 Peters, A.; Liu, E.; Verrier, R. L.; Schwartz, J.; Gold, D. R.; Mittleman, M.; Baliff, J.; Oh, J. A.;
15 Allen, G.; Monahan, K.; Dockery, D. W. (2000) Air pollution and incidence of cardiac
16 arrhythmia. Epidemiology 11: 11-17.
17 Petroschevsky, A.; Simpson, R. W.; Thalib, L.; Rutherford, S. (2001) Associations between
18 outdoor air pollution and hospital admissions in Brisbane, Australia. Arch. Environ.
19 Health 56: 37-52.
20 Piechocki-Minguy, A.; Plaisance, H.; Schadkowski, C.; Sagnier, I.; Saison, J. Y.; Galloo, J. C.;
21 Guillermo, R. (2006) A case study of personal exposure to nitrogen dioxide using a new
22 high sensitive diffusive sampler. Sci. Total Environ. 366: 55-64.
23 Pilotto, L. S.; Douglas, R. M.; Attewell, R. G.; Wilson, S. R. (1997a) Respiratory effects
24 associated with indoor nitrogen dioxide exposure in children. Int. J. Epidemiol. 26: 788-
25 796.
26 Pilotto, L. S.; Douglas, R. M.; Samet, J. M. (1997b) Nitrogen dioxide, gas heating and
27 respiratory illness. Med. J. Aust. 167: 295-296.
28 Pilotto, L. S.; Nitschke, M.; Smith, B. J.; Pisaniello, D.; Ruffin, R. E.; McElroy, H. J.; Martin, J.;
29 Killer, J. E. (2004) Randomized controlled trial of unflued gas heater replacement on
30 respiratory health of asthmatic schoolchildren. Int. J. Epidemiol. 33: 208-214.
31 Pino, P.; Walter, T.; Oyarzun, M.; Villegas, R.; Romieu, I. (2004) Fine particulate matter and
32 wheezing illnesses in the first year of life. Epidemiology 15: 702-708.
33 Pinto, J. P.; Lefohn, A. S.; Shadwick, D. S. (2004) Spatial variability of PM2.5 in urban areas in
34 the United States. J. Air Waste Manage. Assoc. 54: 440-449.
35 Pitts, J. N., Jr. (1987) Nitration of gaseous polycyclic aromatic hydrocarbons in simulated and
36 ambient urban atmospheres: a source of mutagenic nitroarenes. Atmos. Environ. 21:
37 2531-2547.
38 Plaisance, H.; Piechocki-Minguy, A.; Garcia-Fouque, S.; Galloo, J. C. (2004) Influences of
39 meteorological factors on the N02 measurements by passive diffusion tube. Atmos.
40 Environ. 38: 573-580.
March 2008 6-37 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pleijel, H.; Karlsson, G. P.; Gerdin, E. B. (2004) On the logarithmic relationship between N02
2 concentration and the distance from a highroad. Sci. Total Environ. 332: 261-264.
3 Poloniecki, J. D.; Atkinson, R. W.; Ponce de Leon, A.; Anderson, H. R. (1997) Daily time series
4 for cardiovascular hospital admissions and previous day's air pollution in London, UK.
5 Occup. Environ. Med. 54: 535-540.
6 Ponce, N. A.; Hoggatt, K. J.; Wilhelm, M.; Ritz, B. (2005) Preterm birth: the interaction of
7 traffic-related air pollution with economic hardship in Los Angeles neighborhoods. Am.
8 J. Epidemiol. 162: 140-148.
9 Ponce de Leon, A.; Anderson, H. R.; Bland, J. M.; Strachan, D. P.; Bower, J. (1996) Effects of
10 air pollution on daily hospital admissions for respiratory disease in London between
11 1987-88 and 1991-92. In: St Leger, S., ed. The APHEA project. Short term effects of air
12 pollution on health: a European approach using epidemiological time series data. J.
13 Epidemiol. Community Health 50(suppl. 1): S63-S70.
14 Ponka, A. (1991) Asthma and low level air pollution in Helsinki. Arch. Environ. Health 46: 262-
15 270.
16 Ponka, A.; Virtanen, M. (1994) Chronic bronchitis, emphysema, and low-level air pollution in
17 Helsinki, 1987-1989. Environ. Res. 65: 207-217.
18 Ponka, A.; Virtanen, M. (1996) Asthma and ambient air pollution in Helsinki. In: St Leger, S.,
19 ed. The APHEA project. Short term effects of air pollution on health: a European
20 approach using epidemiological time series data. J. Epidemiol. Community Health
21 50(suppl. 1):S59-S62.
22 Ponsonby, A.-L.; Glasgow, N.; Gatenby, P.; Mullins, R.; McDonald, T.; Hurwitz, M.; Pradith,
23 B.; Attewell, R. (2001) The relationship between low level nitrogen dioxide exposure and
24 child lung function after cold air challenge. Clin. Exp. Allergy 31: 1205-1212.
25 Pope, C. A., Ill; Thun, M. J.; Namboodiri, M. M.; Dockery, D. W.; Evans, J. S.; Speizer, F. E.;
26 Heath, C. W., Jr. (1995) Particulate air pollution as a predictor of mortality in a
27 prospective study of U.S. adults. Am. J. Respir. Crit. Care Med. 151: 669-674.
28 Pope, C. A., Ill; Burnett, R. T.; Thun, M. J.; Calle, E. E.; Krewski, D.; Ito, K.; Thurston, G. D.
29 (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine
30 particulate air pollution. JAMA J. Am. Med. Assoc. 287: 1132-1141.
31 Posin, C.; Clark, K.; Jones, M. P.; Patterson, J. V.; Buckley, R. D.; Hackney, J. D. (1978)
32 Nitrogen dioxide inhalation and human blood biochemistry. Arch. Environ. Health 33:
33 318-324.
34 Postlethwait, E. M.; Bidani, A. (1990) Reactive uptake governs the pulmonary air space removal
35 of inhaled nitrogen dioxide. J. Appl. Physiol. 68: 594-603.
36 Postlethwait, E. M.; Bidani, A. (1994) Mechanisms of pulmonary NOz absorption. Toxicology
37 89: 217-237.
38 Postlethwait, E. M.; Langford, S. D.; Bidani, A. (1991) Transfer of N02 through pulmonary
39 epithelial lining fluid. Toxicol. Appl. Pharmacol. 109: 464-471.
March 2008 6-38 DRAFT-DO NOT QUOTE OR CITE
-------
1 Postlethwait, E. M.; Langford, S. D.; Jacobson, L. M.; Bidani, A. (1995) NOz reactive absorption
2 substrates in rat pulmonary surface lining fluids. Free Radical Biol. Med. 19: 553-563.
3 Prescott, G. J.; Cohen, G. R.; Elton, R. A.; Fowkes, F. G. R.; Agius, R. M. (1998) Urban air
4 pollution and cardiopulmonary ill health: a 14.5 year time series study. Occup. Environ.
5 Med. 55: 697-704.
6 Proust, B.; Lacroix, G.; Robidel, F.; Marliere, M.; Lecomte, A.; Vargaftig, B. B. (2002)
7 Interference of a short-term exposure to nitrogen dioxide with allergic airways responses
8 to allergenic challenges in BALB/c mice. Mediators Inflammation 11: 251-260.
9 Raaschou-Nielsen, 0.; Skov, H.; Lohse, C.; Thomsen, B. L.; Olsen, J. H. (1997) Front-door
10 concentrations and personal exposures of Danish children to nitrogen dioxide. Environ.
11 Health Perspect. 105: 964-970.
12 Ramsay, T. 0.; Burnett, R. T.; Krewski, D. (2003) The effect of concurvity in generalized
13 additive models linking mortality to ambient particulate matter. Epidemiology 14: 18-23.
14 Ranzi, A.; Gambini, M.; Spattini, A.; Galassi, C.; Sesti, D.; Bedeschi, M.; Messori, A.; Baroni,
15 A.; Cavagni, G.; Lauriola, P. (2004) Air pollution and respiratory status in asthmatic
16 children: hints for a locally based preventive strategy. AIRE study. Eur. J. Epidemiol. 19:
17 567-576.
18 Rasmussen, T. R.; Kjaergaard, S. K.; Tarp, U.; Pedersen, 0. F. (1992) Delayed effects of N02
19 exposure on alveolar permeability and glutathione peroxidase in healthy humans. Am.
20 Rev. Respir. Dis. 146: 654-659.
21 Raw, G. J.; Coward, S. K. D.; Brown, V. M.; Crump, D. R. (2004) Exposure to air pollutants in
22 English homes. J. Exposure Anal. Environ. Epidemiol. 14(suppl. 1): S85-S94.
23 Rehn, T.; Svartengren, M.; Philipson, K.; Camner, P. (1982) Mukociliaer transport i lunga och
24 naesa samt luftvaegsmotstand efter exponering foer kvaevedioxid [Mucociliary transport
25 in the lung and nose after exposure to nitrogen dioxide]. Vallingby, Sweden: Swedish
26 State Power Board; project KHM technical report no. 40.
27 Reisen, F.; Arey, J. (2005) Atmospheric reactions influence seasonal PAH and nitro-PAH
28 concentrations in the Los Angeles basin. Environ. Sci. Technol. 39: 64-73.
