LJ
I
Integrated Review Plan for the
Primary National Ambient Air Quality
Standards for Nitrogen Dioxide
External Review Draft
-------
[This page intentionally left blank]
-------
EPA-452/P-14-001
February 2014
Integrated Review Plan for the
Primary National Ambient Air Quality
Standards for Nitrogen Dioxide
External Review Draft
U. S. Environmental Protection Agency
National Center for Environmental Assessment
Office of Research and Development
and
Office of Air Quality Planning and Standards
Office of Air and Radiation
Research Triangle Park, North Carolina
February 2014
-------
DISCLAIMER
This draft integrated review plan serves as a public information document and as a
management tool for the U. S. Environmental Protection Agency's National Center for
Environmental Assessment and Office of Air Quality Planning and Standards in conducting the
review of the national ambient air quality standards for nitrogen dioxide. The approach described
in this draft plan may be modified for presentation in the final plan to reflect review by the Clean
Air Scientific Advisory Committee and public comments. Subsequent modifications to the plan
may result from information developed during this review, and in consideration of advice and
comments received from the Clean Air Scientific Advisory Committee and the public during the
course of the review. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
February 2014 1 Draft - Do Not Quote or Cite
-------
TABLE OF CONTENTS
Table of Contents i
List of Figures ii
List of Tables ii
LIST OF ACRONYMS/ABBREVIATIONS iii
1 INTRODUCTION 1-1
1.1 LEGISLATIVE REQUIREMENTS 1-1
1.2 OVERVIEW OF THE NAAQS REVIEW PROCESS 1-3
1.3 REVIEW OF AIR QUALITY CRITERIA FOR OXIDES OF NITROGEN AND
STANDARDS FOR NITROGEN DIOXIDE 1-7
1.4 SCOPE OF THE CURRENT REVIEW 1-11
2 STATUS AND SCHEDULE 2-1
3 KEY POLICY-RELEVANT IS SUES 3-1
3.1 Considerations and Conclusions IN last REVIEW 3-1
3.1.1 Need for Revision 3-2
3.1.2 Elements of Revised Standard 3-5
3.1.2.1 Indicator 3-5
3.1.2.2 Averaging Time 3-6
3.1.2.3 Levels 3-7
3.1.2.4 Form 3-9
3.1.3 Areas of Uncertainty 3-10
3.2 GENERAL APPROACH FOR THE CURRENT Review 3-11
4 SCIENCE ASSESSMENT 4-1
4.1 SCOPE OF THE ISA 4-1
4.2 ORGANIZATION OF THE ISA 4-2
4.3 ASSESSMENT APPROACH 4-3
4.3.1 Introduction 4-3
4.3.2 Literature Search and Selection of Relevant Studies 4-5
4.3.3 Evaluation of Individual Study Quality 4-6
4.3.4 Integration of Evidence and Determination of Causality 4-9
4.3.5 Quality Management 4-10
4.4 SPECIFIC ISSUES TO BE ADDRESSED IN THE ISA 4-10
4.5 SCIENTIFIC AND PUBLIC REVIEW of the ISA 4-20
5 QUANTITATIVE RISK AND EXPOSURE ASSESSMENT 5-1
5.1 OVERVIEW OF RISK AND EXPOSURE ASSESSMENT FROM PRIOR
REVIEW 5-2
5.1.1 Ambient Air Quality Characterization 5-2
5.1.1.1 Key Observations 5-4
5.1.1.2 Key Uncertainties 5-6
February 2014 i Draft - Do Not Quote or Cite
-------
5.1.2 Human Exposure Assessment 5-6
5.1.2.1 Key Observations 5-8
5.1.2.2 Key Uncertainties 5-8
5.1.3 Epidemiological-based Human Health Risk Assessment 5-9
5.1.3.1 Key Observations 5-10
5.1.3.2 Key Uncertainties 5-10
5.2 CONSIDERATION OF QUANTITATIVE ASSESSMENTS FOR THIS REVIEW
5-11
5.2.1 Air Quality Characterization 5-11
5.2.2 Human Exposure Assessment 5-11
5.2.3 Controlled Exposure-based Human Health Risk Assessment 5-17
5.2.4 Epidemiological-based Human Health Risk Assessment 5-17
5.3 SCIENTIFIC AND PUBLIC REVIEW 5-19
6 AMBIENT AIR MONITORING 6-1
6.1 CONSIDERATION OF SAMPLING AND ANALYSIS METHODS 6-1
6.2 CONSIDERATION OF AIR MONITORING NETWORK REQUIREMENTS .. 6-2
7 POLICY ASSESSMENT AND RULEMAKING 7-1
7.1 POLICY ASSESSMENT 7-1
7.2 RULEMAKING 7-2
8 REFERENCES 8-1
APPENDICES
Appendix A: Draft Outline for Integrated Science Assessment for Oxides of Nitrogen
- Health Criteria A-l
LIST OF FIGURES
Figure 1-1. Overview of theNAAQS Review Process 1-4
Figure 3-1. Overview of General Approach for Review of Primary NO2 Standards 3-13
Figure 4-1. General Process for Development of Integrated Science Assessments 4-4
Figure 6-1. NO2 Monitoring Network: Active Monitors as of September 2013 6-4
LIST OF TABLES
Table 1-1. Primary national ambient air quality standards for oxides of nitrogen since 1971.... 1-8
Table 2-1. Anticipated schedule for the current review 2-2
Table 5-1. Information (data, methods, models, etc.) identified as potentially important
and/or newly available to inform the air quality characterization for the current
review 5-13
Table 5-2. Information (data, methods, models, etc.) identified as potentially important
and/or newly available to inform the exposure assessment for the current
review 5-16
Table 5-3. Information (data, methods, models, etc.) identified as potentially important
and/or newly available to inform the epidemiological-based risk assessment
for the current review 5-18
February 2014 ii Draft - Do Not Quote or Cite
-------
AADT
AMMS
ANPR
APEX
AQCD
AQI
AQS
CAA
CASAC
CBS A
CFR
CO
COPD
C-R
CV
CVD
ED
EPA
FEM
FEVi
FR
FRM
GAM
HA
HERO
HONO
HRV
HNO3
IRP
ISA
IUGR
Km
LC
LML
m
MSA
NAAQS
NCEA
NCore
NO
NO2
NO3"
NOX
NOY
LIST OF ACRONYMS/ABBREVIATIONS
Annual Average Daily Traffic
Air Monitoring and Methods Subcommittee
Advanced notice of proposed rulemaking
Air Pollutants Exposure model
Air Quality Criteria Document
Air Quality Index
EPA's Air Quality System
Clean Air Act
Clean Air Scientific Advisory Committee
Core Based Statistical Area
Code of Federal Regulations
Carbon monoxide
Chronic obstructive pulmonary disease
Concentration-response
Cardiovascular
Cardiovascular disease
Emergency department
Environmental Protection Agency
Federal Equivalent Method
Forced expiratory volume in one second, volume of air exhaled in first
second of exhalation
Federal Register
Federal Reference Method
Generalized Additive Models
Hospital admissions
Health and Environmental Research Online
Nitrous acid
Heart rate variability
Nitric acid
Integrated Review Plan
Integrated Science Assessment
Intrauterine growth restriction, intrauterine growth retardation
Kilometer
Local conditions
Lowest measured level
Meters
Metropolitan Statistical Area
National Ambient Air Quality Standards
National Center for Environmental Assessment
National Core Monitoring Network
Nitric oxide
Nitrogen dioxide
Nitrate
NO+NO2
Total oxides of nitrogen (NOx + NOz)
February 2014
in
Draft - Do Not Quote or Cite
-------
NOz
03
OAQPS
OAR
OMB
ORD
PA
PAN
PM
ppb
ppm
PRB
QA
QMP
REA
RIA
RTF
SES
SLAMS
SO2
TBD
Reactive oxides of nitrogen (e.g., HNOs, HONO, PAN, particulate
nitrates)
Ozone
Office of Air Quality Planning and Standards
Office of Air and Radiation
Office of Management and Budget
Office of Research and Development
Policy Assessment
Peroxyacetyl nitrate
Particulate matter
Parts per billion
Parts per million
Policy-relevant background
Quality assurance
Quality Management Plan
Risk and Exposure Assessment
Regulatory Impact Analysis
Research Triangle Park
Socioeconomic status
State and local air monitoring stations
Sulfur dioxide
To be determined
February 2014
IV
Draft - Do Not Quote or Cite
-------
i 1 INTRODUCTION
2 The U.S. Environmental Protection Agency (EPA) is conducting a review of the primary
3 (health-based) national ambient air quality standards (NAAQS) for nitrogen dioxide (NC>2). This
4 draft Integrated Review Plan (IRP) presents the planned approach for the review. This review
5 will provide an integrative assessment of relevant scientific information for oxides of nitrogen
1 9
6 and will focus on the basic elements that define the NAAQS: the indicator, averaging time,
7 form,3 and level.4 The EPA Administrator will consider these elements collectively in evaluating
8 the protection to public health afforded by the primary standards.
9 This document is organized into eight chapters. Chapter 1 summarizes the legislative
10 requirements for the review of the NAAQS (section 1.1), summarizes the review process (section
11 1.2), provides an overview of past reviews of the primary NC>2 NAAQS (section 1.3), and
12 outlines the scope of the current review (section 1.4) Chapter 2 presents the status and schedule
13 for the current review. Chapter 3 provides background on the key issues and uncertainties that
14 informed the final decisions in the last review and presents a set of policy-relevant questions that
15 will serve to focus this review on the critical scientific and policy issues. Chapters 4 through 7
16 discuss the planned scope and organization of key assessment documents, the planned
17 approaches for preparing the documents, plans for scientific and public review of the documents,
18 and specific ambient air quality monitoring considerations. Complete reference citations are
19 provided in chapter 8.
20 1.1 LEGISLATIVE REQUIREMENTS
21 Two sections of the Clean Air Act (CAA) govern the establishment, review, and revision
22 of the NAAQS. Section 108 (42 U.S.C. 7408) directs the Administrator to identify and list
23 certain air pollutants and then to issue air quality criteria for those pollutants. The Administrator
24 is to list those air pollutants that in her "judgment, cause or contribute to air pollution which may
25 reasonably be anticipated to endanger public health or welfare;" "the presence of which in the
26 ambient air results from numerous or diverse mobile or stationary sources;" and "for which . . .
27 [the Administrator] plans to issue air quality criteria..." Air quality criteria are intended to
1 The "indicator" of a standard defines the chemical species or mixture that is measured in determining
whether an area attains the standard. Nitrogen dioxide (NO2) is the indicator for the oxides of nitrogen.
2 The "averaging time" defines the time period over which ambient measurements are averaged (e.g., 1-
hour, 8-hour, 24-hour, annual).
3 The "form" of a standard defines the air quality statistic that is compared to the level of the standard in
determining whether an area attains the standard. For example, the form of the current 1-hour NO2 standard is the
three-year average of the 98th percentile of the annual distribution of 1-hour daily maximum NO2 concentrations.
4 The "level" defines the allowable concentration of the criteria pollutant in the ambient air.
February 2014 1-1 Draft - Do Not Quote or Cite
-------
1 "accurately reflect the latest scientific knowledge useful in indicating the kind and extent of all
2 identifiable effects on public health or welfare which may be expected from the presence of [a]
3 pollutant in the ambient air . . ." 42 U.S.C. 7408(b). Section 109 (42 U.S.C. 7409) directs the
4 Administrator to propose and promulgate "primary" and "secondary" NAAQS for pollutants for
5 which air quality criteria are issued.5 Section 109(b)(l) defines a primary standard as one "the
6 attainment and maintenance of which in the judgment of the Administrator, based on such
7 criteria and allowing an adequate margin of safety, are requisite to protect the public health."6 A
8 secondary standard, as defined in section 109(b)(2), must "specify a level of air quality the
9 attainment and maintenance of which, in the judgment of the Administrator, based on such
10 criteria, is requisite to protect the public welfare from any known or anticipated adverse effects
11 associated with the presence of [the] pollutant in the ambient air."7
12 The requirement that primary standards provide an adequate margin of safety was
13 intended to address uncertainties associated with inconclusive scientific and technical
14 information available at the time of standard setting. It was also intended to provide a reasonable
15 degree of protection against hazards that research has not yet identified. See Lead Industries
16 Association v. EPA, 647 F.2d 1130, 1154 (D.C. Cir 1980); American Petroleum Institute v.
17 Costle, 665 F.2d 1176, 1186 (D.C. Cir. 1981); American Farm Bureau Federation v. EPA, 559
18 F. 3d 512, 533 (D.C. Cir. 2009); Association of Battery Recyclers v. EPA, 604 F. 3d 613, 617-18
19 (D.C. Cir. 2010). Both kinds of uncertainties are components of the risk associated with pollution
20 at levels below those at which human health effects can be said to occur with reasonable
21 scientific certainty. Thus, in selecting primary standards that provide an adequate margin of
22 safety, the Administrator is seeking not only to prevent pollution levels that have been
23 demonstrated to be harmful but also to prevent lower pollutant levels that may pose an
24 unacceptable risk of harm, even if the risk is not precisely identified as to nature or degree. The
25 CAA does not require the Administrator to establish a primary NAAQS at a zero-risk level or at
26 background concentration levels, see Lead Industries v. EPA, 647 F.2d at 1156 n.5l,_Mississippi
27 v. EPA, 723 F. 3d 246, 255, 262-63 (D.C. Cir. 2013), but rather at a level that reduces risk
28 sufficiently so as to protect public health with an adequate margin of safety.
5 As discussed in section 1.4 below, this document describes the review of the primary NO2 standards. The
secondary NO2 standard will be separately reviewed in conjunction with review of the secondary SO2 standard.
6The 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." S. Rep. No. 91-1196, 91st Cong., 2d Sess. 10 (1970).
7 -
Welfare effects as defined in section 302(h) (42 U.S.C. 7602(h)) 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."
February 2014 1-2 Draft - Do Not Quote or Cite
-------
1 In addressing the requirement for an adequate margin of safety, the EPA considers such
2 factors as the nature and severity of the health effects involved, the size of the sensitive group(s),
3 and the kind and degree of uncertainties. The selection of any particular approach to providing an
4 adequate margin of safety is a policy choice left specifically to the Administrator's judgment.
5 See Lead Industries Association v. EPA., 647 F.2d at 1161-62; Mississippi v. EPA., 723 F. 3d at
6 265.