29 Restrepo, C.; Zimmerman, R.; Thurston, G.; Clemente, J.; Gorczynski, J.; Zhong, M.; Blaustein,
30 M.; Chen, L. C. (2004) A comparison of ground-level air quality data with New York
31 State Department of Environmental Conservation monitoring stations data in South
32 Bronx, New York. Atmos. Environ. 38: 5295-5304.
33 Rich, D. Q.; Schwartz, J.; Mittleman, M. A.; Link, M.; Luttmann-Gibson, H.; Catalano, P. J.;
34 Speizer, F. E.; Dockery, D. W. (2005) Association of short-term ambient air pollution
35 concentrations and ventricular arrhythmias. Am. J. Epidemiol. 161: 1123-1132.
36 Rich, D. Q.; Kim, M. H.; Turner, J. R.; Mittleman, M. A.; Schwartz, J.; Catalano, P. J.; Dockery,
37 D. W. (2006a) Association of ventricular arrhythmias detected by implantable
38 cardioverter defibrillator and ambient air pollutants in the St Louis, Missouri
39 metropolitan area. Occup. Environ. Med. 63: 591-596.
40 Rich, D. Q.; Mittleman, M. A.; Link, M. S.; Schwartz, J.; Luttmann-Gibson, H.; Catalano, P. J.;
41 Speizer, F. E.; Gold, D. R.; Dockery, D. W. (2006b) Increased risk of paroxysmal atrial
March 2008 6-39 DRAFT-DO NOT QUOTE OR CITE
-------
1 fibrillation episodes associated with acute increases in ambient air pollution. Environ.
2 Health Perspect. 114: 120-123.
3 Richters, A.; Damji, K. S. (1990) The relationship between inhalation of nitrogen dioxide, the
4 immune system, and progression of a spontaneously occurring lymphoma in AKR mice.
5 J. Environ. Pathol. Toxicol. Oncol. 10: 225-230.
6 Richters, A.; Kuraitis, K. (1981) Inhalation of N02 and blood borne cancer cell spread to the
7 lungs. Arch. Environ. Health 36: 36-39.
8 Richters, A.; Kuraitis, K. (1983) Air pollutants and the facilitation of cancer metastasis. Environ.
9 Health Perspect. 52: 165-168.
10 Richters, A.; Richters, V. (1983) A new relationship between air pollutant inhalation and cancer.
11 Arch. Environ. Health 38: 69-75.
12 Richters, A.; Richters, V.; Alley, W. P. (1985) The mortality rate from lung metastases in
13 animals inhaling nitrogen dioxide (NOz). J. Surg. Oncol. 28: 63-66.
14 Riediker, M.; Williams, R.; Devlin, R.; Griggs, T.; Bromberg, P. (2003) Exposure to particulate
15 matter, volatile organic compounds, and other air pollutants inside patrol cars. Environ.
16 Sci. Technol. 37: 2084-2093.
17 Ristovski, Z. D.; Tass, I.; Morawska, L.; Saxby, W. (2000) Investigation into the emission of fine
18 particles, formaldehyde, oxides of nitrogen and carbon monoxide from natural gas
19 heaters. J. Aerosol. Sci. 31 (suppl. 1): S490-S491.
20 Ritz, B.; Yu, F.; Chapa, G.; Fruin, S. (2000) Effect of air pollution on preterm birth among
21 children born in Southern California between 1989 and 1993. Epidemiology 11: 502-511.
22 Robinson, J. R. (1989) On uncertainty in the computation of global emissions for biomass
23 burning, dim. Change 14: 243-262.
24 Robison, T. W.; Murphy, J. K.; Beyer, L. L.; Richters, A.; Forman, H. J. (1993) Depression of
25 stimulated arachidonate metabolism and superoxide production in rat alveolar
26 macrophages following in vivo exposure to 0.5 ppm N02. J. Toxicol. Environ. Health 38:
27 273-292.
28 Rodgers, M. 0.; Davis, D. D. (1989) A UV-photofragmentation/laser-induced fluorescence
29 sensor for the atmospheric detection of HONO. Environ. Sci. Technol. 23: 1106-1112.
30 Roemer, W.; Hoek, G.; Brunekreef, B.; Haluszka, J.; Kalandidi, A.; Pekkanen, J. (1998) Daily
31 variations in air pollution and respiratory health in a multicentre study: the PEACE
32 project. Eur. Respir. J. 12: 1354-1361.
33 Roemer, W.; Clench-Aas, J.; Englert, N.; Hoek, G.; Katsouyanni, K.; Pekkanen, J.; Brunekreef,
34 B. (1999) Inhomogeneity in response to air pollution in European children (PEACE
35 project). Occup. Environ. Med. 56: 86-92.
36 Roger, L. J.; Horstman, D. H.; McDonnell, W.; Kehrl, H.; Ives, P. J.; Seal, E.; Chapman, R.;
37 Massaro, E. (1990) Pulmonary function, airway responsiveness, and respiratory
38 symptoms in asthmatics following exercise in NOz. Toxicol. Ind. Health 6: 155-171.
March 2008 6-40 DRAFT-DO NOT QUOTE OR CITE
-------
1 Rogge, W. R; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. (1993)
2 Sources of fine organic aerosol. 5. Natural gas home appliances. Environ. Sci. Technol.
3 27: 2736-2744.
4 Rojas-Bracho, L.; Suh, H. H.; Oyola, P.; Koutrakis, P. (2002) Measurements of children's
5 exposures to particles and nitrogen dioxide in Santiago, Chile. Sci. Total Environ. 287:
6 249-264.
7 Rojas-Martinez, R.; Perez-Padilla, R.; Olaiz-Fernandez, G.; Mendoza-Alvarado, L.; Moreno-
8 Macias, H.; Fortoul, T.; McDonnell, W.; Loomis, D.; Romieu, I. (2007a) Lung function
9 growth in children with long-term exposure to air pollutants in Mexico City. Am. J.
10 Respir. Crit. Care Med. 176: 377-384.
11 Rojas-Martinez, R.; Perez-Padilla, R.; Olaiz-Fernandez, G.; Mendoza-Alvarado, L.; Moreno-
12 Macias, H.; Fortoul, T.; McDonnell, W.; Loomis, D.; Romieu, I. (2007b) Lung function
13 growth in children with long-term exposure to air pollutants in Mexico City. Online data
14 supplement. Available: http://ajrccm.atsjournals.Org/cgi/data/176/4/377/DCl/l [3
15 October, 2007].
16 Rombout, P. J. A.; Dormans, J. A. M. A.; Marra, M.; Van Esch, G. J. (1986) Influence of
17 exposure regimen on nitrogen dioxide-induced morphological changes in the rat lung.
18 Environ. Res. 41:466-480.
19 Romieu, L; Ramirez-Aguilar, M.; Moreno-Macias, H.; Barraza-Villarreal, A.; Miller, P.;
20 Hernandez-Cadena, L.; Carbajal-Arroyo, L. A.; Hernandez-Avila, M. (2004) Infant
21 mortality and air pollution: modifying effect by social class. J. Occup. Environ. Med. 46:
22 1210-1216.
23 Romieu, L; Ramirez-Aguilar, M.; Sienra-Monge, J. J.; Moreno-Macias, H.; Del Rio-Navarro, B.
24 E.; David, G.; Marzec, J.; Hernandez-Avila, M.; London, S. (2006) GSTM1 and GSTP1
25 and respiratory health in asthmatic children exposed to ozone. Eur. Respir. J. 28: 953-
26 959.
27 Roorda-Knape, M. C.; Janssen, N. A. H.; De Hartog, J. J.; Van Vliet, P. H. N.; Harssema, H.;
28 Brunekreef, B. (1998) Air pollution from traffic in city districts near major motorways.
29 Atmos. Environ. 32: 1921-1930.
30 Roorda-Knape, M. C.; Janssen, N. A.; De Hartog, J.; Van Vliet, P. H.; Harssema, H.;
31 Brunekreef, B. (1999) Traffic related air pollution in city districts near motorways. Sci.
32 Total Environ. 235: 339-341.
33 Rossi, 0. V. J.; Kinnula, V. L.; Tienari, J.; Huhti, E. (1993) Association of severe asthma attacks
34 with weather, pollen, and air pollutants. Thorax 48: 244-248.
35 Rothman, K. J.; Greenland, S., eds. (1998) Modern epidemiology. 2nd ed. Philadelphia, PA:
36 Lippincott-Raven Publishers.
37 Rotko, T.; Kousa, A.; Aim, S.; Jantunen, M. (2001) Exposures to nitrogen dioxide inEXPOLIS-
38 Helsinki: microenvironment, behavioral and sociodemographic factors. J. Exposure Anal.
39 Environ. Epidemiol. 11: 216-223.
March 2008 6-41 DRAFT-DO NOT QUOTE OR CITE
-------
1 Roy-Burman, P.; Pattengale, P. K.; Sherwin, R. P. (1982) Effect of low levels of nitrogen
2 dioxide inhalation on endogenous retrovirus gene expression. Exp. Mol. Pathol. 36: 144-
3 155.
4 Rubenchik, B. L.; Glavin, A. A.; Galenko, P. M.; Kilkichko, A. A.; Olenick, I. 0.; Artemov, K.
5 V. (1995) Gaseous nitrogen dioxide increases the endogenous synthesis of carcinogenic
6 7V-nitrosodimethylamine in animals. J. Environ. Pathol. Toxicol. Oncol. 14: 111-115.
7 Rubenstein, I.; Bigby, B. G.; Reiss, T. F.; Boushey, H. A., Jr. (1990) Short-term exposure to 0.3
8 ppm nitrogen dioxide does not potentiate airway responsiveness to sulfur dioxide in
9 asthmatic subjects. Am. Rev. Respir. Dis. 141: 381-385.
10 Rubenstein, I.; Reiss, T. F.; Bigby, B. G.; Stites, D. P.; Boushey, H. A., Jr. (1991) Effects of 0.60
11 PPM nitrogen dioxide on circulating and bronchoalveolar lavage lymphocyte phenotypes
12 in healthy subjects. Environ. Res. 55: 18-30.
13 Ruckerl, R.; Ibald-Mulli, A.; Koenig, W. Schneider, A.; Woelke, G.; Cyrys, J.; Heinrich, J.;
14 Marder, V.; Frampton, M.; Wichmann, H. E.; Peters, A. (2006) Air pollution and markers
15 of inflammation and coagulation in patients with coronary heart disease. Environ. Health
16 Perspect. 173:432-441.
17 Rusznak, C.; Devalia, J. L.; Davies, R. J. (1996) Airway response of asthmatic subjects to
18 inhaled allergen after exposure to pollutants. Thorax 51: 1105-1108.
19 Sabin, L. D.; Kozawa, K.; Behrentz, E.; Winer, A. M.; Fitz, D. R.; Pankratz, D. V.; Colome, S.
20 D.; Fruin, S. A. (2005) Analysis of real-time variables affecting children's exposure to
21 diesel-related pollutants during school bus commutes in Los Angeles. Atmos. Environ.
22 39: 5243-5254.
23 Saez, M.; Ballester, F.; Barcelo, M. A.; Perez-Hoyos, S.; Bellido, J.; Tenias, J. M.; Ocana, R.;
24 Figueiras, A.; Arribas, F.; Aragones, N.; Tobias, A.; Cirera, L.; Canada, A.; on behalf of
25 the EMECAM Group. (2002) A combined analysis of the short-term effects of
26 photochemical air pollutants on mortality within the EMECAM project. Environ. Health
27 Perspect. 110:221-228.
28 Salam, M. T.; Millstein, J.; Li, Y.-R; Lurmann, F. W.; Margolis, H. G.; Gilliland, F. D. (2005)
29 Birth outcomes and prenatal exposure to ozone, carbon monoxide, and particulate matter:
30 results from the Children's Health Study. Environ. Health Perspect. 113: 1638-1644.
31 Salmeen, L; Durisin, A. M.; Prater, T. J.; Riley, T.; Schuetzle, D. (1982) Contribution of 1-
32 nitropyrene to direct-acting Ames assay mutagenicities of diesel particulate extracts.
33 Mutat. Res. 104: 17-23.
34 Samet, J. M.; Bell, M. L. (2004) Commentary: nitrogen dioxide and asthma redux. Int. J.
35 Epidemiol. 33: 215-216.
36 Samet, J. M.; Zeger, S. L.; Dominici, F.; Curriero, F.; Coursac, L; Dockery, D. W.; Schwartz, J.;
37 Zanobetti, A. (2000) The national morbidity, mortality, and air pollution study. Part II:
38 morbidity, mortality, and air pollution in the United States. Cambridge, MA: Health
39 Effects Institute; research report no. 94, part II.
40 Samoli, E.; Aga, E.; Touloumi, G.; Nisiotis, K.; Forsberg, B.; Lefranc, A.; Pekkanen, J.;
41 Wojtyniak, B.; Schindler, C.; Niciu, E.; Brunstein, R.; Dodic Fikfak, M.; Schwartz, J.;
March 2008 6-42 DRAFT-DO NOT QUOTE OR CITE
-------
1 Katsouyanni, K. (2006) Short-term effects of nitrogen dioxide on mortality: an analysis
2 within the APHEA project. Eur. Respir. J. 27: 1129-1137.
3 Sandstrom, T.; Andersson, M. C.; Kolmodin-Hedman, B.; Stjernberg, N.; Angstrom, T. (1990)
4 Bronchoalveolar mastocytosis and lymphocytosis after nitrogen dioxide exposure in man:
5 a time-kinetic study. Eur. Respir. J. 3: 138-143.
6 Sandstrom, T.; Stjernberg, N.; Eklund, A.; Ledin, M.-C.; Bjermer, L.; Kolmodin-Hedman, B.;
7 Lindstrom, K.; Rosenhall, L.; Angstrom, T. (1991) Inflammatory cell response in
8 bronchoalveolar lavage fluid after nitrogen dioxide exposure of healthy subjects: a dose-
9 response study. Eur. Respir. J. 4: 332-339.
10 Sandstrom, T.; Helleday, R.; Bjermer, L.; Stjernberg, N. (1992a) Effects of repeated exposure to
11 4 ppm nitrogen dioxide on bronchoalveolar lymphocyte subsets and macrophages in
12 healthy men. Eur. Respir. J. 5: 1092-1096.
13 Sandstrom, T.; Ledin, M.-C.; Thomasson, L.; Helleday, R.; Stjernberg, N. (1992b) Reductions in
14 lymphocyte subpopulations after repeated exposure to 1.5 ppm nitrogen dioxide. Br. J.
15 Ind. Med. 49: 850-854.
16 Sarnat, J. A.; Koutrakis, P.; Suh, H. H. (2000) Assessing the relationship between personal
17 particulate and gaseous exposures of senior citizens living in Baltimore, MD. J. Air
18 Waste Manage. Assoc. 50: 1184-1198.
19 Sarnat, J. A.; Schwartz, J.; Catalano, P. J.; Suh, H. H. (2001) Gaseous pollutants in particulate
20 matter epidemiology: confounders or surrogates? Environ. Health Perspect. 109: 1053-
21 1061.
22 Sarnat, J. A.; Brown, K. W.; Schwartz, J.; Coull, B. A.; Koutrakis, P. (2005) Ambient gas
23 concentrations and personal particulate matter exposures: implications for studying the
24 health effects of particles. Epidemiology 16: 385-395.
25 Sarnat, S. E.; Suh, H. H.; Coull, B. A.; Schwartz, J.; Stone, P. H.; Gold, D. R. (2006) Ambient
26 particulate air pollution and cardiac arrhythmia in a panel of older adults in Steubenville,
27 Ohio. Occup. Environ. Med. 63: 700-706.
28 Sarwar, G.; Corsi, R.; Allen, D.; Weschler, C. (2002a) Production and levels of selected indoor
29 radicals: a modeling assessment. In: Proceedings of 9th International Conference on
30 Indoor Air Quality and Climate, Indoor Air 2002; June-July; Monterey, CA.
31 Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002b) Hydroxyl radicals in
32 indoor environments. Atmos. Environ. 36: 3973-3988.
33 Sasaki, J. C.; Arey, J.; Eastmond, D. A.; Parks, K. K.; Grosovsky, A. J. (1997) Genotoxicity
34 induced in human lymphoblasts by atmospheric reaction products of naphthalene and
35 phenanthrene. Mutat. Res. 393: 23-35.
36 Saxon, A.; Diaz-Sanchez, D. (2005) Air pollution and allergy: you are what you breathe. Nat.
37 Immunol. 6: 223-226.
38 Scarlett, J. F.; Abbott, K. J.; Peacock, J. L.; Strachan, D. P.; Anderson, H. R. (1996) Acute
39 effects of summer air pollution on respiratory function in primary school children in
40 southern England. Thorax 51: 1109-1114.
March 2008 6-43 DRAFT-DO NOT QUOTE OR CITE
-------
1 Schildcrout, J. S.; Sheppard, L.; Lumley, T.; Slaughter, J. C.; Koenig, J. Q.; Shapiro, G. G.
2 (2006) Ambient air pollution and asthma exacerbations in children: an eight-city analysis.
3 Am. J. Epidemiol. 164: 505-517.
4 Schindler, C.; Ackermann-Liebrich, U.; Leuenberger, P.; Monn, C.; Rapp, R.; Bolognini, G.;
5 Bongard, J.-P.; Brandli, 0.; Domenighetti, G.; Karrer, W.; Keller, R.; Medici, T. G.;
6 Perruchoud, A. P.; Schoni, M. H.; Tschopp, J.-M.; Villiger, B.; Zellweger, J.-P.;
7 SAPALDIA Team. (1998) Associations between lung function and estimated average
8 exposure to N02 in eight areas of Switzerland. Epidemiology 9: 405-411.
9 Schindler, C.; Kiinzli, N.; Bongard, J.-P.; Leuenberger, P.; Karrer, W.; Rapp, R.; Monn, C.;
10 Ackermann-Liebrich, U.; Swiss Study on Air Pollution and Lung Diseases in Adults
11 Investigators. (2001) Short-term variation in air pollution and in average lung function
12 among never-smokers. Am. J. Respir. Crit. Care Med. 163: 356-361.
13 Schouten, J. P.; Vonk, J. M.; de Graaf, A. (1996) Short term effects of air pollution on
14 emergency hospital admissions for respiratory disease: results of the APHEA project in
15 two major cities in The Netherlands, 1977-89. In: St Leger, S., ed. The APHEA project.
16 Short term effects of air pollution on health: a European approach using epidemiological
17 time series data. J. Epidemiol. Community Health 50(suppl. 1): S22-S29.
18 Schuetzle, D. (1983) Sampling of vehicle emissions for chemical analysis and biological testing.
19 Environ. Health Perspect. 47: 65-80.
20 Schwartz, J. (2001) Air pollution and blood markers of cardiovascular risk. Environ. Health
21 Perspect. Suppl. 109: 405-409.
22 Schwartz, J.; Dockery, D. W.; Neas, L. M.; Wypij, D.; Ware, J. H.; Spengler, J. D.; Koutrakis,
23 P.; Speizer, F. E.; Ferris, B. G., Jr. (1994) Acute effects of summer air pollution on
24 respiratory symptom reporting in children. Am. J. Respir. Crit. Care Med. 150: 1234-
25 1242.