7 In setting standards that are "requisite" to protect public health and welfare, as provided
8 in section 109(b), the EPA's task is to establish standards that are neither more nor less stringent
9 than necessary for these purposes. In so doing, the EPA may not consider the costs of
10 implementing the standards. See generally, Whitman v. American Trucking Associations, 531
11 U.S. 457, 465-472, 475-76 (2001). Likewise, "[attainability and technological feasibility are not
12 relevant considerations in the promulgation of national ambient air quality standards." American
13 Petroleum Institute v. Costle, 665 F. 2d at 1185.
14 Section 109(d)(l) requires that "not later than December 31, 1980, and at 5-year intervals
15 thereafter, the Administrator shall complete a thorough review of the criteria published under
16 section 108 and the national ambient air quality standards . . . and shall make such revisions in
17 such criteria and standards and promulgate such new standards as may be appropriate ..."
18 Section 109(d)(2) requires that an independent scientific review committee "shall complete a
19 review of the criteria . . . and the national primary and secondary ambient air quality standards. . .
20 and shall recommend to the Administrator any new . . . standards and revisions of existing
21 criteria and standards as may be appropriate . . . ." Since the early 1980s, this independent
o
22 review function has been performed by the Clean Air Scientific Advisory Committee (CASAC).
23 1.2 OVERVIEW OF THE NAAQS REVIEW PROCESS
24 The current process for reviewing the NAAQS includes four major phases: (1) planning,
25 (2) science assessment, (3) risk/exposure assessment, and (4) policy assessment and rulemaking.
26 Figure 1-1 provides an overview of this process, and each phase is described in more detail
27 below.9
Lists of CASAC members and of members of the CASAC Oxides of Nitrogen Primary NAAQS Review
Panel are available at:
http://vosemite.epa.gov/sab/sabproduct.nsfAVebCASAC/CommitteesandMembership7OpenDocument.
9The EPA maintains a website on which key documents developed for NAAQS reviews are made available
(http://www.epa.gOv/ttn/naaqs/X The EPA's NAAQS review process has evolved over time (Jackson, 2009).
Information on the current process is available at: http://www.epa.gov/ttn/naaqs/review.html. As discussed in
section 1.3 below, this process was generally followed in the primary NO2 NAAQS review completed in 2010 with
the exception that there was not a separate Policy Assessment document issued. Rather, the Risk and Exposure
Assessment (U.S. EPA, 2008b) included a policy assessment chapter (i.e., Chapter 10).
February 2014 1-3 Draft - Do Not Quote or Cite
-------
Workshop nn ^ In
science-policy issues
Peer-reviewed
^^ EPA ^H
^T proposed ^H
tegrated Review Plan (
Dclcy-relevant issues an
RP): timeline and key
Integrated Scie
synthesi
nee Assessment (ISA): evaluation and
3 of most policy-relevant studies
i
Risk/Exposure Assessment (REA):
on key results, observations, and uncertainties
1
Policy Assessment (PA): staff analysis of
interpretation of information in the ISA and REA
Interagency
^L decisions on ^^H
^^ otandarHc ^| ^^^^
Agency decision
„ making and draft
proposal notice
Clean Air Scientific
Advisory Committee
(CASAC) review
Public comment
r
Public hearings
and comments
on proposal
Agency decision
making and draft
final notice
Interagency
review
Figure 1-1. Overview of the NAAQS review process.
February 2014
1-4
Draft - Do Not Quote or Cite
-------
1 The planning phase of the NAAQS review process begins with a science policy
2 workshop, which is intended to identify issues and questions to frame the review. Drawing from
3 the workshop discussions, a draft IRP is prepared j ointly by EPA's National Center for
4 Environmental Assessment (NCEA), within the Office of Research and Development (ORD),
5 and EPA's Office of Air Quality Planning and Standards (OAQPS), within the Office of Air and
6 Radiation (OAR).10 The draft IRP is made available for CAS AC review and for public comment.
7 The final IRP is prepared in consideration of CASAC and public comments. This document
8 presents the current plan and specifies the schedule for the entire review, the process for
9 conducting the review, and the key policy-relevant science issues that will guide the review.
10 The second phase of the review, science assessment, involves the preparation of an
11 Integrated Science Assessment (ISA) and supplementary materials. The ISA, prepared by
12 NCEA, provides a concise review, synthesis, and evaluation of the most policy-relevant science,
13 including key science judgments that are important to the design and scope of exposure and risk
14 assessments, as well as other aspects of the NAAQS review. The ISA (and any supplementary
15 materials that may be developed) provides a comprehensive assessment of the current scientific
16 literature pertaining to known and anticipated effects on public health and welfare associated
17 with the presence of the pollutant in the ambient air, emphasizing information that has become
18 available since the last air quality criteria review in order to reflect the current state of
19 knowledge. As such, the ISA forms the scientific foundation for each NAAQS review and is
20 intended to provide information useful in forming judgments about air quality indicator(s),
21 form(s), averaging time(s) and level(s) for the NAAQS. The current review process generally
22 includes production of a first and second draft ISA, both of which undergo CASAC and public
23 review prior to completion of the final ISA. Chapter 4 below provides a more detailed
24 description of the planned scope, organization, and assessment approach for the ISA and any
25 supporting materials that may be developed.
26 In the third phase, the risk/exposure assessment phase, OAQPS staff considers
27 information and conclusions presented in the ISA, with regard to support provided for the
28 development of quantitative assessments of the risks and/or exposures for health and/or welfare
29 effects. As an initial step, staff prepares a planning document (REA Planning Document) that
30 considers the extent to which newly available scientific evidence and tools/methodologies
31 warrant the conduct of quantitative risk and exposure assessments. As discussed in Chapter 5
10In this review of the primary NAAQS for NO2, a draft plan for development of the ISA was prepared by
NCEA prior to development of this draft IRP. The draft plan for development of the ISA was made available for
public comment and was the subject of a consultation with CASAC (78 FR 26026; 78 FR 27234). Comments
received during that consultation have been considered in preparation of chapter 4 in this draft IRP. Further
comments received on this draft IRP will be considered in developing a final IRP and a second draft ISA.
February 2014 1-5 Draft - Do Not Quote or Cite
-------
1 below, the REA Planning Document focuses on the degree to which important uncertainties in
2 the last review may be addressed by new information available in this review. Specifically, the
3 document considers the extent to which newly available data, methods, and tools might be
4 expected to appreciably affect the assessment results, or address important gaps in our
5 understanding of the exposures and risks associated with NO2. To the extent warranted, this
6 document outlines a general plan, including scope and methods, for conducting assessments. The
7 REA Planning Document is generally prepared in conjunction with the first draft ISA11 and is
8 presented for consultation with CASAC and for public comment. When an assessment is
9 performed, one or more drafts of each risk and exposure assessment document (REA) undergoes
10 CASAC and public review. The REA provides concise presentations of methods, key results,
11 observations, and related uncertainties. Chapter 5 below discusses consideration of potential
12 quantitative human health-related assessments for this review.
13 The review process ends with the policy assessment and rulemaking phase. The Policy
14 Assessment (PA) is prepared prior to issuance of proposed and final rules. The PA provides a
15 transparent presentation of OAQPS staff analyses and conclusions regarding the adequacy of the
16 current standards and, if revision is considered, what revisions may be appropriate. The PA
17 integrates and interprets the information from the ISA and REA to frame policy options for
18 consideration by the Administrator. Such an evaluation of policy implications is intended to help
19 "bridge the gap" between the Agency's scientific assessments, presented in the ISA and
20 REA(s), and the judgments required of the EPA Administrator in determining whether it is
21 appropriate to retain or revise the NAAQS. In so doing, the PA is also intended to facilitate
22 CASAC's advice to the Agency and recommendations to the Administrator on the adequacy of
23 the existing standards and, as pertinent, on revisions that may be appropriate to consider, as
24 provided for in the CAA. In evaluating the adequacy of the current standards and, as appropriate,
25 a range of potential alternative standards, the PA considers the available scientific evidence and,
26 as available, quantitative risk and exposure analyses together with related limitations and
27 uncertainties. The PA focuses on the information that is most pertinent to evaluating the basic
28 elements of national ambient air quality standards: indicator, averaging time, form, and level.
29 One or more drafts of a PA are released for CASAC review and public comment prior to
30 completion of the final PA.
31 Following issuance of the final PA and consideration of conclusions presented therein,
32 the Agency develops and publishes a notice of proposed rulemaking that communicates the
33 Administrator's proposed decisions regarding the standards review. A draft notice undergoes
nThe current review of the primary NO2 standards is an exception to this. As indicated in Table 2-1 below,
the draft REA planning document will be made available for public comment and consultation with CASAC
subsequent to the CASAC review of the first draft ISA.
February 2014 1-6 Draft - Do Not Quote or Cite
-------
1 9
1 interagency review involving other federal agencies prior to publication. Materials upon which
2 the proposed decision is based, including the documents described above, are made available to
3 the public in the regulatory docket for the review.13 A public comment period, during which
4 public hearings are generally held, follows publication of the notice of proposed rulemaking.
5 Taking into account comments received on the proposed rule,14 the Agency develops a final rule
6 which undergoes interagency review prior to publication to complete the rulemaking process.
7 Chapter 7 below discusses the development of the PA and the rulemaking steps for this review.
8 1.3 REVIEW OF AIR QUALITY CRITERIA FOR OXIDES OF
9 NITROGEN AND STANDARDS FOR NITROGEN DIOXIDE
10 In 1971, the EPA added nitrogen oxides to the list of criteria pollutants under section
11 108(a)(l) of the CAA and issued the initial air quality criteria (36 FR 1515, January 30, 1971;
12 U.S. EPA, 1971). Based on these air quality criteria, the EPA promulgated NAAQS for nitrogen
13 oxides using NC>2 as the indicator (36 FR 8186, April 30, 1971). Both primary and secondary
14 standards were set at 100 |ig/m3 (equal to 0.053 parts per million (ppm)), annual average. Since
15 then, the Agency has completed multiple reviews of the air quality criteria and primary
16 standards, as summarized in Table 1-1.
17
12 Where implementation of the proposed decision would have an annual effect on the economy of $100
million or more (e.g., by necessitating the implementation of emissions controls), the EPA develops and releases a
draft regulatory impact analysis (RIA) concurrent with the notice of proposed rulemaking. This activity is conducted
under Executive Order 12866. The RIA is conducted completely independent of the rulemaking process and, by
statute, is not considered in decisions regarding the review of the NAAQS.
13 All documents in the docket are listed in the www.regulations.gov index. Publically available docket
materials are available either electronically at www.regulations.gov or in hard copy at the Air and Radiation Docket
and Information Center. The docket ID number for this review is EPA-HQ-OAR-2013-0146.
14When issuing the final rulemaking, the Agency responds to all significant comments on the proposed rule.
February 2014 1-7 Draft - Do Not Quote or Cite
-------
1
2
3
Table 1-1. Primary national ambient air quality standards for oxides of nitrogen since
1971.
Final
Rule/Decision
1971
36 FR 8186
Apr 30, 1971
1985
50 FR 25532
Jun 19, 1985
1996
61 FR 52852
Oct8, 1996
2010
75 FR 6474
Feb 9, 2010
Indicator
NO2
Averaging
Time
1 year
Level
53 ppb15
Form
Annual arithmetic average
Primary NO2 standard retained, without revision.
Primary NO2 standard retained, without revision.
NO2
1 hour
lOOppb
3 -year average of the 98th
percentile of the annual
distribution of daily maximum
1-hour concentrations
Primary annual NO2 standard retained, without revision.
4
5
6
7
8
9
10
11
12
The EPA retained the primary and secondary NO2 standards, without revision, in reviews
completed in 1985 and 1996 (50 FR 25532, June 19, 1985; 61 FR 52852, October 8, 1996). In
the latter of the two decisions, the EPA concluded that "the existing annual primary standard
appears to be both adequate and necessary to protect human health against both long- and short-
term NO2 exposures" and that "retaining the existing annual standard is consistent with the
scientific data assessed in the Criteria Document (U.S. EPA, 1993) and the Staff Paper (U.S.
EPA, 1995) and with the advice and recommendations of CASAC" (61 FR 52854, October 8,
1996).16
15 The initial standard level of the annual NO2 standard was 100 ug/m3 which is equal to 0.053 ppm or 53
parts per billion (ppb). The units for the standard level were officially changed to ppb in the final rule issued in 2010
(75 FR 6531, February 9, 2010).
16 In presenting the rationale for the final decision, the EPA noted that "a 0.053 ppm annual standard would
keep annual NO2 concentrations considerably below the long-term levels for which serious chronic effects have
been observed in animals" and that "[r]etaining the existing standard would also provide protection against short-
term peak NO2 concentrations at the levels associated with mild changes in pulmonary function and airway
responsiveness observed in controlled human [exposure] studies" (60 FR 52874, 52880, October 11, 1995).
February 2014
1-8
Draft - Do Not Quote or Cite
-------
1 The last review of the air quality criteria for oxides of nitrogen (health criteria) and the
2 primary NO2 standard was initiated in December 2005 (70 FR 73236, December 9, 2005).17'18
3 The Agency's plans for conducting the review were presented in the Integrated Review Plan for
4 the Primary National Ambient Air Quality Standard for Nitrogen Dioxide (2007 IRP, U.S. EPA,
5 2007a), which included consideration of comments received during a CAS AC consultation as
6 well as public comment on a draft IRP. The scientific assessment for the review was described in
7 the 2008 Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (2008 ISA,
8 U.S. EPA, 2008a), multiple drafts of which received review by CAS AC and the public. The EPA
9 also conducted quantitative human risk and exposure assessments, after consultation with
10 CASAC and after receiving public comment on a draft analysis plan (U.S. EPA, 2007b). These
11 technical analyses were presented in the Risk and Exposure Assessment to Support the Review of
12 the NO2 Primary National Ambient Air Quality Standard (2008 REA, U. S. EPA, 2008b),
13 multiple drafts of which received CASAC and public review.