26 Schwartz, J.; Litonjua, A.; Suh, H.; Verrier, M.; Zanobetti, A.; Syring, M.; Nearing, B.; Verrier,
27 R.; Stone, P.; MacCallum, G.; Speizer, F. E.; Gold, D. R. (2005) Traffic related pollution
28 and heart rate variability in a panel of elderly subjects. Thorax 60: 455-461.
29 Seaton, A.; Dennekamp, M. (2003) Hypothesis: ill health associated with low concentrations of
30 nitrogen dioxide-an effect of ultrafine particles? Thorax 58: 1012-1015.
31 Segala, C.; Fauroux, B.; Just, J.; Pascual, L.; Grimfeld, A.; Neukirch, F. (1998) Short-term effect
32 of winter air pollution on respiratory health of asthmatic children in Paris. Eur. Respir. J.
33 11:677-685.
34 Segala, C.; Poizeau, D.; Neukirch, F.; Aubier, M.; Samson, J.; Gehanno, P. (2004) Air pollution,
35 passive smoking, and respiratory symptoms in adults. Arch. Environ. Health 59: 669-676.
36 Seinfeld, J. H.; Pandis, S. N. (1998) Atmospheric chemistry and physics: from air pollution to
37 climate change. New York, NY: John Wiley & Sons, Inc.
38 Selgrade, M. K.; Mole, M. L.; Miller, F. J.; Hatch, G. E.; Gardner, D. E.; Hu, P. C. (1981) Effect
39 of NOz inhalation and vitamin C deficiency on protein and lipid accumulation in the lung.
40 Environ. Res. 26: 422-437.
March 2008 6-44 DRAFT-DO NOT QUOTE OR CITE
-------
1 Shalamberidze, 0. P.; Tsereteli, N. T. (1971) Effect of low concentrations of sulfur and nitrogen
2 dioxides on the estrual cycle and reproductive functions of experimental animals. Hyg.
3 Sanit. (USSR) 36: 178-182.
4 Sherwin, R. P.; Carlson, D. A. (1973) Protein content of lung lavage fluid of guinea pigs exposed
5 to 0.4 ppm nitrogen dioxide: disc-gel electrophoresis for amount and types. Arch.
6 Environ. Health 27: 90-93.
7 Shindell, D. T.; Faluvegi, G.; Stevenson, D. S.; Krol, M. C.; Emmons, L. K.; Lamarque, J.-F.;
8 Petron, G.; Dentener, F. J.; Ellingsen, K.; Schultz, M. G.; Wild, 0.; Amann, M.;
9 Atherton, C. S.; Bergmann, D. J.; Bey, I.; Butler, T.; Cofala, J.; Collins, W. J.; Derwent,
10 R. G.; Doherty, R. M.; Drevet, J.; Eskes, H. J.; Fiore, A. M.; Gauss, M.; Hauglustaine, D.
11 A.; Horowitz, L. W.; Isaksen, I. S. A.; Lawrence, M. G.; Montanaro, V.; Miiller, J.-F.;
12 Pitari, G.; Prather, M. J.; Pyle, J. A.; Rast, S.; Rodriguez, J. M.; Sanderson, M. G.;
13 Savage, N. H.; Strahan, S. E.; Sudo, K.; Szopa, S.; Unger, N.; Van Noije, T. P. C.; Zeng,
14 G. (2006) Multimodel simulations of carbon monoxide: comparison with observations
15 and projected near-future changes. J. Geophys. Res. [Atmos.] 111 (D 19306):
16 10.1029/2006JD007100.
17 Shorter, J. H.; Herndon, S.; Zahniser, M. S.; Nelson, D. D.; Wormhoudt, J.; Demerjian, K. L.;
18 Kolb, C. E. (2005) Real-time measurements of nitrogen oxide emissions from in-use New
19 York City transit buses using a chase vehicle. Environ. Sci. Technol. 39: 7991-8000.
20 Shy, C. M.; Love, G. J. (1980) Recent evidence on the human health effects of nitrogen dioxide.
21 In: Lee, S. D., ed. Nitrogen oxides and their effects on health: [papers from the
22 symposium "Biological studies of environmental pollutants, part II. Health effects of
23 nitrogen oxides"]; April 1979; Honolulu, HI. Ann Arbor, MI: Ann Arbor Science
24 Publishers, Inc.; pp. 291-305.
25 Siegel, P. D.; Al-Humadi, N. H.; Nelson, E. R.; Lewis, D. M.; Hubbs, A. F. (1997) Adjuvant
26 effect of respiratory irritation on pulmonary allergic sensitization: time and site
27 dependency. Toxicol. Appl. Pharmacol. 144: 356-362.
28 Silkoff, P. E.; Zhang, L.; Dutton, S.; Langmack, E. L.; Vedal, S.; Murphy, J.; Make, B. (2005)
29 Winter air pollution and disease parameters in advanced chronic obstructive pulmonary
30 disease panels residing in Denver, Colorado. J. Allergy Clin. Immunol. 115: 337-344.
31 Simoni, M.; Carrozzi, L.; Baldacci, S.; Scognamiglio, A.; Di Pede, F.; Sapigni, T.; Viegi, G.
32 (2002) The Po River delta (north Italy) indoor epidemiological study: effects of pollutant
33 exposure on acute respiratory symptoms and respiratory function in adults. Arch.
34 Environ. Health 57: 130-136.
35 Simoni, M.; Scognamiglio, A.; Carrozzi, L.; Baldacci, S.; Angino, A.; Pistelli, F.; Di Pede, F.;
36 Viegi, G. (2004) Indoor exposures and acute respiratory effects in two general population
37 samples from a rural and an urban area in Italy. J. Exposure Anal. Environ. Epidemiol.
38 14(suppl. 1):S144-S152.
39 Simpson, R.; Denison, L.; Petroeschevsky, A.; Thalib, L.; Williams, G. (2000) Effects of
40 ambient particle pollution on daily mortality in Melbourne 1991-1996. J. Exposure Anal.
41 Environ. Epidemiol. 10: 488-496.
March 2008 6-45 DRAFT-DO NOT QUOTE OR CITE
-------
1 Simpson, R.; Williams, G.; Petroeschevsky, A.; Best, T.; Morgan, G.; Denison, L.; Hinwood, A.;
2 Neville, G. (2005a) The short-term effects of air pollution on hospital admissions in four
3 Australian cities. Aust. N. Z. J. Public Health 29: 213-221.
4 Simpson, R.; Williams, G.; Petroeschevsky, A.; Best, T.; Morgan, G.; Denison, L.; Hinwood, A.;
5 Neville, G.; Neller, A. (2005b) The short-term effects of air pollution on daily mortality
6 in four Australian cities. Aust. N. Z. J. Public Health 29: 205-212.
7 Singer, B. C.; Hodgson, A. T.; Hotchi, T.; Kim, J. J. (2004) Passive measurement of nitrogen
8 oxides to assess traffic-related pollutant exposure for the East Bay Children's Respiratory
9 Health Study. Atmos. Environ. 38: 393-403.
10 Singh, H. B.; Salas, L.; Herlth, D.; Kolyer, R.; Czech, E.; Avery, M.; Crawford, J. H.; Pierce, R.
11 B.; Sachse, G. W.; Blake, D. R.; Cohen, R. C.; Bertram, T. H.; Perring, A.; Wooldridge,
12 P. J.; Dibb, J.; Huey, G.; Hudman, R. C.; Turquety, S.; Emmons, L. K.; Flocke, F.; Tang,
13 Y.; Carmichael, G. R.; Horowitz, L. W. (2007) Reactive nitrogen distribution and
14 partitioning in the North American troposphere and lowermost stratosphere. J. Geophys.
15 Res. [Atmos.] 112(D12S04): 10.1029/2006JD007664.
16 Slordal, L. H.; Walker, S. E.; Solberg, S. (2003) The urban air dispersion model EPISODE
17 applied in AirQUIS2003. Technical description. Kjeller, Norway: Norwegian Institute for
18 Air Research (NILU); report no. TR 12/2003.
19 Smith, B. J.; Nitschke, M.; Pilotto, L. S.; Ruffin, R. E.; Pisaniello, D. L.; Wilson, K. J. (2000)
20 Health effects of daily indoor nitrogen dioxide exposure in people with asthma. Eur.
21 Respir.J. 16:879-885.
22 Solomon, C.; Christian, D. L.; Chen, L. L.; Welch, B. S.; Kleinman, M. T.; Dunham, E.; Erie, D.
23 J.; Balmes, J. R. (2000) Effect of serial-day exposure to nitrogen dioxide on airway and
24 blood leukocytes and lymphocyte subsets. Eur. Respir. J. 15: 922-928.
25 Son, B.; Yang, W.; Breysse, P.; Chung, T.; Lee, Y. (2004) Estimation of occupational and
26 nonoccupational nitrogen dioxide exposure for Korean taxi drivers using a
27 microenvironmental model. Environ. Res. 94: 291-296.
28 S0rensen, M.; Loft, S.; Andersen, H. V.; Raaschou-Nielsen, 0.; Skovgaard, L. T.; Knudsen, L.
29 E.; Nielsen, I. V.; Hertel, 0. (2005) Personal exposure to PM2.5, black smoke and N02 in
30 Copenhagen: relationship to bedroom and outdoor concentrations covering seasonal
31 variation. J. Exposure Anal. Environ. Epidemiol. 15: 413-422.