14 Over the course of the last review, the EPA made several changes to the NAAQS review
15 process. An important change was the discontinuation of the Staff Paper, which traditionally
16 contained staff evaluations to bridge the gap between the Agency's science assessments and the
17 judgments required of the EPA Administrator in determining whether it was appropriate to retain
18 or revise the NAAQS.19 In the course of reviewing the second draft REA, however, CASAC
19 expressed the view that the document would be incomplete without the addition of a policy
20 assessment chapter presenting an integration of evidence-based considerations and risk and
21 exposures assessment results. CASAC stated that such a chapter would be "critical for
22 considering options for the NAAQS for NO2 (Samet, 2008a). In addition, within the period of
23 CASAC's review of the second draft REA, the EPA's Deputy Administrator indicated in a letter
24 to the CASAC chair, addressing earlier CASAC comments on the NAAQS review process, that
25 the risk and exposure assessment would include "a broader discussion of the science and how
26 uncertainties may effect decisions on the standard" and "all analyses and approaches for
27 considering the level of the standard under review, including risk assessment and weight of
28 evidence methodologies" (Peacock, 2008, p. 3). Accordingly, the final 2008 REA included a
29 policy assessment chapter that considered the scientific evidence in the 2008 ISA and the
"Documents related to the current review as well as reviews completed in 2010 and 1996 are available at:
http://www.epa.gOv/ttn/naaqs/standards/nox/s nox index.html.
18The EPA conducted a separate review of the secondary NO2 NAAQS jointly with a review of the
secondary SO2 NAAQS. The Agency retained those secondary standards, without revision, to address the direct
effects on vegetation of exposure to gaseous oxides of nitrogen and sulfur (77 FR 20218, April 3, 2012).
19 Initial changes to the NAAQS review process included a policy assessment document reflecting Agency
(rather than staff) views published as an advanced notice of public rulemaking (ANPR). Under this process, the
ANPR would have been reviewed by CASAC (Peacock, 2006).
February 2014 1-9 Draft - Do Not Quote or Cite
-------
1 exposure and risk results presented in other chapters of the 2008 REA as they related to the
2 adequacy of the then current primary annual NC>2 standard and potential alternative standards for
3 consideration (U.S EPA, 2008b, chapter 10).20 CASAC discussed the final version of the 2008
4 REA, with an emphasis on the policy assessment chapter, during a public teleconference on
5 December 5, 2008 (73 FR 66895, November 12, 2008). Following that teleconference, CASAC
6 offered comments and advice on the primary NC>2 standard in a letter to the Administrator
7 (Samet, 2008b).
8 As discussed in more detail in section 3.1 below, after considering an integrative
9 synthesis of the body of evidence on human health effects associated with the presence of NC>2 in
10 the air and the exposure and risk information, the Administrator determined that the existing
11 primary NC>2 NAAQS, based on an annual arithmetic average, was not sufficient to protect the
12 public health from the array of effects that could occur following short-term exposures to
13 ambient NC>2. In so doing, the Administrator particularly noted the potential for adverse health
14 effects to occur following exposures to elevated NC>2 concentrations that can occur around major
15 roads (75 FR 6482). In a notice published in the Federal Register on July 15, 2009, the EPA
16 proposed to supplement the existing primary annual NC>2 standard by establishing a new short-
17 term standard (74 FR 34404). In a notice published in the Federal Register on February 9, 2010,
18 the EPA finalized a new short-term NC>2 standard with a level of 100 ppb, based on the 3-year
19 average of the 98th percentile of the annual distribution of daily maximum 1-hour concentrations.
20 The EPA also retained the existing primary annual NC>2 standard with a level of 53 ppb, annual
21 average (75 FR 6474). The Agency's final decision included consideration of CASAC (Samet,
22 2009) and public comments on the proposed rule.
23 Revisions to the NAAQS were accompanied by revisions to the data handling procedures,
91
24 the ambient air monitoring and reporting requirements, and the Air Quality Index (AQI). As
25 described in section 6.2 below, one aspect of the new monitoring network requirements included
20 Subsequent to the completion of the 2008 REA, EPA Administrator Jackson called for additional key
changes to the NAAQS review process including reinstating a policy assessment document that contains staff
analysis of the scientific bases for alternative policy options for consideration by senior Agency management prior
to rulemaking (Jackson, 2009). As discussed in Chapter 7 of this document, a Policy Assessment will be developed
for this review.
21 The current federal regulatory measurement methods for NO2 are specified in 40 CFR part 50, Appendix
F and 40 CFR part 53. Consideration of ambient air measurements with regard to judging attainment of the
standards is specified in 40 CFR part 50, Appendix S. The NO2 monitoring network requirements are specified in 40
CFR part 58, Appendix D, section 4.3. The EPA revised the AQI for NO2 to be consistent with the revised primary
NO2 NAAQS as specified in 40 CFR part 58 Appendix G. Guidance on the approach for implementation of the new
standard was described in the Federal Register notices for the proposed and final rules (74 FR 34404; 75 FR 6474).
February 2014 1-10 Draft - Do Not Quote or Cite
-------
1 requirements for States to locate monitors near heavily trafficked roadways in large urban areas
2 and in other locations where maximum NC>2 concentrations can occur. Subsequent to the 2010
3 rulemaking, the EPA revised the deadlines by which the near-road monitors are to be operational
4 in order to implement a phased deployment approach (78 FR 16184, March 14, 2013). As
5 discussed in section 6.2 below, the near-road NO2 monitors will become operational between
6 January 1, 2014 and January 1, 2017.
7 1.4 SCOPE OF THE CURRENT REVIEW
8 Section 108(c) of the CAA specifies that the air quality criteria relating to NC>2 include
9 consideration of nitric and nitrous acids, nitrites, nitrates, nitrosamines, and other derivatives of
10 oxides of nitrogen, including multiple gaseous and particulate species. This includes gases such
11 as NC>2 and nitric oxide (NO) as well as their gaseous and particulate reaction products (e.g.,
12 organic and inorganic nitrates and nitrites, nitro-polycyclic aromatic hydrocarbons) (U.S. EPA,
13 2013b, section 2.2, Figure 2-1). Collectively, we refer to this set of species as NOy. Id. As in
14 previous reviews, this review will focus on effects associated with the gaseous NOy species.
15 Effects associated with the particulate species (e.g., nitrate) are addressed in the review of the
16 NAAQS for particulate matter (PM) (78 FR 30866, January 15, 2013; U.S. EPA, 2009).22
17 Consistent with the review completed in 2010, this review is focused on the primary
18 standards and as such will only consider relevant scientific information related to potential health
19 effects associated with exposure to oxides of nitrogen. The EPA is separately reviewing the
20 secondary standard for oxides of nitrogen in conjunction with a review of the secondary SO2
21 standard (78 FR 53452, August 29, 2013).23
22 When referring to the group of gaseous oxidized nitrogen compounds as a whole, the ISA and other
assessment documents developed in this review will use the term "oxides of nitrogen." Based on the definition
commonly used in the scientific literature, the abbreviation NOX will refer specifically to the sum of NO2 and NO
concentrations (U.S. EPA, 2013b, section 2.2).
23 Additional information on the ongoing review of the secondary NO2 and SO2 standards is available at:
http://www.epa.gov/ttn/naaqs/standards/no2so2sec/index.html.
February 2014 1-11 Draft - Do Not Quote or Cite
-------
i 2 STATUS AND SCHEDULE
2 In February 2012, the EPA announced the initiation of the current periodic review of the
3 air quality criteria for oxides of nitrogen and the primary NC>2 NAAQS and issued a call for
4 information in the Federal Register (77 FR 7149, February 10, 2012). Also, as an initial step in
5 the NAAQS review process described in section 1.1 above, the EPA invited a wide range of
6 scientific experts (from EPA and outside organizations) to participate in a workshop to discuss
7 the policy-relevant science to inform the development of this draft IRP. Id. These experts
8 represented a variety of scientific disciplines, including epidemiology, human and animal
9 toxicology, statistics, risk/exposure analysis, and atmospheric science. This workshop was held
10 February 29 to March 1, 2012 in Research Triangle Park, NC and provided an opportunity for
11 the participants to broadly discuss the key policy-relevant issues around which the EPA would
12 structure this review of the primary NC>2 NAAQS and the most meaningful new science that
13 would be available to inform our understanding of these issues.24 Based in part on the workshop
14 discussions, the EPA developed the Draft Plan for Development of the Integrated Science
15 Assessment (ISA) for Nitrogen Oxides - Health Criteria (U.S. EPA, 2013a)25 and this draft IRP
16 outlining the schedule, the process, and the policy-relevant science issues identified as key to
17 guiding the evaluation of the air quality criteria for oxides of nitrogen and the review of the
18 primary NO2 NAAQS.
19 Table 2-1 outlines the schedule under which the Agency is currently conducting this
20 review. The scope of the review and the key documents to be prepared during the review are
21 discussed throughout the rest of this document.
24 Workshop materials are available in the rulemaking docket accessible through
http://www.regulations.gov. Docket ID number EPA-HQ-OAR-2013-0146.
25 The EPA released a draft plan outlining the plans for developing the ISA for CASAC consultation and
public review (78 FR 26026, May 3, 2013). The EPA held a consultation with CASAC on this draft plan during a
public teleconference on June 5, 2013 (78 FR 27234, May 9, 2013). CASAC and public comments on the draft plan
were considered in developing Chapter 4 of this draft IRP.
February 2014 2-1 Draft - Do Not Quote or Cite
-------
Table 2-1. Anticipated schedule for the current review.
Stage of Review
Major Milestone
Target Date
Integrated Review
Plan (IRP)
Literature Search
Federal Register Call for Information
Workshop on science/policy issues
Draft plan for developing ISA
CAS AC consultation on draft ISA plan
Draft IRP
CAS AC review of draft IRP
Final IRP
Ongoing
February 10, 2012
February 29 - March 1, 2012
May 2013
June 5, 2013
February 2014
March 12-13, 2014
June 2014
Integrated Science
Assessment (ISA)
First draft ISA
CAS AC public meeting for review of first draft ISA
Second draft ISA
CASAC/public review of second draft ISA
Final ISA
November 2013
March 12-13, 2014
August 2014
October 2014
February 2015
Risk/Exposure
Assessment (REA)
REA Planning Document
CAS AC consultation/public review of REA Planning
Document
If warranted,
First draft REA
CAS AC/public review of first draft REA
Second draft REA
CAS AC/public review of second draft REA
Final REA
September 2014
October 2014
TBD
Policy Assessment
(PA)/Rulemaking
First draft PA
CAS AC/public review of first draft PA
Second draft PA26
CAS AC/public review of second draft PA
Final PA
Notice of proposed rulemaking
Notice of final rulemaking
January 2015
February 2015
October 2015
November 2015
April 2016
September 2016
June 2017
The anticipated schedule presented in Table 2-1 includes preparation of two draft PAs for CAS AC and
public review. However, in NAAQS reviews where a new REA is not developed and where staff preliminarily
conclude in a first draft PA that it is appropriate to consider retaining the current standards, without revision, the
EPA may decide that there is no new substantive information that we would intend to add that would provide a basis
for preparing a second draft PA. In NAAQS reviews in which the newly available information calls into question the
adequacy of the current standard(s), a second draft PA is typically prepared to include staff consideration of
potential alternative standards. If the Agency determines that a second draft PA is not warranted, CAS AC and public
comments on the first draft PA will be considered in preparing the final PA and the schedule adjusted accordingly.
February 2014
2-2
Draft - Do Not Quote or Cite
-------
i 3 KEY POLICY-RELEVANT ISSUES
2 In each NAAQS review, an initial step is to address the following overarching question:
3 • Does the currently available scientific evidence and exposure/risk-based
4 information support or call into question the adequacy of the protection
5 afforded by the current standard(s)?
6 As appropriate, reviews also address a second overarching question:
7 • What alternative standards, if any, are supported by the currently available
8 scientific evidence and exposure/risk-based information, and are appropriate
9 for consideration?
10 To inform our evaluation of these overarching questions in the current review, we have identified
11 key policy-relevant issues to be considered. These key issues reflect aspects of the health effects
12 evidence, air quality information, and exposure/risk information that, in our judgment, are likely
13 to be particularly important to informing the Administrator's decisions. They build upon the key
14 issues that were important in previous reviews.
15 Section 3.1 below describes the key considerations and conclusions from the last review
16 with regard to the adequacy of the primary NC>2 standards (section 3.1.1), and with regard to the
17 elements for a revised suite of standards judged in that review to provide requisite public health
18 protection (section 3.1.2). Section 3.2 summarizes our general approach for reviewing the
19 primary NC>2 standards in the current review and outlines the key policy-relevant issues. These
20 issues are presented as a series of questions that will frame our approach to considering the
21 extent to which the available evidence and information support retaining or revising the current
22 primary standards for NC>2.
23 3.1 CONSIDERATIONS AND CONCLUSIONS IN LAST REVIEW
24 The last review of the primary NO2 NAAQS was completed in 2010 (75 FR 6474). In
25 consideration of health effects evidence and air quality and exposure/risk information available
26 in that review, the EPA established a new short-term standard to provide increased public health
97
27 protection, including for asthmatics and other at-risk populations, against an array of adverse
27 As used here and similarly throughout this document, the term population refers to persons having a
quality or characteristic in common, such as a specific pre-existing illness or a specific age or lifestage, with
lifestage referring to a distinguishable time frame in an individual's life characterized by unique and relatively stable
behavioral and/or physiological characteristics that are associated with development and growth. Identifying at-risk
populations includes consideration of intrinsic (e.g., genetic or developmental aspects) or acquired (e.g., disease or
smoking status) factors that increase the risk of health effects occurring with exposure to oxides of nitrogen as well
as extrinsic, nonbiological factors such as those related to socioeconomic status, reduced access to health care, or
exposure.
February 2014 3-1 Draft - Do Not Quote or Cite
-------
1 respiratory health effects that had been linked to short-term NC>2 exposures (75 FR 6498 to 6502,
2 February 9, 2010; U.S. EPA, 2008a, sections 3.1.7 and 5.3.2.1; Table 5.3-1) (75 FR 6502).
3 Specifically, EPA established a short-term standard defined by the 3-year average of the 98*
4 percentile of the yearly distribution of daily maximum 1-hour NC>2 concentrations, with a level
5 of 100 ppb. In addition to setting the new 1-hour standard, EPA retained the annual standard of
6 53 ppb (75 FR 6502). Together, the two standards were concluded to provide protection for
7 susceptible groups against adverse respiratory health effects associated with short-term
8 exposures to NO2 and effects potentially associated with long-term exposures. As discussed
9 further in section 6.2 below, in conjunction with the revised primary NC>2 NAAQS the EPA also
10 established a two-tiered monitoring network composed of: (1) near-road monitors which would
11 be placed in locations of expected maximum 1-hour NC>2 concentrations near heavily trafficked
12 roads in urban areas and (2) monitors located to characterize areas with the highest expected NO2
13 concentrations at the neighborhood and larger spatial scales (also referred to as "area-wide"
14 monitors) (75 FR 6505 to 6506, February 9, 2010).
15 Key policy-relevant aspects of the Administrator's decisions with regard to need to revise
16 the primary NC>2 NAAQS, and with regard to the elements of the revised standard, are described
17 below in sections 3.1.1 and 3.1.2, respectively. Areas of uncertainty identified in the last review
18 are noted in section 3.1.3.
19 3.1.1 Need for Revision
20 The 2010 decision to revise the existing primary NC>2 standard was based on the extensive
21 body of evidence published through early 2008 and assessed in the 2008 ISA (U.S. EPA, 2008a),
22 including the assessment of the policy-relevant aspects of that evidence;28 the quantitative
23 exposure and risk analyses presented in the REA (U.S. EPA, 2008b); the advice and
24 recommendations of CASAC (Samet, 2008b); and public comments (U.S. EPA, 2010). The
25 scientific evidence included controlled human exposure studies providing evidence of airway
26 hyperresponsiveness in asthmatics following short-term exposures to NC>2 concentrations as low
27 as 100 ppb, and epidemiological studies reporting associations between short-term NC>2 and
28 respiratory effects in locations that would have met the annual standard. The quantitative
29 analyses presented in the 2008 REA included exposure and risk estimates for air-quality adjusted
30 to just meet the annual standard. Based on the evidence and exposure/risk information, and based
31 on CASAC's advice that "the current NAAQS does not protect the public's health and that it
32 should be revised" (Samet, 2008b, p. 2), the Administrator concluded that the existing primary
As noted in section 1.3 above, due to changes in the NAAQS process, the last review of the NO2
NAAQS did not include a separate Policy Assessment. Rather, the REA for that review included a Policy
Assessment chapter.
February 2014 3-2 Draft - Do Not Quote or Cite
-------
1 annual NC>2 standard alone was not sufficient to protect public health from the array of
2 respiratory effects that had been reported following short-term exposures to oxides of nitrogen
3 (75 FR 6488 to 6490, February 9, 2010).
4 As an initial consideration in reaching this decision, the Administrator noted that the
5 evidence relating short-term (minutes to hours) NC>2 exposures to respiratory morbidity was
6 judged in the ISA to be "sufficient to infer a likely causal relationship" (75 FR 6489; U.S. EPA,
7 2008a, sections 3.1.7 and 5.3.2.1).29 This evidence included a large body of epidemiological
8 studies reporting associations between short-term NC>2 concentrations measured at central-site
9 monitors and respiratory-related symptoms, emergency department visits, and hospital
10 admissions. Overall, the 2008 ISA characterized the epidemiological evidence as consistent, in
11 that associations were reported in studies conducted in numerous locations with a variety of
12 methodological approaches, and coherent, in that the studies reported associations with
13 respiratory health outcomes that were logically linked together. In addition, a number of these
14 associations were statistically significant, particularly the more precise effect estimates (U.S.
15 EPA, 2008a, section 5.3.2.1). In studies that evaluated concentration-response (C-R)
16 relationships, they appeared linear within the observed range of data with "little evidence of any
17 effect threshold" (U.S. EPA, 2008a, sections 4.2 and 5.3.2.9). In considering the epidemiological
18 evidence, the Administrator acknowledged that the interpretation of the studies is complicated by
19 the fact that on-road vehicle exhaust emissions are a nearly ubiquitous source of combustion
20 pollutant mixtures than include NC>2, but additionally noted ISA analyses of co-pollutants
21 generally found that NC>2 associations remained robust in multi-pollutant models (75 FR 6489).
22 The evidence also included controlled human exposure studies that evaluated airway
23 hyperresponsiveness in asthmatics following short-term (30-minute to 2-hour) exposures to NC>2
24 concentrations at or above 100 ppb, as well as supporting evidence from animal toxicological
25 studies (U.S. EPA, 2008a, sections 3.1.3 and 5.4). The EPA drew two broad conclusions
26 regarding airway responsiveness in asthmatics following NC>2 exposures. First, that NC>2
27 exposure may enhance the sensitivity to allergen-induced decrements in lung function and
28 increase the allergen-induced airway inflammatory response following 30-minute exposures of
29 asthmatic adults to NO2 concentrations as low as 260 ppb. (U.S. EPA, 2008a, section 5.3.2.1,
30 Figure 3.1-2). Second, that exposure to NC>2 resulted in small but significant increases in
31 nonspecific airway hyperresponsiveness in healthy and asthmatic adults. In asthmatics, the ISA
32 concluded that such increases were observed following 1-hour exposures to 100 ppb NO2 (U.S.
29In contrast, the evidence relating long-term (weeks to years) NO2 exposures to adverse health effects was
judged to be either "suggestive but not sufficient to infer a causal relationship" (respiratory morbidity) or
"inadequate to infer the presence or absence of a causal relationship" (mortality, cancer cardiovascular effects,
reproductive/developmental effects) (75 FR 6478).
February 2014 3-3 Draft - Do Not Quote or Cite
-------
1 EPA,2008a, sections 3.1.3.2; 5.3.2.1). The EPA further concluded that the majority of
2 asthmatics may experience NCVrelated airway hyperresponsiveness following short-term NC>2
3 exposures between 100 and 300 ppb (U.S. EPA, 2008a, Table 3.1-3; U.S. EPA, 2008b, p. 283).
4 Enhanced airway responsiveness could have important clinical implications for asthmatics since
5 transient increases in airway responsiveness following NC>2 exposure have the potential to
6 increase symptoms and worsen asthma control (74 FR 34415, July 15, 2009; U.S. EPA, 2008a,
7 sections 5.3.2.1 and 5.4). An update to a meta-analysis of data for nonspecific airway
8 responsiveness that had been considered in the previous review provided support to the
9 conclusions on exposure concentrations eliciting effects (Folinsbee, 1992; U.S. EPA, 1993, 60
10 FR 52818, October 11, 1995; U.S. EPA, 2008a, section 3.1.3.2, Tables 3.1-2 and 3.1-3).30
11 The exposure-and risk-based information further informed the Administrator's decisions
12 regarding adequacy of the then-existing NC>2 primary standard. The Administrator took note of
13 the REA conclusion that risks estimated for air quality adjusted upward to simulate just meeting
14 the current standard could reasonably be concluded to be important from a public health
15 perspective, while additionally recognizing the uncertainties associated with adjusting air quality
16 in such analyses (75 FR 6489).31 For air quality adjusted to just meet the existing annual
17 standard, the REA findings given particular attention by the Administrator included the
18 following: "a large percentage (8-9%) of respiratory-related ED visits in Atlanta could be
19 associated with short-term NC>2 exposures; most asthmatics in Atlanta could be exposed on
20 multiple days per year to NC>2 concentrations at or above 300 ppb, and most locations evaluated
21 could experience on-/near-road NO2 concentrations above 100 ppb on more than half of the days
22 in a given year" (75 FR 6489; U.S. EPA, 2008b, section 10.3.2).32 The 2008 REA additionally
30The changes made to the analysis were to remove the results of one allergen study and to add the results
from a non-specific responsiveness study, and to discuss results for an additional exposure concentration (i.e., 100
ppb) (U.S. EPA, 2008a, section 3.1.3.2).
31 As described further in chapter 5 below, the 2008 REA considered air quality data from the existing
network of ambient monitors as well as data from controlled human exposure studies and epidemiological studies to
model exposure to NO2 and to estimate health risks associated with short-term exposures. Additionally, recognizing
that large segments of the public live, work, go to school, or travel on or near roads, the 2008 REA also estimated
linti ini^.,^ o^^iii^iiikj \J-L iii^ j_/un_<-i-i-s^ i.iv^/2 »»v/iiv^ £)*-' i*J iJW'J.i.v/v/
exposures that would occur in these particular locations.
32 -
' Estimates were developed for: (1) an "as-is" scenario in which it estimated the health risks associated
with short-term exposure to NO2 at actual recent air-quality concentrations, which were lower than what was
permitted by the then current annual NO2 standard; (2) a "just meets" scenario in which it estimated the health risks
associated with air quality adjusted upward to simulate just meeting the then current annual standard; and (3) other
scenarios for potential alternative standards. The 2008 REA's health risk estimates were based on actual or modeled
ambient concentrations at pre-2010 air quality monitors. Those monitors primarily measured NO2 concentrations
that were representative of a broad geographic area (e.g., area-wide ambient measurements ) rather than
concentrations at specific locations where the highest concentrations of NO2 were likely to be found (e.g., maximum
or peak ambient measurements including near major roadways).
Area-wide monitors are defined as those sited at neighborhood, urban, and regional scales, as well as those
monitors sited at either a micro- or middle-scale that are representative of many such locations in the same Core
February 2014 3-4 Draft - Do Not Quote or Cite
-------
1 found that, under the "as is" scenario (i.e., recent air quality concentrations), individuals
2 spending time on or near roads could expect to experience short-term NC>2 exposures above
3 health effect benchmark levels of concern33 multiple times per year.
4 In reaching the conclusion on adequacy of the then-existing standard, the Administrator
5 also considered advice received from CASAC. In their advice, CASAC agreed that the primary
6 concern in the review was to protect against health effects that have been associated with short-
7 term NC>2 exposures. CASAC also agreed that the annual standard alone was not sufficient to
8 protect public health against the types of exposures that could lead to these health effects. As
9 noted in their letter to the EPA Administrator, ''CASAC concurs with EPA's judgment that the
10 current NAAQS does not protect the public's health and that it should be revised" (Samet,
11 2008b).
12 Based on the considerations summarized above, the Administrator concluded that the
13 then-existing NC>2 primary NAAQS was not requisite to protect public health with an adequate
14 margin of safety and that the standard should be revised in order to provide increased public
15 health protection against respiratory effects associated with short-term exposures, particularly for
16 susceptible populations such as asthmatics, children, and older adults (75 FR 6490). Upon
17 consideration of approaches to revising the standard, the Administrator concluded that it was
18 appropriate to set a new short-term standard, as described below.
19 3.1.2 Elements of Revised Standard
20 In considering appropriate revisions in the last review, each of the four basic elements of
21 the NAAQS (indicator, averaging time, level, and form) was evaluated. The rationale for
22 decisions on those elements is summarized below.
23 3.1.2.1 Indicator
24 In previous reviews, the EPA focused on NC>2 as the most appropriate indicator for oxides
25 of nitrogen because the available scientific information regarding health effects was largely
26 indexed by NC>2. In the review completed in 2010, controlled human exposure studies and animal
27 toxicological studies provided specific evidence for health effects following exposures to NC>2.
28 In addition, epidemiological studies typically reported effects associated with NC>2
29 concentrations though the degree to which monitored NC>2 reflected actual NO2 concentrations,
Based Statistical Area(CBSA) (40 CFR58.1). The introduction and first use of the term "area-wide" was in the
final rule for the last primary NO2 NAAQS review as part of the NO2 minimum monitoring requirements (75 FR
6504, February 9, 2010). The term was formally defined in the final rule for the most recent PM NAAQS review (78
FR 3235 and 3281 to 3282, January 15, 2013). The underlying spatial scales are defined in 40 CFR part 58
Appendix D, section 1.2(b).
33 Health effect benchmark levels evaluated in the 2008 REA ranged from 100 to 300 ppb based on
increased airway hyperresponsiveness in asthmatics (from controlled human exposure studies) (U.S. EPA, 2008b,
section 6.2).
February 2014 3-5 Draft - Do Not Quote or Cite
-------
1 as opposed to NC>2 plus other gaseous oxides of nitrogen, was recognized as an uncertainty (75
2 FR 6490, February, 9, 2010; U.S. EPA 2008b, section 2.2.3).
3 Based on the information available in the last review, and consistent with the views of
4 CASAC (Samet, 2008b, p.2; Samet, 2009, p.2), the Agency concluded it was appropriate to
5 continue to use NC>2 as the indicator for a standard that was intended to address effects associated
6 with exposure to NC>2, alone or in combination with other gaseous oxides of nitrogen. In so
7 doing, the EPA recognized that measures leading to reductions in population exposures to NC>2
8 will also reduce exposures to other oxides of nitrogen (75 FR 6490).
9 3.1.2.2 Averaging times
10 In considering the most appropriate averaging time for the NC>2 primary NAAQS, the
11 Administrator noted the available scientific evidence as assessed in the ISA, the air quality
12 analyses presented in the REA, the conclusions of the policy assessment chapter of the REA,
13 CASAC recommendations, and public comments received (75 FR 6490). Her key considerations
14 are summarized below.
15 When considering averaging time, the Administrator first noted that the evidence relating
16 short-term (minutes to hours) NC>2 exposures to respiratory morbidity was judged in the ISA to
17 be "sufficient to infer a likely causal relationship" (U.S. EPA, 2008a, section 5.3.2.1) while the
18 evidence relating long-term (weeks to years) NO2 exposures to adverse health effects was judged
19 to be either "suggestive but not sufficient to infer a causal relationship" (respiratory morbidity)
20 or "inadequate to infer the presence or absence of a causal relationship" (mortality, cancer,
21 cardiovascular effects, reproductive/developmental effects) (U.S. EPA, 2008a, sections 5.3.2.4-
22 5.3.2.6). The Administrator concluded that these judgments most directly supported an
23 averaging time that focused protection on short-term exposures to NC>2.
24 As had been done in previous reviews of the NC>2 NAAQS, the Administrator next noted
25 that it is instructive to evaluate the potential for a standard based on annual average NO2
26 concentrations, as was the existing standard at the time of the 2010 review, to provide protection
27 against short-term NC>2 exposures. To this end, the Administrator considered REA analyses that
28 indicated a relatively large degree of variability in ratios of short-term (i.e., 1-hour and 24-hour)
29 NC>2 concentrations to annual average concentrations, suggesting that a standard based on annual
30 average NO2 concentrations would not likely be an effective or efficient approach to focus
31 protection on short-term NC>2 exposures. For example, these analyses indicated that in some
32 areas the existing annual standard could allow 1-hour daily maximum NC>2 concentrations of
33 about 400 ppb, while in other areas the annual standard could limit 1-hour daily maximum NC>2
34 concentrations to about 150 ppb. Thus, for purposes of protecting against the range of 1-hour
35 NC>2 exposures, the Administrator agreed with the REA conclusion that a standard based on
February 2014 3-6 Draft - Do Not Quote or Cite
-------
1 annual average concentrations would likely require more control than necessary in some areas
2 and less control than necessary in others, depending on the standard level selected.