32 Spengler, J. D.; Brauer, M.; Samet, J. M.; Lambert, W. E. (1993) Nitrous acid in Albuquerque,
33 New Mexico, homes. Environ. Sci. Technol. 27: 841-845.
34 Spengler, J.; Schwab, M.; Ryan, P. B.; Colome, S.; Wilson, A. L.; Billick, L; Becker, E. (1994)
35 Personal exposure to nitrogen dioxide in the Los Angeles Basin. J. Air Waste Manage.
36 Assoc. 44: 39-47.
37 Spicer, C. W.; Coutant, R. W.; Ward, G. F.; Joseph, D. W.; Gaynor, A. J.; Billick, I. H. (1989)
38 Rates and mechanisms of N02 removal from indoor air by residential materials. Environ.
39 Int. 15: 643-654.
40 Spicer, C. W.; Kenny, D. V.; Ward, G. F.; Billick, I. H. (1993) Transformations, lifetimes, and
41 sources of N02, HONO, and HN03 in indoor environments. Air Waste 43: 1479-1485.
March 2008 6-46 DRAFT-DO NOT QUOTE OR CITE
-------
1 Spix, C.; Anderson, H. R.; Schwartz, J.; Vigotti, M. A.; LeTertre, A.; Vonk, J. M.; Touloumi, G.;
2 Balducci, F.; Piekarski, T.; Bacharova, L.; Tobias, A.; Ponka, A.; Katsouyanni, K. (1998)
3 Short-term effects of air pollution on hospital admissions of respiratory diseases in
4 Europe: a quantitative summary of APHEA study results. Arch. Environ. Health 53: 54-
5 64.
6 Steerenberg, P. A.; Nierkens, S.; Fischer, P. H.; Van Loveren, H.; Opperhuizen, A.; Vos, J. G.;
7 Van Amsterdam, J. G. (2001) Traffic-related air pollution affects peak expiratory flow,
8 exhaled nitric oxide, and inflammatory nasal markers. Arch. Environ. Health 56: 167-
9 174.
10 Steerenberg, P. A.; Bischoff, E. W. M. A.; de Klerk, A.; Verlaan, A. P. J.; Jongbloets, L. M. N.;
11 Van Loveren, H.; Opperhuizen, A.; Zomer, G.; Heisterkamp, S. H.; Hady, M.; Spieksma,
12 F. T. M.; Fischer, P. H.; Dormans, J. A. M. A.; van Amsterdam, J. G. C. (2003) Acute
13 effect of air pollution on respiratory complaints, exhaled NO and biomarkers in nasal
14 lavages of allergic children during the pollen season. Int. Arch. Allergy Immunol. 131:
15 127-137.
16 Steinbacher, M.; Zellweger, C.; Schwarzenbach, B.; Bugmann, S.; Buchmann, B.; Ordonez, C.;
17 Prevot, A. S. H.; and Hueglin, C. (2007) Nitrogen oxides measurements at rural sites in
18 Switzerland: bias of conventional measurement techniques. J. Geophys. Res. (Atmos.)
19 112(011307): 10.1029/2006JD007971.
20 Steinvil, A.; Kordova-Biezuner, L.; Shapira, I.; Berliner, S.; Rogowski, 0. (2007) Short-term
21 exposure to air pollution and inflammation-sensitive biomarkers. Environ. Res. 106: 51-
22 61.
23 Stemmler, K.; Ammann, M.; Bonders, C.; Kleffmann, J.; George, C. (2006) Photosensitized
24 reduction of nitrogen dioxide on humic acid as a source of nitrous acid. Nature (London,
25 U.K.) 440: 195-198.
26 Stephens, R. J.; Freeman, G.; Crane, S. C.; Furiosi, N. J. (1971) Ultrastructural changes in the
27 terminal bronchiole of the rat during continuous, low-level exposure to nitrogen dioxide.
28 Exp. Mol. Pathol. 14: 1-19.
29 Stephens, R. J.; Freeman, G.; Evans, M. J. (1972) Early response of lungs to low levels of
30 nitrogen dioxide: light and electron microscopy. Arch. Environ. Health 24: 160-179.
31 Stevenson, D.; Dentener, F. J.; Schultz, M. G.; Ellingsen, K.; Van Noije, T. P. C.; Wild, 0.;
32 Zeng, G.; Amann, M.; Atherton, C. S.; Bell, N.; Bergmann, D. J.; Bey, L; Butler, T.;
33 Cofala, J.; Collins, W. J.; Derwent, R. G.; Doherty, R. M.; Drevet, J.; Eskes, H. J.; Fiore,
34 A. M.; Gauss, M.; Hauglustaine, D. A.; Horowitz, L. W.; Isaksen, I. S. A.; Krol, M. C.;
35 Lamarque, J.-F.; Lawrence, M. G.; Montanaro, V.; Miiller, J.-F.; Pitari, G.; Prather, M. J.;
36 Pyle, J. A.; Rast, S.; Rodriguez, J. M.; Sanderson, M. G. (2006) Multimodel ensemble
37 simulations of present-day and near-future tropospheric ozone. J. Geophys. Res. [Atmos.]
38 111(008301): 10.1029/2005JD006338.
39 Stieb, D. M.; Burnett, R. T.; Beveridge, R. C.; Brook, J. R. (1996) Association between ozone
40 and asthma emergency department visits in Saint John, New Brunswick, Canada.
41 Environ. Health Perspect. 104: 1354-1360.
March 2008 6-47 DRAFT-DO NOT QUOTE OR CITE
-------
1 Stieb, D. M.; Beveridge, R. C.; Brook, J. R.; Smith-Doiron, M.; Burnett, R. T.; Dales, R. E.;
2 Beaulieu, S.; Judek, S.; Mamedov, A. (2000) Air pollution, aeroallergens and
3 cardiorespiratory emergency department visits in Saint John, Canada. J. Exposure Anal.
4 Environ. Epidemiol. 10: 461-477.
5 Stieb, D. M.; Judek, S.; Burnett, R. T. (2002) Meta-analysis of time-series studies of air pollution
6 and mortality: effects of gases and particles and the influence of cause of death, age, and
7 season. J. Air Waste Manage. Assoc. 52: 470-484.
8 Stieb, D. M.; Judek, S.; Burnett, R. T. (2003) Meta-analysis of time-series studies of air pollution
9 and mortality: update in relation to the use of generalized additive models. J. Air Waste
10 Manage. 53: 258-261.
11 Stolzel, M.; Peters, A.; Wichmann, H.-E. (2003) Daily mortality and fine and ultrafine particles
12 in Erfurt, Germany. In: Revised analyses of time-series studies of air pollution and health.
13 Special report. Boston, MA: Health Effects Institute; pp. 231-240. Available:
14 http://www.healtheffects.org/Pubs/TimeSeries.pdf [18 October, 2004].
15 Strand, V.; Salomonsson, P.; Lundahl, J.; Bylin, G. (1996) Immediate and delayed effects of
16 nitrogen dioxide exposure at an ambient level on bronchial responsiveness to histamine in
17 subjects with asthma. Eur. Respir. J. 9: 733-740.
18 Strand, V.; Rak, S.; Svartengren, M.; Bylin, G. (1997) Nitrogen dioxide exposure enhances
19 asthmatic reaction to inhaled allergen in subjects with asthma. Am. J. Respir. Crit. Care
20 Med. 155:881-887.
21 Strand, V.; Svartengren, M.; Rak, S.; Barck, C.; Bylin, G. (1998) Repeated exposure to an
22 ambient level of N02 enhances asthmatic response to nonsymptomatic allergen dose. Eur.
23 Respir. J. 12:6-12.
24 Stutz, J.; Alicke, B.; Ackermann, R.; Geyer, A.; Wang, S.; White, A. B.; Williams, E. J.; Spicer,
25 C. W.; Fast, J. D. (2004) Relative humidity dependence of HONO chemistry in urban
26 areas. J. Geophys. Res. [Atmos.] 109: 10.1029/2003JD004135.
27 Sun, H.-L.; Chou, M.-C.; Lue, K.-H. (2006) The relationship of air pollution to ED visits for
28 asthma differ between children and adults. Am. J. Emerg. Med. 24: 709-713.
29 Sunyer, J.; Spix, C.; Quenel, P.; Ponce-de-Leon, A.; Ponka, A.; Barumandzadeh, T.; Touloumi,
30 G.; Bacharova, L.; Wojtyniak, B.; Vonk, J.; Bisanti, L.; Schwartz, J.; Katsouyanni, K.
31 (1997) Urban air pollution and emergency admissions for asthma in four European cities:
32 the APHEA project. Thorax 52: 760-765.
33 Suzuki, A. K.; Tsubone, H.; Ichinose, T.; Oda, H.; Kubota, K. (1981) [Effects of subchronic
34 nitrogen dioxide exposure on arterial blood pHa, Pacoz and Paoz in rats]. Nippon
35 Eiseigaku Zasshi 36: 816-823.
36 Suzuki, T.; Terada, N.; Ikeda, S.; Ohsawa, M.; Endo, K.; Mizoguchi, I. (1984) [Effect of N02
37 exposure on the activity of angiotensin converting enzyme in lung]. Kenkyu Nenpo
38 Tokyo-toritsu Eisei Kenkyusho 35: 279-285.
39 Svartengren, M.; Strand, V.; Bylin, G.; Jarup, L.; Pershagen, G. (2000) Short-term exposure to
40 air pollution in a road tunnel enhances the asthmatic response to allergen. Eur. Resp. J.
41 15:716-724.