3 In next considering the level of support available for specific short-term averaging times,
4 the Administrator noted that the policy assessment chapter of the REA considered evidence from
5 both experimental and epidemiologic studies. Controlled human exposure studies and animal
6 toxicological studies provided evidence that NC>2 exposures from less than 1-hour up to 3-hours
7 can result in respiratory effects such as increased airway responsiveness and inflammation (ISA,
8 section 5.3.2.7). She specifically noted the ISA conclusion that exposures of asthmatic adults to
9 100 ppb NO2 for 1-hour can result in small but significant increases in nonspecific airway
10 responsiveness (U.S. EPA, 2003a, section 5.3.2.1). In addition, the epidemiologic literature
11 provided support for short-term averaging times ranging from approximately 1-hour up to 24-
12 hours (U.S. EPA, 2003a, section 5.3.2.7). Based on this, the Administrator concluded that a
13 primary concern with regard to averaging time is the degree of protection provided against 1-
14 hour NC>2 concentrations. Based on REA analyses of ratios between 1-hour and 24-hour NC>2
15 concentrations, she further concluded that a standard based on 1-hour daily maximum NC>2
16 concentrations could also be effective at protecting against 24-hour NO2 exposures.
17 Based on the above, the Administrator judged that it was appropriate to set a new NO2
18 standard with a 1-hour averaging time. She concluded that such a standard can effectively limit
19 short-term (i.e., 1- to 24-hours) exposures that have been linked to adverse respiratory effects.
20 She also retained the existing annual standard to continue to provide protection for effects
21 potentially associated with long-term exposures to oxides of nitrogen (75 FR 6502). These
22 decisions were consistent with CASAC advice to establish a primary short-term standard for
23 oxides of nitrogen based on using 1-hour maximum NC>2 concentrations and to retain the current
24 annual standard34 (Samet, 2008b, p. 2; Samet, 2009, p. 2).
25 3.1.2.3 Levels
26 With consideration of the available health effects evidence, exposure and risk analyses,
27 and air quality information, the Administrator set the level of the new 1-hour NC>2 standard at
28 100 ppb. This standard was focused on limiting the maximum 1-hour NC>2 concentrations in
29 ambient air, including in locations near major roadways where the highest ambient NC>2
34 CASAC advised that "the findings of the REA do not provide assurance that a short-term standard based
on the one-hour maximum will necessarily protect the populations from long-term exposures at levels potentially
leading to adverse health effects" therefore, it recommended retaining the existing annual standard (Samet, 2008b, p.
2).
February 2014 3-7 Draft - Do Not Quote or Cite
-------
1 concentrations can occur in urban areas (75 FR 6474).35 In establishing this new standard, the
2 Administrator emphasized the importance of protecting against exposures to peak concentrations
3 of NC>2, such as those that can occur around major roadways. Available evidence and
4 information suggested that roadways account for the majority of exposures to peak NO2
5 concentrations and, therefore, are important contributors to NO2-associated public health risks.
6 In setting the level of the new 1-hour standard at 100 ppb, the Administrator noted that
7 there is no bright line clearly directing the choice of level. Rather, the choice of what is
8 appropriate is a public health policy judgment entrusted to the Administrator. This judgment
9 must include consideration of the strengths and limitations of the evidence and the appropriate
10 inferences to be drawn from the evidence and the exposure and risk assessments.
11 The Administrator judged that the existing evidence from controlled human exposure
12 studies supported the conclusion that the NCVinduced increase in airway responsiveness at or
13 above 100 ppb presents a risk of adverse effects for some asthmatics, especially those with more
14 serious (i.e., more than mild) asthma. The Administrator noted that the risks associated with
15 increased airway responsiveness cannot be fully characterized based on available controlled
16 human exposure studies, and thus she was not able to determine whether the increased airway
17 responsiveness experienced by asthmatics in these studies is an adverse health effect. However,
18 the Administrator concluded that asthmatics, particularly those suffering from more severe
19 asthma, warrant protection from the risk of adverse effects associated with the NO2-induced
20 increase in airway responsiveness. Therefore, the Administrator concluded that the controlled
21 human exposure evidence supported setting a standard level no higher than 100 ppb to reflect a
22 cautious approach to the uncertainty regarding the adversity of the effect. However, those
23 uncertainties led her to also conclude that this evidence did not support setting a standard level
24 lower than 100 ppb.
25 The Administrator also considered the more serious health effects reported in NO2
26 epidemiologic studies. She noted that a new standard focused on protecting against maximum 1-
27 hour NC>2 concentrations in ambient air anywhere in an area, with a level of 100 ppb and an
28 appropriate form (as discussed below), would be expected to limit area-wide36 NO2
29 concentrations to below those in locations where epidemiologic studies had reported associations
30 with respiratory-related hospital admissions or emergency department visits. The Administrator
35In conjunction with this new standard, the Administrator established a 2-tiered monitoring network that
included monitors sited to measure the maximum NO2 concentrations near major roadways, as well as monitors
sited to measure maximum area-wide NO2 concentrations (section 6, below).
36As discussed above, area-wide concentrations refer to those measured by monitors that have been sited to
characterize ambient concentrations at the neighborhood and larger spatial scales (see also section 6, below).
February 2014 3-8 Draft - Do Not Quote or Cite
-------
1 also concluded that such a 1-hour standard would be consistent with the REA conclusions based
2 on the NC>2 exposure and risk information.
3 Given the above considerations and the comments received on the proposal, the
4 Administrator judged it appropriate to set a 1-hour standard focused on limiting the maximum
5 allowable NO2 concentrations that can occur anywhere in an area, with a level of 100 ppb.
6 Specifically, she concluded that such a standard, with an appropriate form as discussed below,
7 would provide a significant increase in public health protection compared to that provided by the
8 annual standard alone and would be expected to protect against the respiratory effects that have
9 been linked with NC>2 exposures in both controlled human exposure and epidemiologic studies.
10 This includes limiting exposures at and above 100 ppb for the vast majority of people, including
11 those in at-risk groups, and maintaining area-wide NC>2 concentrations well below those in
12 locations where key U.S. epidemiologic studies had reported that ambient NO2 is associated with
13 clearly adverse respiratory health effects, as indicated by increased hospital admissions and
14 emergency department visits. The Administrator also noted that a standard level of 100 ppb was
15 consistent with the consensus recommendation of CAS AC.
16 In setting the standard level at 100 ppb rather than a lower level, the Administrator also
17 acknowledged the uncertainties associated with the scientific evidence. She noted that a 1-hour
18 standard with a level lower than 100 ppb would only result in significant further public health
19 protection if, in fact, there is a continuum of serious, adverse health risks caused by exposure to
20 NO2 concentrations below 100 ppb and/or associated with area-wide NO2 concentrations well-
21 below those in locations where key U.S. epidemiologic studies had reported associations with
22 respiratory-related emergency department visits and hospital admissions. Based on the available
23 evidence, the Administrator did not believe that such assumptions were warranted. Taking into
24 account the uncertainties that remained in interpreting the evidence from available controlled
25 human exposure and epidemiologic studies, the Administrator noted that the likelihood of
26 obtaining benefits to public health with a standard set below 100 ppb decreases while the
27 likelihood of requiring reductions in ambient concentrations that go beyond those that are needed
28 to protect public health increases.
29 3.1.2.4 Forms
30 The "form" of a standard defines the air quality statistic that is to be compared to the level of
31 the standard in determining whether an area attains the standard. The Agency recognizes that for
32 short-term standards, concentration-based forms which reflect consideration of a statistical
33 characterization of an entire distribution of air quality data with a focus on a single statistical metric,
34 such as the 98th or 99th percentile, can better reflect pollutant-associated health risks than forms based
35 on expected exceedances. This is the case because concentration-based forms give proportionally
36 greater weight to days when pollutant concentrations are well above the level of the standard than to
February 2014 3-9 Draft - Do Not Quote or Cite
-------
1 days when the concentrations are just above the level of the standard.37 In addition, when averaged
2 over three years, these concentration-based forms are judged to provide an appropriate balance
3 between limiting peak pollutant concentrations and providing a stable regulatory target, facilitating
4 the development of stable implementation programs (75 FR 6492).
5 In the last review, the EPA considered two specific concentration-based forms (i.e., the 98th
6 and 99th percentile concentrations), averaged over 3 years, for the new 1-hour NO2 standard. The
7 focus on the upper percentiles of the distribution was based, in part, on evidence of health effects
8 associated with short-term NO2 exposures from experimental studies which provided information on
9 specific exposure concentrations that were linked to respiratory effects. The Agency proposed to
10 adopt either a 99th percentile or a 4th highest form, averaged over 3 years and also solicited comment
11 on both 98th percentile and 7th or 8th highest forms (74 FR 34430, July 15, 2008). Given the potential
12 for instability in the higher percentile concentrations and the absence of data from the proposed
13 two-tier monitoring network (e.g., around major roadways), CASAC, in a letter to the Administrator
14 following issuance of the Agency's proposed rule, recommended a form based on the 3-year average
15 of the 98th percentile of the distribution of 1-hour daily maximum NO2 concentrations (Samet, 2009,
16 p. 2).
17 In reaching her final decision in the last review, the Administrator recognized that the public
18 health protection provided by the new 1 -hour NO2 standard was based in large part on: (1) the
19 approach used to set the standard and (2) the level of the standard in conjunction with the form of the
20 standard (75 FR 6493, February 9, 2010). Given that the EPA set a new primary 1-hour NO2 standard
21 that focused on limiting the maximum allowable NO2 concentration in ambient air, the Agency
22 agreed with CASAC that an appropriate consideration with regard to form was the extent to which
23 specific statistics could be unstable at locations where maximum NO2 concentrations are expected
24 (e.g., including near major roads).
25 Given the limited available information on the variability in peak NO2 concentrations near
26 important sources of NO2 such as near major roadways, and given the recommendation from CASAC
27 of the potential for instability in the 99th percentile concentrations, the Administrator judged it
28 appropriate to set the form based on the 3-year average of the 98th percentile of the annual
29 distribution of daily maximum 1-hour NO2 concentrations. Id. In addition, consistent with CASAC
30 advice (Samet, 2008b, p. 2; Samet, 2009, p.2), the EPA retained the form of the annual standard (75
31 FR6502).
32 3.1.3 Areas of Uncertainty
33 While the available scientific information informing the last review was stronger and
34 more consistent than in previous reviews, and provided a strong basis for decision making in that
37 Compared to an exceedance-based form, a concentration-based form reflects the magnitude of the
exceedance of a standard level not just the fact that such an exceedance occurred.
February 2014 3-10 Draft - Do Not Quote or Cite
-------
1 review, the Agency recognized that areas of uncertainty remained. These were generally related
2 to: (1) understanding the role of NC>2 in the complex ambient mixture which includes a range of
3 co-occurring pollutants (e.g., PM2.5, ozone, CO, 802); (2) understanding the extent to which
4 monitored ambient NC>2 concentrations used in epidemiological studies reflect exposures in
5 study populations and the range of ambient concentrations over which we continue to have
6 confidence in the health effects observed in the epidemiological studies; (3) understanding the
7 extent to which the magnitude and potential adversity of NCVinduced respiratory effects
8 reported in controlled human exposures studies can be characterized; (4) understanding the NO2
9 concentration gradients around important sources, such as major roads, and relating those
10 gradients to broader ambient monitoring concentrations; and (5) an improved characterization of
11 NC>2 exposures and risk including alternative approaches for estimated risks associated with air
12 quality simulated to just meet current or alternative standards.
13 3.2 GENERAL APPROACH FOR THE CURRENT REVIEW
14 The approach for this review builds on the substantial body of work done during the
15 course of the last review, taking into account the more recent scientific information and air
16 quality data now available to inform our understanding of the key policy-relevant issues. The
17 approach described below is most fundamentally based on using the EPA's assessment of the
18 current scientific evidence and associated quantitative analyses to inform the Administrator's
19 judgments regarding primary standards for oxides of nitrogen that are requisite to protect public
20 health with an adequate margin of safety. This approach will involve translating scientific and
21 technical information into the basis for addressing a series of key policy-relevant questions using
22 both evidence- and exposure/risk-based considerations.38
23 Figure 3-1 summarizes the general approach, including consideration of the policy-
24 relevant questions which will frame the current review. The ISA, REA (if warranted), and PA
25 developed in this new review will provide the basis for addressing the key policy-relevant
26 questions and will inform the Administrator's judgment as to the adequacy of the current primary
27 NC>2 standards and decisions as to whether to retain or revise these standards. This approach
28 recognizes that the available health effects evidence generally reflects a continuum, consisting of
29 ambient concentrations at which scientists generally agree that health effects are likely to occur,
30 through lower concentrations at which the likelihood and magnitude of the response become
31 increasingly uncertain. Furthermore, this approach is consistent with the requirements of the
32 NAAQS provisions of the CAA and with how the EPA and the courts have historically
33 interpreted the CAA. As discussed in section 1.1 above, these provisions require the
38 Evidence-based considerations include those related to the health effects evidence assessed and
characterized in the ISA. Exposure/risk-based considerations draw from the results of the quantitative analyses.
February 2014 3-11 Draft - Do Not Quote or Cite
-------
1 Administrator to establish primary standards that, in the Administrator's judgment, are requisite
2 to protect public health with an adequate margin of safety. In so doing, the Administrator seeks
3 to establish standards that are neither more nor less stringent than necessary for this purpose. The
4 CAA does not require that primary standards be set at a zero-risk level, but rather at a level that
5 avoids unacceptable risks to public health. The four basic elements of the NAAQS (i.e.,
6 indicator, averaging time, form, and level) will be considered collectively in evaluating the
7 health protection afforded by the current or any alternative standards considered.