March 2008 6-48 DRAFT-DO NOT QUOTE OR CITE
-------
1 Tabacova, S.; Balabaeva, L. (1988) Nitrogen dioxide embryotoxicity and lipid peroxidation.
2 Teratology 38: 29A.
3 Tabacova, S.; Nikiforov, B.; Balabaeva, L. (1985) Postnatal effects of maternal exposure to
4 nitrogen dioxide. Neurobehav. Toxicol. Teratol. 7: 785-789.
5 Tager, I. B.; Balmes, J.; Lurmann, F.; Ngo, L.; Alcorn, S.; Kiinzli, N. (2005) Chronic exposure to
6 ambient ozone and lung function in young adults. Epidemiology 16: 751-759.
7 Takano, H.; Yanagisawa, R.; Inoue, K.-L; Shimada, A.; Ichinose, T.; Sadakane, K.; Yoshino, S.;
8 Yamaki, K.; Morita, M.; Yoshikawa, T. (2004) Nitrogen dioxide air pollution near
9 ambient levels is an atherogenic risk primarily in obese subjects: a brief communication.
10 Exp. Biol. Med. 229: 361-364.
11 Tanaka, H.; Honma, S.; Nishi, M.; Igarashi, T.; Teramoto, S.; Nishio, F.; Abe, S. (1998) Acid
12 fog and hospital visits for asthma: an epidemiological study. Eur. Respir. J. 11: 1301-
13 1306.
14 Tenias, J. M.; Ballester, F.; Rivera, M. L. (1998) Association between hospital emergency visits
15 for asthma and air pollution in Valencia, Spain. Occup. Environ. Med. 55: 541-547.
16 Tenias, J. M.; Ballester, F.; Perez-Hoyos, S.; Rivera, M. L. (2002) Air pollution and hospital
17 emergency room admissions for chronic obstructive pulmonary disease in Valencia,
18 Spain. Arch. Environ. Health 57: 41-47.
19 Tepper, J. S.; Costa, D. L.; Winsett, D. W.; Stevens, M. A.; Doerfler, D. L.; Watkinson, W. P.
20 (1993) Near-lifetime exposure of the rat to a simulated urban profile of nitrogen dioxide:
21 pulmonary function evaluation. Fund. Appl. Toxicol. 20: 88-96.
22 Thompson, A. J.; Shields, M. D.; Patterson, C. C. (2001) Acute asthma exacerbations and air
23 pollutants in children living in Belfast, Northern Ireland. Arch. Environ. Health 56: 234-
24 241.
25 Timonen, K. L.; Pekkanen, J. (1997) Air pollution and respiratory health among children with
26 asthmatic or cough symptoms. Am. J. Respir. Crit. Care Med. 156: 546-552.
27 Timonen, K. L.; Pekkanen, J.; Tiittanen, P.; Salonen, R. 0. (2002) Effects of air pollution on
28 changes in lung function induced by exercise in children with chronic respiratory
29 symptoms. Occup. Environ. Med. 59: 129-134.
30 Timonen, K. L.; Hoek, G.; Heinrich, J.; Bernard, A.; Brunekreef, B.; De Hartog, J.; Hameri, K.;
31 Ibald-Mulli, A.; Mirme, A.; Peters, A.; Tiittanen, P.; Kreyling, W. G.; Pekkanen, J.
32 (2004) Daily variation in fine and ultrafine particulate air pollution and urinary
33 concentrations of lung Clara cell protein CC16. Occup. Environ. Med. 61: 908-914.
34 Tobias, A.; Campbell, M. J.; Saez, M. (1999) Modelling asthma epidemics on the relationship
35 between air pollution and asthma emergency visits in Barcelona, Spain. Eur. J.
36 Epidemiol. 15: 799-803.
37 Tokiwa, H.; Ohnishi, Y. (1986) Mutagenicity and carcinogenicity of nitroarenes and their
38 sources in the environment. Crit. Rev. Toxicol. 17: 23-60.
39 Tokiwa, H.; Nakanishi, Y.; Sera, N.; Hara, N.; Inuzuka, S. (1998) Analysis of environmental
40 carcinogens associated with the incidence of lung cancer. Toxicol. Lett. 99: 33-41.
March 2008 6-49 DRAFT-DO NOT QUOTE OR CITE
-------
1 Tolbert, P. E.; Mulholland, J. A.; Macintosh, D. L.; Xu, F.; Daniels, D.; Devine, 0. J.; Carlin, B.
2 P.; Klein, M.; Dorley, J.; Butler, A. J.; Nordenberg, D. F.; Frumkin, H.; Ryan, P. B.;
3 White, M. C. (2000) Air quality and pediatric emergency room visits for asthma in
4 Atlanta, Georgia. Am. J. Epidemiol. 151: 798-810.
5 Tolbert, P. E.; Klein, M.; Peel, J. L.; Sarnat, S. E.; Sarnat, J. A. (2007) Multipollutant modeling
6 issues in a study of ambient air quality and emergency department visits in Atlanta. J.
7 Exposure Sci. Environ. Epidemiol. 17(Suppl. 2s): S29-S35.
8 Triche, E. W.; Belanger, K.; Bracken, M. B.; Beckett, W. S.; Holford, T. R.; Gent, J. F.;
9 McSharry, J.-E.; Leaderer, B. P. (2005) Indoor heating sources and respiratory symptoms
10 in nonsmoking women. Epidemiology 16: 377-384.
11 Tsai, S.-S.; Goggins, W. B.; Chiu, H.-F.; Yang, C.-Y. (2003) Evidence for an association
12 between air pollution and daily stroke admissions in Kaohsiung, Taiwan. Stroke 34:
13 2612-2616.
14 Tsai, S.-S.; Cheng, M.-H.; Chiu, H.-F.; Wu, T.-N.; Yang, C.-Y. (2006) Air pollution and hospital
15 admissions for asthma in a tropical city: Kaohsiung, Taiwan. Inhalation Toxicol. 18: 549-
16 554.
17 Tsubone, H.; Suzuki, A. K. (1984) Reflex cardiopulmonary responses by stimulation to type J
18 receptors in rats exposed to N02. J. Toxicol. Environ. Health 13: 905-917.
19 Tunnicliffe, W. S.; Burge, P. S.; Ayres, J. G. (1994) Effect of domestic concentrations of
20 nitrogen dioxide on airway responses to inhaled allergen in asthmatic patients. Lancet
21 344: 1733-1736.
22 U.S. Census Bureau. (2006) American Housing Survey for the United States: 2005. Washington,
23 DC: U.S. Department of Housing and Urban Development. (Current Housing Reports,
24 Series H150/05). Available: http://www.census.gov/prod/2006pubs/hl50-05.pdf (24
25 January, 2008).
26 U.S. Code. (2003a) Clean Air Act, §108, air quality criteria and control techniques. U. S. C. 42:
27 §7408.
28 U.S. Code. (2003b) Clean Air Act, §109, national ambient air quality standards. U. S. C. 42:
29 §7409.
30 U.S. Code. (2005) Clean Air Act, §302, definitions. U. S. C. 42: §7602(h).
31 U.S. Court of Appeals for the District of Columbia. (1980) Lead Industries v. U.S.
32 Environmental Protection Agency. 647 F2d 1130, 1154 (DC Cir. 1980).
33 U.S. Court of Appeals for the District of Columbia. (1981) American Petroleum Institute v.
34 Costle. 665 F2d 1176, 1186 (DC Cir. 1981).
35 U.S. Environmental Protection Agency. (1982) Air quality criteria for oxides of nitrogen.
36 Research Triangle Park, NC: Office of Health and Environmental Assessment,
37 Environmental Criteria and Assessment Office; EPA report no. EPA-600/8-82-026.
38 Available from: NTIS, Springfield, VA; PB83-131011.
39 U.S. Environmental Protection Agency. (1993) Air quality criteria for oxides of nitrogen.
40 Research Triangle Park, NC: Office of Health and Environmental Assessment,
March 2008 6-50 DRAFT-DO NOT QUOTE OR CITE
-------
1 Environmental Criteria and Assessment Office; report nos. EPA/600/8-9!/049aF-cF. 3v.
2 Available from: NTIS, Springfield, VA; PB95-124533, PB95-124525, and PB95-124517.
3 U.S. Environmental Protection Agency. (1995) Review of the national ambient air quality
4 standards for nitrogen dioxide: assessment of scientific and technical information.
5 Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no.
6 EPA/452/R-95-005.
7 U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related
8 photochemical oxidants. Research Triangle Park, NC: Office of Research and
9 Development; report nos. EPA/600/AP-93/004aF-cF. 3v. Available from: NTIS,
10 Springfield, VA; PB96-185582, PB96-185590, and PB96-185608. Available:
11 http://cfpub2.epa.gov/ncea/.
12 U.S. Environmental Protection Agency. (1998) SLAMS/NAMS/PAMS network review
13 guidance. Research Triangle Park, NC: Office of Air Quality Planning and Standards;
14 EPA-454/R-98-003. Available:
15 http://www.epa.gov/ttn/amtic/files/ambient/criteria/reldocs/netrev98.pdf (22 February,
16 2008).
17 U.S. Environmental Protection Agency. (2004) Air quality criteria for particulate matter.
18 Research Triangle Park, NC: National Center for Environmental Assessment; report no.
19 EPA/600/P-99/002aF-bF. 2v. Available: http://cfpub.epa.gov/ncea/ [9 November, 2004].
20 U.S. Environmental Protection Agency. (2005) Guidelines for carcinogen risk assessment.
21 Washington, DC: Risk Assessment Forum; report no. EPA/630/P-03/001F. Available:
22 http://cfpub.epa.gov/ncea/index.cfm [30 November, 2005].
23 U.S. Environmental Protection Agency. (2006) Air quality criteria for ozone and related
24 photochemical oxidants. Research Triangle Park, NC: National Center for Environmental
25 Assessment; report no. EPA/600/R-05/004aF-cF. 3v. Available:
26 http://cfpub.epa.gov/ncea/ [24 March, 2006].
27 U.S. Environmental Protection Agency. (2007) Integrated plan for the primary National Ambient
28 Air Quality Standard for Nitrogen Dioxide. Research Triangle Park, NC: National Center
29 for Environmental Assessment.
30 U.S. Senate. (1970) National Air Quality Standards Act of 1970: report of the Committee on
31 Public Works, United States Senate together with individual views to accompany S.
32 4358. Washington, DC: Committee on Public Works; report no. CONG/91-1196.
33 U.S. Supreme Court. (2001) Whitman v. American Trucking Association. 531 U.S. 457 (nos. 99-
34 1257 and 99-1426).
35 Vagaggini, B.; Paggiaro, P. L.; Giannini, D.; Franco, A. D.; Cianchetti, S.; Carnevali, S.;
36 Taccola, M.; Bacci, E.; Bancalari, L.; Dente, F. L.; Giuntini, C. (1996) Effect of short-
37 term N02 exposure on induced sputum in normal, asthmatic and COPD subjects. Eur.
38 Respir.J. 9: 1852-1857.
39 Van Der Zee, S. C.; Hoek, G.; Boezen, H. M.; Schouten, J. P.; Van Wijnen, J. H.; Brunekreef, B.
40 (1999) Acute effects of urban air pollution on respiratory health of children with and
41 without chronic respiratory symptoms. Occup. Environ. Med. 56: 802-813.
March 2008 6-51 DRAFT-DO NOT QUOTE OR CITE
-------
1 Van Der Zee, S. C.; Hoek, G.; Boezen, M. H.; Schouten, J. P.; Van Wijnen, J. H.; Brunekreef, B.
2 (2000) Acute effects of air pollution on respiratory health of 50-70 yr old adults. Eur.
3 Respir.J. 15:700-709.
4 Van Noije, T. P. C.; Eskes, H. J.; Dentener, F. J.; Stevenson, D. S.; Ellingsen, K.; Schultz, M. G.;
5 Wild, 0.; Amann, M.; Atherton, C. S.; Bergmann, D. J.; Bey, I.; Boersma, K. F.; Butler,
6 T.; Cofala, J.; Drevet, J.; Fiore, A. M.; Gauss, M.; Hauglustaine, D. A.; Horowitz, L. W.;
7 Isaksen, I. S. A.; Krol, M. C.; Lamarque, J.-F.; Lawrence, M. G.; Martin, R. V.;
8 Montanaro, V.; Miiller, J.-F.; Pitari, G.; Prather, M. J.; Pyle, J. A.; Richter, A.;
9 Rodriguez, J. M.; Savage, N. H.; Strahan, S. E.; Sudo, K.; Szopa, S.; Van Roozendael, M.
10 (2006) Multi-model ensemble simulations of tropospheric N02 compared with GOME
11 retrievals for the year 2000. Atmos. Chem. Phys. Discuss. 6: 2965-3047.
12 Van Strien, R. T.; Gent, J. F.; Belanger, K.; Triche, E.; Bracken, M. B.; Leaderer, B. P. (2004)
13 Exposure to N02 and nitrous acid and respiratory symptoms in the first year of life.
14 Epidemiology 15:471-478.
15 Varshney, C. K.; Singh, A. P. (2003) Passive samplers for NOX monitoring: a critical review.
16 Environmentalist 23: 127-136.
17 Velsor, L. W.; Postlethwait, E. M. (1997) N02-induced generation of extracellular reactive
18 oxygen is mediated by epithelial lining layer antioxidants. Am. J. Physiol. 17: L1265-
19 L1275.
20 Victorin, K.; Busk, L.; Cederberg, H.; Magnusson, J. (1990) Genotoxic activity of 1,3-butadiene
21 and nitrogen dioxide and their photochemical reaction products in Drosophila and in the
22 mouse bone marrow micronucleus assay. Mutat. Res. 228: 203-209.
23 Vigotti, M. A.; Chiaverini, F.; Biagiola, P.; Rossi, G. (2007) Urban air pollution and emergency
24 visits for respiratory complaints in Pisa, Italy. J. Toxicol. Environ. Health Part A 70: 266-
25 269.
26 Villeneuve, P. J.; Chen, L.; Stieb, D.; Rowe, B. H. (2006) Associations between outdoor air
27 pollution and emergency department visits for stroke in Edmonton, Canada. Eur. J.
28 Epidemiol. 21: 689-700.
29 Vinzents, P. S.; M011er, P.; S0rensen, M.; Knudsen, L. E.; Herte, L. Q.; Jensen, F. P.; Schibye,
30 B.; Loft, S. (2005) Personal exposure to ultrafine particles and oxidative DNA damage.
31 Environ. Health Perspect. 113: 1485-1490.
32 Von Klot, S.; Wolke, G.; Tuch, T.; Heinrich, J.; Dockery, D. W.; Schwartz, J.; Kreyling, W. G.;
33 Wichmann, H. E.; Peters, A. (2002) Increased asthma medication use in association with
34 ambient fine and ultrafine particles. Eur. Respir. J. 20: 691-702.
35 Von Klot, S.; Peters, A.; Aalto, P.; Bellander, T.; Berglind, N.; D'Ippoliti, D.; Elosua, R.;
36 Hermann, A.; Kulmala, M.; Lanki, T.; Lowel, H.; Pekkanen, J.; Picciotto, S.; Sunyer, J.;
37 Forastiere, F.; Health Effects of Particles on Susceptible Subpopulations (HEAPSS)
38 Study Group. (2005) Ambient air pollution is associated with increased risk of hospital
39 cardiac readmissions of myocardial infarction survivors in five European cities.
40 Circulation 112: 3073-3079.
41 Wagner, W. D.; Duncan, B. R.; Wright, P. G.; Stokinger, H. E. (1965) Experimental study of
42 threshold limit of N02. Arch. Environ. Health 10: 455-466.
March 2008 6-52 DRAFT-DO NOT QUOTE OR CITE
-------
1 Wainman, T.; Weschler, C.; Lioy, P.; Zhang, J. (2001) Effects of surface type and relative
2 humidity on the production and concentration of nitrous acid in a model indoor
3 environment. Environ. Sci. Technol. 35: 2200-2206.
4 Wallace, L. A.; Emmerich, S. J.; Howard-Reed, C. (2004) Source strengths of ultrafine and fine
5 particles due to cooking with a gas stove. Environ. Sci. Technol. 38: 2304-2311.
6 Wang, X.-K.; Lu, W.-Z. (2006) Seasonal variation of air pollution index: Hong Kong case study.
7 Chemosphere 63: 1261-1272.
8 Wang, J. H.; Devalia, J. L.; Duddle, J. M.; Hamilton, S. A.; Davies, R. J. (1995a) Effect of six-
9 hour exposure to nitrogen dioxide on early-phase nasal response to allergen challenge in
10 patients with a history of seasonal allergic rhinitis. J. Allergy Clin. Immunol. 96: 669-
11 676.
12 Wang, J. H.; Duddle, J.; Devalia, J. L.; Davies, R. J. (1995b) Nitrogen dioxide increases
13 eosinophil activation in the early-phase response to nasal allergen provocation. Int. Arch.
14 Allergy Immunol. 107: 103-105.
15 Wang, J. H.; Devalia, J. L.; Rusznak, C.; Bagnall, A.; Sapsford, R. J.; Davies, R. J. (1999) Effect
16 of fluticasone propionate aqueous nasal spray on allergen-induced inflammatory changes
17 in the nasal airways of allergic rhinitics following exposure to nitrogen dioxide. Clin.
18 Exp. Allergy 29: 234-240.
19 Ward, D. J.; Miller, M. R.; Walters, S.; Harrison, R. M.; Ayres, J. G. (2000) Impact of correcting
20 peak flow for nonlinear errors on air pollutant effect estimates from a panel study. Eur.
21 Respir.J. 15: 137-140.
22 Ward, D. J.; Roberts, K. T.; Jones, N.; Harrison, R. M.; Ayres, J. G.; Hussain, S.; Walters, S.
23 (2002) Effects of daily variation in outdoor particulates and ambient acid species in
24 normal and asthmatic children. Thorax 57: 489-502.
25 Wayne, R. P.; Barnes, L; Biggs, P.; Burrows, J. P.; Canosa-Mas, C. E.; Hjorth, J.; Le Bras, G.;
26 Moortgat, G. K.; Perner, D.; Poulet, G.; Restelli, G.; Sidebottom, H. (1991) The nitrate
27 radical: physics, chemistry, and the atmosphere. Atmos. Environ. Part A 25: 1-203.
28 Weichenthal, S.; Dufresne, A.; Infante-Rivard, C.; Joseph, L. (2007) Indoor ultrafine particle
29 exposures and home heating systems: a cross-sectional survey of Canadian homes during
30 the winter months. J. Exposure Sci. Environ. Epidemiol. 17: 288-297.
31 Wellenius, G. A.; Schwartz, J.; Mittleman, M. A. (2005a) Air pollution and hospital admissions
32 for ischemic and hemorrhagic stroke among medicare beneficiaries. Stroke 36: 2549-
33 2553.
34 Wellenius, G. A.; Bateson, T. F.; Mittleman, M. A.; Schwartz, J. (2005b) Particulate air pollution
35 and the rate of hospitalization for congestive heart failure among medicare beneficiaries
36 in Pittsburgh, Pennsylvania. Am. J. Epidemiol. 161: 1030-1036.