8 We note that the final decision on the adequacy of the current standards and, if
9 appropriate, potential alternative standards, is largely a public health policy judgment to be made
10 by the Administrator. The Administrator's final decision must draw upon scientific information
11 and analyses about health effects, population exposure and risks, as well as judgments about how
12 to consider the range and magnitude of uncertainties that are inherent in the scientific evidence
13 and analyses. As in the previous review as well as other recent NAAQS reviews, the EPA will
14 consider the implications of placing more or less weight or emphasis on different aspects of the
15 scientific evidence and exposure/risk-based information to inform the public health policy
16 judgments that the Administrator will make in reaching final decisions on whether to retain or
17 revise the current standards in this review.
February 2014 3-12 Draft - Do Not Quote or Cite
-------
Adequacy of Current Suite of NO2 Standards?
I
Evidence-based Considerations
> Does currently available evidence and
related uncertainties strengthen or call into
question prior conclusions?
^ Evidence of health effects not
previously identified?
^Evidence of effects at lower
concentrations than previously observed or
in areas that would have likely met current
standards?
^Expanded understanding of at-risk
lifestages and populations?
^Does newly available information call into
question any of the basic elements of the
standards?
Risk/Exposure-based Considerations
> Nature, magnitude, and uncertainties of
estimated exposures and risks remaining
upon just meeting the current suite of
standards?
> Roles of annual and 1-hour standards?
> Importance of remaining risks from public
health perspective?
>Uncertainties in the exposure and risk
estimates?
Does information
call into question
adequacy of current
suite of NO,
standards?
Consideration of Potential Alternative Suites of Standards
Consider
retaining
current suite
of NO,
standards
Elements of Potential Alternative Standards
> Indicator
"> Averaging times
^Forms
>Levels
Potential alternative standards for consideration
Figure 3-1. Overview of General Approach for Review of Primary NOi Standards
February 2014
3-13
Draft - Do Not Quote or Cite
-------
1 The initial overarching question in reviewing the adequacy of the current suite of primary
2 NC>2 NAAQS is whether the available body of scientific evidence, assessed in the ISA and used
3 as a basis for developing or interpreting any risk/exposure analyses, supports or calls into
4 question the scientific conclusions reached in the last review regarding health effects related to
5 exposures to oxides of nitrogen. The evaluation of the available scientific evidence and
6 risk/exposure information with regard to adequacy of the current standards will focus on key
7 policy-relevant issues by addressing a series of questions including the following:
8 • To what extent has new information altered the scientific support for the occurrence of
9 health effects as a result of short- and/or long-term exposure to oxides of nitrogen in the
10 ambient air?
11 o What evidence is available from recent studies focused on specific chemical
12 components within the broader group of oxides of nitrogen (e.g., NC>2, NO, NOx) to
13 inform our understanding of the nature of exposures that are linked to various health
14 outcomes?
15 o To what extent is key scientific evidence becoming available to improve our
16 understanding of the health effects associated with various time periods of exposures,
17 including peak (e.g., 1-hour) and chronic exposures (e.g., more than one month to
18 years)?
19 o At what pollutant concentrations do these health effects occur?
20 o Is there evidence of effects at exposure concentrations lower than previously
21 observed or in areas that would likely have met the current primary NC>2 standards?
22 o To what extent is new information available to improve the characterization of the
23 magnitude and/or potential adversity of NO2-induced respiratory effects reported in
24 controlled human exposure studies?
25 o To what extent is new information available to improve our understanding of the
26 range of ambient concentrations over which we continue to have confidence in the
27 health effects observed in the epidemiological studies?
28 o To what extent are health effects associated with exposures to oxides of nitrogen,
29 including NC>2, as opposed to one or more co-occurring pollutants (e.g., PM2.5, ozone,
30 SO2)?
31 o Has new information altered our understanding of human lifestages and populations
32 that are particularly at increased risk for experiencing health effects associated with
33 exposure to oxides of nitrogen?
34 o Is there new information to shed light on the nature the of exposure-response
35 relationship in different at-risk lifestages and/or populations?
36 o Is there new or emerging evidence on health effects beyond respiratory effects in
37 asthmatics or effects in high exposure populations (e.g., people living, working, or
38 going to school in near-road environments) that suggest potential additional at-risk
39 lifestages and populations should be given increased focus in this review?
February 2014 3-14 Draft - Do Not Quote or Cite
-------
1 • To what extent is new information available to improve our understanding of the NC>2
2 concentration gradients around important sources, such as major roads and combustion
3 sources, and to relate those gradients to broader ambient monitoring concentrations?
4 • To what extent does risk or exposure information suggest that exposures of concern are
5 likely to occur with recent ambient NC>2 concentrations or with concentrations that just
6 meet the current primary NC>2 standards?
7 o Are the estimated risks/exposures considered in this review of sufficient magnitude
8 such that the health effects might reasonably be judged to be important from a public
9 health perspective?
10 o What are the important uncertainties associated with any risk/exposure estimates?
11 • To what extent have important uncertainties identified in the last review been reduced
12 and/or have new uncertainties emerged?
13 • To what extent does newly available information reinforce or call into question any of the
14 basic elements of the current primary NC>2 standards?
15 If the evidence suggests that revision of the current standards might be appropriate, the
16 EPA will address a second overarching question related to what alternative standards are
17 appropriate for consideration. Specifically, we will evaluate how the scientific information and
18 assessments inform decisions regarding the basic elements of the primary NC>2 NAAQS:
19 indicator, averaging time, level, and form. These elements will be considered collectively in
20 evaluating the health protection afforded by the current or any alternative standards considered.
21 With regard to consideration of alternative standards, specific policy-relevant questions that will
22 be addressed include the following:
23 • To what extent does any new information provide support for consideration of a different
24 indicator for oxides of nitrogen in addition to or in place of NC>2?
25 • To what extent does the health effects evidence evaluated in the ISA, air quality analyses,
26 and, if available, new REA provide support for considering any different averaging
27 times?
28 • To what extent do air quality analyses and other information provide support for
29 consideration of alternative standard forms!
30 • What range of alternative standard levels should be considered based on the scientific
31 evidence evaluated in the ISA, air quality analyses and, if available, new REA39?
32 • What are the important uncertainties and limitations in the available evidence and
33 assessments and how might those uncertainties and limitations be taken into
39 As outlined in Table 2-1 and discussed in Chapter 6 below, the REA Planning Document will consider
the extent to which newly available scientific evidence and tools/methodologies warrant the conduct of new
quantitative risk and exposure assessments. To the extent completely new assessments are not developed for this
review, assessments from the last review may be interpreted in light of the newly available information in
addressing the key policy questions for the review.
February 2014 3-15 Draft - Do Not Quote or Cite
-------
1 consideration in identifying alternative standard indicators, averaging times, forms
2 and/or levels!
February 2014 3-16 Draft - Do Not Quote or Cite
-------
i 4 SCIENCE ASSESSMENT
2 The ISA comprises the science assessment phase of the NAAQS review process. As
3 outlined in section 1.4 above, the purpose of the current review is to inform the review of the
4 primary NC>2 standards only.40 Hence, the ISA will focus on updating the air quality criteria
5 associated with health effects evidence only.41
6 4.1 SCOPE OF THE ISA
7 The ISA provides an updated critical evaluation and synthesis of the current scientific
8 literature pertaining to known and anticipated effects on public health associated with the
9 presence of oxides of nitrogen in the ambient air, including the nature of any remaining or newly
10 identified uncertainties and limitations associated with the health evidence. Discussions in the
11 ISA will primarily focus on scientific evaluations that can inform the key policy questions
12 described in section 3.2 above. Although emphasis will be placed on the discussion of the health
13 effects information, other scientific information will also be presented and evaluated in order to
14 provide a better understanding of the following issues: (1) the sources of oxides of nitrogen to
15 ambient air; (2) measurement of and recent ambient concentrations of oxides of nitrogen
16 including NC>2, including subsequent fate and transport in the environment; and (3) important
17 considerations related to characterizing potential population exposures to oxides of nitrogen. The
18 process for evaluating and synthesizing scientific literature and addressing key policy questions
19 is detailed in the Preamble to the ISA.
20 The ISA is not intended to provide a detailed literature review but rather, will draw from
21 the existing body of evidence to synthesize the current state of knowledge on the most relevant
22 issues pertinent to the review of the primary NC>2 NAAQS. The ISA serves to revise the
23 scientific assessment available at the time of the last review. Thus, the ISA will build on the
24 conclusions of the last review of the air quality criteria for oxides of nitrogen as presented in the
25 2008 ISA and focus on peer-reviewed literature published since that document42 as well as on
26 any new interpretations of previously available literature. Key findings, conclusions, and
27 uncertainties from the 2008 ISA will be briefly summarized at the beginning of the ISA and of
40 As outlined in section 1.4 above, evidence related to potential welfare (e.g., ecosystem) effects of oxides
of nitrogen will be considered separately in the science assessment conducted as part of the review of the secondary
NAAQS for NO2 and SO2.
41In this review of the primary NAAQS for NO2, a draft plan for development of the ISA was prepared by
NCEA prior to development of this draft IRP. The draft plan for development of the ISA was made available for
public comment and was the subject of a consultation with CASAC (78 FR 26026; 78 FR 27234). Comments
received during that consultation have been considered in preparation of chapter 4 in this draft IRP. Further
comments received on this draft IRP will be considered in preparing the final IRP and the second draft ISA.
42 For the current ISA, searches were conducted for studies published beginning in January 2008.
February 2014 4-1 Draft - Do Not Quote or Cite
-------
1 individual sections. Important older studies may be discussed in detail to reinforce key concepts
2 and conclusions and/or if they are open to reinterpretation in light of newer data. Older studies
3 also may be the primary focus in some subject areas or scientific disciplines where research
4 efforts have subsided, and these older studies remain the definitive works available in the
5 literature. Emphasis will be placed on studies that examine health effects relevant to humans and
6 concentrations of oxides of nitrogen that represent the range of human exposures across ambient
7 microenvironments. Other studies, generally at higher exposure concentrations, may be included
8 if they contain unique data, such as previously unreported effects, evidence of the potential
9 biological mechanism(s) for an observed effect, or information on concentration-response
10 relationships.
11 4.2 ORGANIZATION OF THE ISA
12 The organization of the ISA for the health criteria of oxides of nitrogen will be consistent
13 with that used in the recent assessments for other criteria pollutants (e.g., ISA for Ozone and
14 Related Photochemical Oxidants, U.S. EPA, 2013c). The ISA will begin with a discussion of
15 major legal and historical aspects of prior NAAQS reviews as well as procedures for the
16 assessment of scientific information. An integrative synthesis chapter will summarize the key
17 information for each topic area, the causal determinations for relationships between exposure to
18 oxides of nitrogen and health effects, information describing the extent to which health effects
19 can be attributable specifically to oxides of nitrogen, and other uncertainties related to the
20 interpretation of scientific information. The integrative synthesis chapter also will discuss policy-
21 relevant issues such as the exposure averaging times and lags associated with health effects, the
22 concentration-response relationships including whether or not the evidence supports
23 identification of a discernible threshold below which effects are not likely to occur, and the
24 public health significance of effects associated with exposure to oxides of nitrogen. Subsequent
25 chapters are organized by subject area (see draft outline of the ISA in Appendix A) and contain
26 the detailed evaluation of results from recent studies integrated with previous findings (see
27 section 4.4 for specific issues to be addressed). Sections for each major health effect category
28 (e.g., respiratory effects) conclude with a causal determination about the relationship with
29 relevant exposures to oxides of nitrogen. The ISA will conclude with a chapter that examines
30 exposure and health outcome data to draw conclusions about potential at-risk lifestages and
31 populations.
32 The ISA may be supplemented with other materials if additional documentation is
33 required to support information contained within the ISA. These supplementary materials may
34 include more detailed and comprehensive coverage of relevant publications and may accompany
35 the ISA or be available in electronic form as output from the Health and Environmental Research
February 2014 4-2 Draft - Do Not Quote or Cite
-------
1 Online (HERO) database developed by EPA (http://hero.epa.gov/). Supplementary information
2 that is available in the HERO database will be presented as electronic links in the ISA.
3 4.3 ASSESSMENT APPROACH
4 4.3.1 Introduction
5 The NCEA-RTP is responsible for preparing the ISA. In each NAAQS review,
6 development of the science assessment begins with a "Call for Information" published in the
7 Federal Register. This notice announces EPA's initiation of activities in the preparation of the
8 ISA for the specific NAAQS review and invites the public to assist through the submission of
9 research studies in the identified subject areas. This and subsequent key components of the
10 process currently followed for the development of an ISA (i.e., the development process) are
11 presented in Figure 4-1 and are described in greater detail in the Preamble to the ISA. Section 1.2
12 above briefly describes how the ISA fits into the larger NAAQS review process.
13 Important aspects of the development of the ISA are described in the sections below, including
14 the approach for searching the literature and identifying relevant publications and informing
15 specific policy-relevant questions that are intended to guide the assessment. These
16 responsibilities are undertaken by expert authors of the ISA chapters which include EPA staff
17 with extensive knowledge in their respective fields and extramural scientists solicited by EPA for
18 their expertise in specific fields. The process for scientific and public review of drafts of the ISA
19 is described in section 4.5 below.
February 2014 4-3 Draft - Do Not Quote or Cite
-------
Literature Search and
Study Selection
Evaluation of Individual Stud1
After study selection, the quality of individual studies is evaluated by EPA or outside experts in the fields of
atmospheric science, exposure assessment, dosimetry, animal toxicology, controlled human exposure studies,
epidemiology, ecology and other welfare effects, considering the design, methods, conduct, and documentation of
each study. Strengths and limitations of individual studies that may affect the interpretation of the study are
considered.
Develop Initial Sections
Review and summarize new study results and
findings and conclusions from previous
assessments by category of outcome/effect and
by discipline, e.g., toxicological studies of lung
function.
Peer Input Consultation
Review of initial draft materials by scientists
from both outside and within EPA in public
meeting or public teleconference.
Evaluation, Synthesis and Integration of Evidence
Integrate evidence from scientific disciplines - for example, toxicological, controlled human exposure and
epidemiologic study findings for particular health outcome. Evaluate evidence for related groups of endpoints or
outcomes to draw conclusions regarding health or welfare effect categories, integrating health or welfare effects
evidence with information on mode of action and exposure assessment.
Development of Scientific Conclusions and Causal Determinations
Characterize weight of evidence and develop judgments regarding causality for health or welfare effect categories.
Develop conclusions regarding concentration- or dose-response relationships, potentially at-risk populations,
lifestages, or ecosystems.