37 Weschler, C. J.; Shields, H. C. (1996) The conversion (reduction) of nitrogen dioxide to nitric
38 oxide as a consequence of charcoal filtration. In: Yoshizawa, S.; Kimura, K.; Ikeda, K.;
39 Tanabe, S.; Iwata, T., eds. Indoor Air '96: proceedings of the 7th international conference
40 on indoor air quality and climate, v. 3, July; Nagoya, Japan. Toykyo, Japan: Indoor Air
41 '96; pp. 453-458.
March 2008 6-53 DRAFT-DO NOT QUOTE OR CITE
-------
1 Weschler, C. J.; Shields, H. C. (1997) Potential reactions among indoor pollutants. Atmos.
2 Environ. 31:3487-3495.
3 Weschler, C. J.; Hodgson, A. T.; Wooley, J. D. (1992) Indoor chemistry: ozone, volatile organic
4 compounds, and carpets. Environ. Sci. Technol. 26: 2371-2377.
5 Weschler, C. J.; Shields, H. C.; Naik, D. V. (1994) Indoor chemistry involving 03, NO, and N02
6 as evidenced by 14 months of measurements at a site in southern California. Environ. Sci.
7 Technol. 28: 2120-2132.
8 Westerdahl, D.; Fruin, S.; Sax, T.; Fine, P. M.; Sioutas, C. (2005) Mobile platform
9 measurements of ultrafine particles and associated pollutant concentrations on freeways
10 and residential streets in Los Angeles. Atmos. Environ. 39: 3597-3610.
11 Wheeler, A.; Zanobetti, A.; Gold, D. R.; Schwartz, J.; Stone, P.; Suh, H. H. (2006) The
12 relationship between ambient air pollution and heart rate variability differs for individuals
13 with heart and pulmonary disease. Environ. Health Perspect. 114: 560-566.
14 Wichmann, H.-E.; Spix, C.; Tuch, T.; Wolke, G.; Peters, A.; Heinrich, J.; Kreyling, W. G.;
15 Heyder, J. (2000) Daily mortality and fine and ultrafine particles in Erfurt, Germany. Part
16 I: role of particle number and particle mass. Cambridge, MA: Health Effects Institute;
17 research report no. 98.
18 Williams, E. J.; Parrish, D. D.; Fehsenfeld, F. C. (1987) Determination of nitrogen oxide
19 emissions from soils: results from a grassland site in Colorado, United States. J. Geophys.
20 Res. [Atmos.] 92: 2173-2179.
21 Witschi, H. (1988) Ozone, nitrogen dioxide and lung cancer: a review of some recent issues and
22 problems. Toxicology 48: 1-20.
23 Witten, A.; Solomon, C.; Abbritti, E.; Arjomandi, M.; Zhai, W.; Kleinman, M.; Balmes, J.
24 (2005) Effects of nitrogen dioxide on allergic airway responses in subjects with asthma.
25 J. Occup. Environ. Med. 47: 1250-1259.
26 Wolff, G. T. (1993) On a N0x-focused control strategy to reduce 03. J. Air Waste Manage.
27 Assoc. 43: 1593-1596.
28 Wolff, G. T. (1996) Closure by the Clean Air Scientific Advisory Committee (CASAC) on the
29 draft Air Quality Criteria for Particulate Matter [letter to Carol M. Browner,
30 Administrator, U.S. EPA]. Washington, DC: U.S. Environmental Protection Agency,
31 Clean Air Scientific Advisory Committee.; EPA-SAB-CASAC-LTR-96-005; March 15.
32 Wong, T. W.; Lau, T. S.; Yu, T. S.; Neller, A.; Wong, S. L.; Tarn, W.; Pang, S. W. (1999) Air
33 pollution and hospital admissions for respiratory and cardiovascular diseases in Hong
34 Kong. Occup. Environ. Med. 56: 679-683.
35 Wong, G. W.; Ko, F. W.; Lau, T. S.; Li, S. T.; Hui, D.; Pang, S. W.; Leung, R.; Fok, T. F.; Lai,
36 C. K. (2001) Temporal relationship between air pollution and hospital admissions for
37 asthmatic children in Hong Kong. Clin. Exp. Allergy 31: 565-569.
38 Wong, C.-M.; Atkinson, R. W.; Anderson, H. R.; Hedley, A. J.; Ma, S.; Chau, P. Y.-K.; Lam, T.-
39 H. (2002) A tale of two cities: effects of air pollution on hospital admissions in Hong
40 Kong and London compared. Environ. Health Perspect. 110: 67-77.
March 2008 6-54 DRAFT-DO NOT QUOTE OR CITE
-------
1 Woodruff, T. J.; Parker, J. D.; Kyle, A. D.; Schoendorf, K. C. (2003) Disparities in exposure to
2 air pollution during pregnancy. Environ. Health Perspect. Ill: 942-946.
3 Woodwell, D. A.; Cherry, D. K. (2004) National Ambulatory Medical Care Survey: 2002
4 summary. Hyattsville, MD: National Center for Health Statistics; DHHS publication no.
5 (PHS) 2004-1250. (Advance data from vital and health statistics; no. 346). Available:
6 http://www.cdc.gov/nchs/data/ad/ad346.pdf [3 August, 2005].
7 Wright, R. J.; Subramanian, S. V. (2007) Advancing a multilevel framework for epidemiologic
8 research on asthma disparities. Chest 132: 757-769.
9 Wu, Y -C.; Batterman S. (2006) Proximity of schools in Detroit, Michigan to automobile and
10 truck traffic. J. Exposure Sci. Environ. Epidemiol. 10: 1038.
11 Yamanaka, S. (1984) Decay rates of nitrogen oxides in a typical Japanese living room. Environ.
12 Sci. Technol. 18: 566-570.
13 Yanagisawa, Y.; Nishimura, H. (1982) A badge-type personal sampler for measurement of
14 personal exposure to N02 and NO in ambient air. Environ. Int. 8: 235-242.
15 Yang, C. Y.; Chen, C. J. (2007) Air pollution and hospital admissions for chronic obstructive
16 pulmonary disease in a subtropical city: Taipei, Taiwan. J. Toxicol. Environ. Health A.
17 70: 1214-1219.
18 Yang, Q.; Chen, Y.; Shi, Y.; Burnett, R. T.; McGrail, K. M.; Krewski, D. (2003) Association
19 between ozone and respiratory admissions among children and the elderly in Vancouver,
20 Canada. Inhalation Toxicol. 15: 1297-1308.
21 Yang, C.-Y.; Chen, Y.-S.; Yang, C.-H.; Ho, S.-C. (2004a) Relationship between ambient air
22 pollution and hospital admissions for cardiovascular diseases in Kaohsiung, Taiwan. J.
23 Toxicol. Environ. Health Part A 67: 483-493.
24 Yang, W.; Lee, K.; Chung, M. (2004b) Characterization of indoor air quality using multiple
25 measurements of nitrogen dioxide. Indoor Air 14: 105-111.
26 Yang, Q.; Chen, Y.; Krewski, D.; Burnett, R. T.; Shi, Y.; McGrail, K. M. (2005) Effect of short-
27 term exposure to low levels of gaseous pollutants on chronic obstructive pulmonary
28 disease hospitalizations. Environ. Res. 99: 99-105.
29 Yang, C. Y.; Chen, C. C.; Chen, C. Y.; Kuo, H. W. (2007) Air pollution and hospital admissions
30 for asthma in a subtropical city: Taipei, Taiwan. J. Toxicol. Environ. Health A. 70: 111-
31 117.
32 Zanobetti, A.; Schwartz, J. (2006) Air pollution and emergency admissions in Boston, MA. J.
33 Epidemiol. Community Health 60: 890-895.
34 Zielinska, B.; Arey, J.; Atkinson, R.; Winer, A. M. (1989) The nitroarenes of molecular weight
35 247 in ambient particulate samples collected in southern California. Atmos. Environ. 23:
36 223-229.
37 Zielinska, B.; Sagebiel, J.; McDonald, J. D.; Whitney, K.; Lawson, D. R. (2004) Emission rates
38 and comparative chemical composition from selected in-use diesel and gasoline-fueled
39 vehicles. J. Air Waste Manage. Assoc. 54: 1138-1150.
March 2008 6-55 DRAFT-DO NOT QUOTE OR CITE
-------
1 Zipprich, J. L.; Harris, S. A.; Fox, J. C.; Borzelleca, J. F. (2002) An analysis of factors that
2 influence personal exposure to nitrogen oxides in residents of Richmond, Virginia. J.
3 Exposure Anal. Environ. Epidemiol. 12: 273-285.
4 Zota, A.; Adamkiewicz, G.; Levy, J. I.; Spengler, J. D. (2005) Ventilation in public housing:
5 implications for indoor nitrogen dioxide concentrations. Indoor Air 15: 393-401.
March 2008 6-56 DRAFT-DO NOT QUOTE OR CITE
-------
4>EPA
Please make all necessary changes in the below label, PRESORTED STANDARD
detach copy or copy, and return to the address in the upper POSTAGE & FEES PAID
left-hand corner. EPA
United States DCDMIT M^ r* Q^
Environmental Protection _ PERMIT Na Q-35
Agency If you do not wish to receive these reports CHECK HERE LJ;
detach copy or copy, and return to the address in the upper
left-hand corner.
National Center for
Environmental Assessment
Research Triangle Park, NC 27711
Official Business
Penalty for Private Use
$300
EPA/600/R-07/093aB
March 2008
------- |