3
4
Draft Integrated Science Assessment
Evaluation and integration of newly published studies
after each draft
Clean Air Scientific Advisory Committee
Independent review of draft documents for scientific
quality and sound implementation of causal
framework; anticipated review of two drafts of ISA in
public meetings.
Public Comments
Comments on draft ISA solicited by EPA
Final Integrated Science Assessment
Source: Modified from Figure II of the Preamble to the ISA (U.S. EPA, 2013b).
Figure 4-1. General process for development of Integrated Science Assessments.
February 2014
4-4
Draft - Do Not Quote or Cite
-------
1
2 4.3.2 Literature Search and Selection of Relevant Studies
3 The NCEA-RTP uses a structured approach to identify relevant studies for consideration
4 and inclusion in the ISA. A Federal Register notice is published to announce the initiation of a
5 review and request information, including relevant literature, from the public. In addition,
6 publications are identified by the EPA through a recursive multi-tiered literature search process
7 that includes extensive manual and computer-aided citation mining of computer databases on
8 specific topics in a variety of disciplines using as keywords terms such as oxides of nitrogen,
9 NOx, NC>2, NO, nitric acid, peroxyacetyl nitrate, and total reactive nitrogen. The search strategies
10 are designed a priori and iteratively modified to optimize identification of pertinent published
11 papers. Papers are identified for inclusion in several additional ways: specialized searches on
12 specific topics; relational searches that identify recent publications that have cited references
13 from previous assessments; identification of relevant literature by expert scientists;
14 recommendations from the public and CASAC during the call for information and external
15 review process; and review of citations in previous assessments. These search methods are used
16 to identify recent research published or accepted for publication since the 2008 ISA for Oxides
17 of Nitrogen, i.e., starting in January 2008 through approximately two months before the release
18 of the second external review draft of the ISA (target of July/August 2014, see Table 2-1).
19 Studies published after that date may also be included in the ISA if they provide new information
20 that impacts one or more key scientific issues. Studies published after the ISA cut-off date also
21 may be considered in subsequent phases of the NAAQS review, after assessing whether they
22 provide new information that impacts key scientific issues.
23 Once identified through the multipronged search strategy, studies are reviewed for
24 relevance. Relevant publications are epidemiological, lexicological, and controlled human
25 exposure studies that examine health effects in relation to exposure to oxides of nitrogen as well
26 as studies on sources, emissions, atmospheric chemistry, human exposure, dosimetry, and modes
27 of action of oxides of nitrogen. Relevant publications include studies and reports that have
28 undergone scientific peer review and have been published or accepted for publication. Some
29 publications are excluded as not being relevant based on screening of the title. Publications
30 considered for inclusion in the ISA after reading the title are documented in the HERO database.
31 From the group of considered references, studies and reports are selected for inclusion in
32 the ISA based on review of the abstract and full text. The selection process is based on the extent
33 to which the study is potentially informative and policy-relevant. Potentially policy-relevant and
34 informative studies include those that provide a basis for or describe the relationship between the
35 criteria pollutant and effects, in particular, those studies that offer innovation in method or design
36 and studies that reduce uncertainty on critical issues. Uncertainty can be addressed, for example,
February 2014 4-5 Draft - Do Not Quote or Cite
-------
1 by analyses of potential confounding or effect modification by copollutants or other factors,
2 analyses of concentration-response or dose-response relationships, or analyses related to time
3 between exposure and response. In keeping with the purpose to accurately reflect the latest
4 scientific knowledge, the ISA generally will emphasize studies published since the 2008 ISA.
5 However, evidence from previous studies will be included to integrate with results from recent
6 studies, and in some cases, characterize the key policy-relevant information in a particular
7 subject area or scientific discipline. Analyses conducted by the EPA using publicly available
8 data, for example, air quality and emissions data, also are considered for inclusion in the ISA.
9 The combination of approaches described above is intended to produce the comprehensive
10 collection of pertinent studies needed to address the key scientific issues that form the basis of
11 the ISA. References are cited in the ISA by a hyperlink to the HERO database and also are
12 compiled into reference lists.
13 4.3.3 Evaluation of Individual Study Quality
14 After selecting studies for inclusion, individual study quality is evaluated by considering
15 the design, methods, conduct, and documentation of each study but not considering whether the
16 study results are positive, negative, or null. This uniform approach aims to consider the strengths,
17 limitations, and possible roles of chance, confounding, and other biases that may affect the
18 interpretation of the results from individual studies. In assessing the scientific quality of studies,
19 the following parameters are considered:
20 • How clearly were the study design, study groups, methods, data, and results presented to
21 allow for study evaluation?
22 • To what extent are the air quality data, exposure, or dose metrics of adequate quality to
23 serve as credible exposure indicators?
24 • Were the study populations, subjects, or animal models adequately selected, and are they
25 sufficiently well defined to allow for meaningful comparisons between study or exposure
26 groups?
27 • Are the statistical analyses appropriate, properly performed, and properly interpreted?
28 Do the analytical methods provide adequate sensitivity and precision to support study
29 conclusions?
30 • Are likely covariates (i.e., potential confounding factors, modifying factors) adequately
31 controlled for or taken into account in the study design or statistical analyses?
32 • Are the health endpoint measurements meaningful, valid, and reliable?
33 Additional considerations specific to particular scientific disciplines are discussed below.
34 Atmospheric Science and Exposure Assessment
February 2014 4-6 Draft - Do Not Quote or Cite
-------
1 Atmospheric science and exposure assessment studies focus on measurement of,
2 chemistry, fate, and transport of, and exposure to ambient air pollution using quality-assured
3 field, experimental, and/or modeling techniques. The most informative measurement-based
4 studies will include detailed descriptive statistics for high-quality measurements made at varying
5 spatial and temporal scales. These studies will also include a clear and comprehensive
6 description of measurement techniques and quality control procedures used. Quality control
7 metrics (e.g., method detection limits) and quantitative relationships between and within
8 pollutant measurements (e.g., regression model coefficients, intercepts, and fit statistics) should
9 be provided when appropriate. Measurements including contrasting conditions for various time
10 periods (e.g., weekday/weekend, season), populations, geographic regions, land use types (e.g.,
11 urban/rural), and proximity to various source sectors are particularly useful. The most
12 informative modeling-based studies will incorporate appropriate chemistry, transport, dispersion,
13 and/or exposure modeling techniques with a clear and comprehensive description of model
14 science, evaluation procedures, and metrics.
15 Exposure measurement error, which refers to the uncertainty associated with the exposure
16 metrics used to represent exposure of an individual or population, can be an important
17 contributor to uncertainty in air pollution epidemiological study results. Exposure measurement
18 error can influence epidemiological associations observed between ambient pollutant
19 concentrations and health outcomes by biasing effect estimates toward or away from the null and
20 widening confidence intervals around those estimates (Zeger et al., 2000). Factors that could
21 influence exposure estimates include, but are not limited to, non-ambient sources of exposure,
22 topography of the natural and built environment, meteorology, air quality measurement
23 instrument errors, model uncertainties, time-activity patterns, and the infiltration of outdoor
24 pollutants into indoor environments. Additional information present in high-quality exposure
25 studies includes location and activity information from diaries, questionnaires, global positioning
26 system data, or other means, as well as information on commuting patterns.
27 Epidemiology
28 In evaluating quality of epidemiological studies, the EPA additionally considers whether a
29 given study: (1) presents quantitative information on associations of health effects with short- or
30 long-term exposures that represent ambient concentrations of oxides of nitrogen across various
31 microenvironments; (2) examines health effects of specific oxides of nitrogen; (3) assesses
32 oxides of nitrogen as a component of a complex mixture of air pollutants by considering
33 concentrations of copollutants, correlations of oxides of nitrogen with these copollutants,
34 potential copollutant interactions (e.g., synergistic effects of oxides of nitrogen with other
35 pollutants), potential copollutant confounding (e.g., bias of associations observed between oxides
February 2014 4-7 Draft - Do Not Quote or Cite
-------
1 of nitrogen and health endpoints by the effects of copollutants), and other methods to assess the
2 independent effect of oxides of nitrogen; (4) evaluates health endpoints not previously
3 extensively researched; (5) evaluates lifestages or populations that potentially are at increased
4 risk of health effects related to oxides of nitrogen; (6) examines other potential confounding
5 factors or effect modifiers (e.g., socioeconomic status [SES]); and/or (7) examines important
6 methodological issues (e.g., lag or time period between exposure and effects, model
7 specifications, thresholds, mortality displacement) related to the health effects of exposure to
8 oxides of nitrogen. Among epidemiological studies characterized as high quality by these
9 parameters, emphasis will be given to multicity studies that employ standard methodological
10 analyses for evaluating effects of oxides of nitrogen across cities, provide overall estimates for
11 effects by pooling information across cities, and examine consistency of results across cities. To
12 address specific issues relevant to standard setting in the U.S., such as regional heterogeneity in
13 effects, emphasis will be placed on studies that involve exposures and population characteristics
14 that are relevant to current U.S. populations (e.g., studies conducted in the U.S. or Canada).
15 Controlled Human Exposure and Animal Toxicology
16 Controlled human exposure and animal toxicological studies experimentally evaluate the
17 health effects of administered exposures in human volunteers and animal models under highly
18 controlled laboratory conditions. Controlled human exposure studies are also referred to as
19 human clinical studies. These experiments allow investigators to expose subjects or animal
20 models to known concentrations of oxides of nitrogen under carefully regulated environmental
21 conditions and/or activity levels. In addition to the general quality considerations discussed
22 previously, evaluation of controlled human exposure and animal toxicological studies includes
23 assessing the design and methodology of each study with focus on: (1) characterization of the
24 intake dose, dosing regimen (e.g., duration), and exposure route; (2) characterization of the
25 pollutant(s) (i.e., oxides of nitrogen species); (3) sample size and statistical power to detect
26 differences; and (4) control of other variables that could influence the occurrence of effects. The
27 evaluation of study design generally includes consideration of factors that minimize bias in
28 results such as randomization, blinding and allocation concealment of study subjects,
29 investigators, and research staff, and unexplained loss of animals or withdrawal/exclusion of
30 subjects. Additionally, studies must include appropriate control groups and exposures to allow
31 for accurate interpretation of results relative to exposure. Emphasis is placed on studies that
32 address concentration-dependent responses or time-course of responses. Also, with the
33 recognition that controlled human exposure studies typically are conducted in young adults and
34 healthy individuals, emphasis will be placed on studies that investigate potentially at-risk
35 lifestages or populations (e.g., with pre-existing disease). In addition, consideration will be given
February 2014 4-8 Draft - Do Not Quote or Cite
-------
1 to studies that investigate exposure to oxides of nitrogen separately and in combination with
2 other pollutants such as ozone, PM, and sulfur dioxide.
3 Controlled human exposure or animal toxicological studies involving exposures that
4 approximate expected human exposures in terms of concentration, duration, and route of
5 exposure are of particular interest. Relevant pollutant exposures are considered to be those
6 generally within two orders of magnitude of ambient concentrations measured across various
7 microenvironments. Studies using higher concentration exposures or doses will be considered to
8 the extent that they provide information relevant to understanding mode of action or
9 mechanisms, interspecies variation, or at-risk human lifestages and populations. In vitro studies
10 may be included if they provide mechanistic insight or support results demonstrated in vivo.
11 4.3.4 Integration of Evidence and Determination of Causality
12 As described in the Preamble to the ISA, the EPA uses a consistent and transparent basis
13 for the integration of scientific evidence and evaluation of the causal nature of air
14 pollution-related health effects in the ISA. The evidence evaluated from previous and recent
15 studies is integrated across scientific disciplines and related health effects to form causal
16 determinations. Evaluation of human health effects is informed by controlled human exposure,
17 epidemiological, and toxicological studies. Other information including mechanistic evidence,
18 toxicokinetics, and exposure assessment may be highlighted if it is relevant to the evaluation of
19 health effects and if it is of sufficient importance to affect the overall evaluation. The relative
20 importance of different sources of evidence to the conclusions varies by pollutant or assessment,
21 as does the availability of different sources of evidence for causality determination. In judgments
22 of causality, scientists will also evaluate uncertainty in the scientific evidence, considering issues
23 such as generalizing results from a small number of controlled human exposure subjects to the
24 larger population; quantitative extrapolations of observed pollutant-induced pathophysiological
25 alterations from laboratory animals to humans; confounding by co-exposure to other ambient
26 pollutants, meteorological factors, or other factors; the potential for effects to be due to exposure
27 to air pollution mixtures; and the influence of exposure measurement error on epidemiological
28 study findings.
29 The EPA uses a framework to provide a consistent and transparent basis for classifying
30 the weight of available evidence according to a five-level hierarchy: (1) causal relationship; (2)
31 likely to be a causal relationship; (3) suggestive of a causal relationship; (4) inadequate to infer a
32 causal relationship; and (5) not likely to be a causal relationship (U.S. EPA, 2013c, Table II). In
33 the framework, key considerations in drawing conclusions about causality include consistency of
34 findings for an endpoint across studies, coherence of the evidence across disciplines and across
35 related endpoints, and biological plausibility, including key events that inform modes of action.
February 2014 4-9 Draft - Do Not Quote or Cite
-------
1 Causal determinations are developed for major outcome categories (e.g., respiratory effects) or
2 more specific groups of related endpoints and for the range of exposure concentrations of oxides
3 of nitrogen that are representative of those across various ambient microenvironments. Findings
4 based on higher exposure concentrations may be considered if they inform biological plausibility
5 and potential modes of action. Causal determinations are based on the confidence in the body of
6 evidence, considering study design and quality and strengths and weaknesses in the overall
7 collection of studies across disciplines. In discussing the causal determination, the EPA
8 characterizes the evidence on which the judgment is based, including weight of evidence for
9 individual endpoints within the outcome category or group of related endpoints.
10 4.3.5 Quality Management
11 NCEA participates in the Agency-wide Quality Management System, which requires the
12 development of a Quality Management Plan (QMP). Implementation of the NCEA QMP ensures
13 that all data generated or used by NCEA scientists are "of the type and quality needed and
14 expected for their intended use" and that all information disseminated by NCEA adheres to a
15 high standard for quality including objectivity, utility, and integrity. Quality assurance (QA)
16 measures detailed in the QMP are being employed for the current primary NC>2 NAAQS review,
17 including the development of the ISA for the health criteria of oxides of nitrogen. The NCEA
18 QA staff is responsible for the review and approval of quality-related documentation. NCEA
19 scientists are responsible for the evaluation of all inputs to the ISA, including primary (new) and
20 secondary (existing) data, to ensure their quality is appropriate for their intended purpose. NCEA
21 adheres to Data Quality Objectives, which identify the most appropriate inputs to the science
22 assessment and provide QA instruction for researchers citing secondary information. The
23 approaches utilized to search the literature and criteria for study selection and evaluation were
24 detailed in the two preceding subsections. Generally, NCEA scientists rely on scientific
25 information found in peer-reviewed journal articles, books, and government reports. The ISA
26 also can include information that is integrated or reduced from multiple sources to create new
27 figures, tables, or summation, which is subject to rigorous quality assurance measures to ensure
28 their accuracy.
29 4.4 SPECIFIC ISSUES TO BE ADDRESSED IN THE ISA
30 The ISA for oxides of nitrogen will contain information relevant to considering whether
31 it is appropriate to retain or revise the current primary NC>2 standards. Decisions on the specific
32 content of the ISA will be guided by policy-relevant questions that frame the entire review of the
33 primary NC>2 NAAQS as outlined in section 3.2 above. These policy-relevant questions are
34 related to two overarching issues. The first issue is whether new evidence reinforces or calls into
35 question the evidence presented and evaluated in the last primary NO2 NAAQS review with
February 2014 4-10 Draft - Do Not Quote or Cite
-------
1 respect to factors such as the concentrations of oxides of nitrogen associated with health effects
2 and plausibility of health effects caused by exposure to oxides of nitrogen. The second issue is
3 whether uncertainties from the last review have been reduced and/or whether new uncertainties
4 have emerged. The ISA also will address a set of more specific policy-relevant questions related
5 to the available scientific evidence that stem from these issues. These questions were derived
6 from the last primary NC>2 NAAQS review, as well as from discussions of the scientific evidence
7 that occurred at the February/March 2012 kickoff workshop for the current review (77 FR 7149,
8 February 10, 2012); a CASAC consultation on the draft plan for development of the ISA (U.S.
9 EPA, 2013a; 78 FR 27234, May 9, 2013); and a public workshop that included review of initial
10 draft materials for the ISA (78 FR 27374, May 10, 2013). The specific questions to be addressed
11 in the ISA are listed below by topic area. In the ISA, these topic areas will be discussed in
12 separate chapters or sections.
13 Atmospheric Science and Ambient Concentrations
14 The ISA will present and evaluate data related to ambient concentrations of oxides of
15 nitrogen; sources leading to the presence of oxides of nitrogen in the atmosphere; and chemical
16 reactions that determine the formation, degradation, and lifetime of oxides of nitrogen in the
17 atmosphere. Key conclusions from the 2008 ISA were that motor vehicles and power plants are
18 the major U.S. sources of NOx emissions and that ambient concentrations of NC>2 display
19 heterogeneity across spatial and temporal scales, are higher near roadways, and are correlated
20 with ambient concentrations of several other traffic-related pollutants (U.S. EPA, 2008a, section
21 5.2.1). The formation and reactions of NC>2 are strongly influenced by volatile organic
22 compounds and 63. The relationships among these pollutants are critical to the understanding of
23 ambient NC>2 concentrations. In the current review, with regard to air quality and atmospheric
24 chemistry, specific policy-relevant questions that will be addressed include:
25 • What new information is available to inform our understanding of the atmospheric
26 chemistry of oxides of nitrogen? How does new information characterize the role of
27 atmospheric chemistry in determining relationships among oxides of nitrogen species?
28 What new information is available with respect to nitroaromatics and nitropyrenes, which
29 have shown toxic effects and thus, may be important in assessing health effects from
30 multipollutant exposures? How does the near-source environment (e.g., near major
31 highways or large combustion sources) influence chemistry of oxides of nitrogen?
32 • What new information exists regarding characterization of sources of ambient oxides of
33 nitrogen in both urban and rural environments? What are the relevant spatial and
34 temporal scales for considering ambient emissions of oxides of nitrogen? What new
35 information is available regarding existing and emerging sources of energy and impacts
36 on emissions of oxides of nitrogen?
37 • To what extent have new methods been developed to improve measurements of oxides of
38 nitrogen, particularly those that measure NC>2 directly? How have these new methods
February 2014 4-11 Draft - Do Not Quote or Cite
-------
1 reduced interference problems in measuring oxides of nitrogen? What limitations still
2 remain?
3 • Based on recent air quality and emissions data, what do we know about recent emissions
4 and resulting ambient concentrations of oxides of nitrogen? How have emissions and
5 concentrations of NOx and of NC>2 changed since the 2008 ISA? To what extent can new
6 data sources (e.g., satellites) or air quality analyses be used to improve the
7 characterization of ambient concentrations of oxides of nitrogen?
8 • What spatial and temporal patterns can be seen in ambient NC>2 and NOx concentrations?
9 In particular, what spatial and temporal patterns can be seen on a micro-scale near
10 sources including major roadways and combustion sources such as power plants and
11 biomass burning? What do ambient air quality characterizations (including examinations
12 of the influence of meteorological parameters) indicate regarding spatial patterns on
13 neighborhood, urban, regional, and national scales?
14 • Based on air quality and emissions data for oxides of nitrogen and atmospheric chemistry
15 models, what are likely background concentrations of oxides of nitrogen in the absence of
16 anthropogenic emissions?
17 • What new information is available to characterize the influence of meteorological
18 parameters on micro- to neighborhood-scale concentrations of oxides of nitrogen?
19 Human Exposure
20 The ISA will evaluate the factors that influence exposure to ambient oxides of nitrogen
21 and the measurement error and other uncertainties associated with extrapolation of ambient
22 concentrations to personal exposures to oxides of nitrogen of ambient origin, particularly in the
23 context of interpreting results from epidemiological studies. The evaluation will build upon the
24 discussion in the 2008 ISA, which concluded that measurement error associated with using
25 ambient NC>2 concentrations obtained from central site monitors as measures of exposure in
26 epidemiological studies tended to bias the magnitude of associations between ambient NO2 and
27 health effects toward the null (U.S. EPA, 2008a, section 5.2.2). Uncertainties related to exposure
28 measurement error differ by the exposure period of interest as most epidemiological studies of
29 short-term exposure (e.g., population-level studies using time-series analyses, field/panel studies)
30 rely on temporal variation in exposure while epidemiological studies of long-term exposure (e.g.,
31 longitudinal cohort studies) rely on spatial variability of exposure. In the current review, with
32 regard to exposure, specific policy-relevant questions that will be addressed include:
33 • How have modeling techniques such as sub-grid scale modeling within chemical
34 transport models, air quality dispersion models, and land use regression models been
35 advanced in recent years? What new information is available regarding modeled
36 estimates of spatially-resolved (at the micro-, middle-, and neighborhood scales) ambient
37 NC>2 and NOx concentrations used for exposure assessment?
38 • How have ambient models been merged with stochastic population exposure models
39 recently to improve estimates of exposure? What advancements have been made
February 2014 4-12 Draft - Do Not Quote or Cite
-------
1 regarding validation of stochastic population exposure models and their ability to
2 estimate source attribution for exposures to NO2 or NOx?
3 • What are the relationships between oxides of nitrogen measured at stationary monitoring
4 sites and personal exposure? What evidence is available regarding these relationships in
5 environments near roads or other sources?
6 • What new information exists about the relationship between NO, NC>2, and NOx
7 concentrations and indicators of near-source pollution including distance to sources (e.g.,
8 major roadways) and source activity levels (e.g, traffic counts)?
9 • What studies are available to examine the relationship between near-road oxides of
10 nitrogen, on-road oxides of nitrogen, and in-vehicle exposures to oxides of nitrogen?
11 Given the concern over short-term exposures at or less than one hour in duration, are the
12 directly emitted NO2/NOx ratios sufficiently high such that on-road NO2 exposure is a
13 significant component of total NO2 exposure?
14 • To what extent is information available characterizing how well the current near-road
15 NO2 monitoring sites represent exposures to populations living near major roads?
16 • What new information exists regarding indoor exposures to oxides of nitrogen, including
17 those generated indoors and those that infiltrate from outdoors? What new information is
18 available regarding how oxides of nitrogen are generated indoors?
19 • What new information exists regarding characterization of error in exposure assessment
20 of oxides of nitrogen and how it influences personal-ambient exposure relationships?
21 • What information is available regarding differences in exposure patterns for oxides of
22 nitrogen and personal-ambient exposure relationships among various lifestages and
23 populations?
24 • What are the implications for epidemiology for assessing health effects of exposures to
25 oxides of nitrogen when there are instrumentation errors, such as measurements of
26 ambient concentrations being subject to interferences from other nitrogen compounds?
27 • What new information exists regarding oxides of nitrogen measurements in a
28 multipollutant context? To what extent do NO2 measurements serve as surrogates of
29 exposure to other gaseous pollutants (e.g., carbon monoxide, nitrous acid), particle phase
30 pollutants (e.g., ultrafine particles, black carbon, organic carbon, transition metals)
31 generated by traffic or other combustion sources, or a mixture of traffic-related
32 pollutants?
33 • What new information is available regarding the interaction of oxides of nitrogen with
34 organic compounds emitted from home cleaning and deodorizing products to form
35 organic nitrates indoors that may influence human exposure to NO2?
36 Dosimetry and Modes of Action
37 The ISA will evaluate literature focusing on dosimetry and modes of action that may
38 underlie the health outcomes associated with exposure to NO2 and/or NO. These topic areas will
39 be evaluated using both human and animal data. The 2008 ISA concluded that ambient-relevant
40 concentrations of inhaled NO2 are consumed by constituents of the epithelial lining fluid of the
February 2014 4-13 Draft - Do Not Quote or Cite
-------
1 respiratory tract, including antioxidants, to form secondary reaction products (U.S. EPA, 2008a,
2 section 2.6). These secondary reaction products initiate a cascade of events that are thought to be
3 responsible for health effects observed in association with NC>2 exposure. Additionally, findings
4 of MVinduced changes in airway responsiveness, airway inflammation, and lung host defenses
5 were described as key mechanistic support for NO2-related respiratory effects such as respiratory
6 symptoms and ED visits. In the current review, specific policy-relevant questions related to
7 dosimetry and modes of action that will be addressed include:
8 • What new information is available to inform our understanding of the potential biological
9 mechanisms underlying responses to NC>2 and/or NO exposures at or near
10 environmentally-relevant concentrations, with a focus on response pathway(s) and
11 exposure-dose-response relationships?
12 • What information is available to characterize intra- and inter-individual variability in
13 biological responses following exposure to NC>2 and/or NO?
14 • What are the effects of host factors such as lifestage, sex, pre-existing disease, genetic
15 background, and physical activity on the uptake of NO2 and/or NO and cellular and tissue
16 responses as well as biological mechanisms that may underlie health effects associated
17 with exposure to oxides of nitrogen?
18 • What information is available to discern the relative contributions to internal NO2 and/or
19 NO of: (1) ambient exposures to NO2 and/or NO; (2) dietary consumption of nitrite and
20 nitrate which undergo transformation to NO; and (3) endogenous formation of NO2
21 and/or NO?
22 • What NO2 and/or NO reaction products, including oxides of nitrogen metabolites, can be
23 found in the cells, tissues, or fluids of the respiratory tract and in the systemic circulation
24 that may serve as markers of NO2 and/or NO exposure and effect?
25 • What biological processes, from the molecular to whole organ level, can be qualitatively
26 compared across species?
27 • To what extent can the inhalation dosimetry of NO2 and/or NO be extrapolated between
28 species, qualitatively or quantitatively?
29 • Do interactions between inhaled NO2 and/or NO and other inhaled pollutants influence
30 the mechanisms underlying the toxic potential of NO2 and/or NO?
31 Health Effects
32 In the 2008 ISA, the health effects evidence for oxides of nitrogen was largely indexed by
33 studies of NO2 with the bulk of the evidence provided by short-term exposure studies evaluating
34 respiratory effects. The EPA will build on this assessment and evaluate the newly available
35 literature related to respiratory, cardiovascular, reproductive, and developmental health effects,
36 mortality, and cancer associated with exposure to oxides of nitrogen. Depending on data
37 availability, other health effects also may be evaluated, for example, those related to the central
38 nervous system or gastrointestinal system. Health effects that occur following both short- and
February 2014 4-14 Draft - Do Not Quote or Cite
-------
1 long-term exposures as examined in epidemiological, controlled human exposure, and animal
2 toxicological studies will be evaluated. Efforts will be directed at identifying the concentrations
3 at which effects are observed, including those in potential at-risk lifestages and populations, and
4 assessing the role of oxides of nitrogen within the broader mixture of ambient pollutants. The
5 discussion of health effects also will be integrated with relevant information on dosimetry and
6 modes of action. In the current review, with regard to consideration of health effects associated
7 with short-and long-term exposure to oxides of nitrogen, specific policy-relevant questions that
8 will be addressed include the following:
9 Short-Term Exposure
10 • What do controlled human exposure, animal toxicological, and epidemiological studies
11 indicate regarding the relationship between short-term exposures to oxides of nitrogen
12 and health effects of concern (e.g., respiratory effects, cardiovascular effects, premature
13 mortality)?
14 • How does evidence for health effects associated with oxides of nitrogen compare among
15 healthy individuals, those with pre-existing disease states (e.g., people with asthma or
16 cardiovascular disease), particular lifestages, or groups characterized by other factors that
17 potentially modify risk (e.g., genetic, nutritional)?
18 • At what ambient concentrations of oxides of nitrogen are associations with the various
19 health effects observed in epidemiological studies most well characterized?
20 • To what extent does the scientific evidence support the occurrence of health effects
21 associated with short-term (i.e., minutes up to 1 month) exposure to oxides of nitrogen at
22 ambient concentrations that are lower than those previously demonstrated? If so, what
23 uncertainties are related to these associations and are the health effects in question
24 important from a public health perspective?
25 • How do results of recent studies or new interpretations of previous findings expand our
26 understanding of the relationship between short-term exposure to oxides of nitrogen and
27 airway hyperresponsiveness or other lung function changes, inflammation, host defense
28 against infection disease, and respiratory symptoms?
29 • What are the effects of oxides of nitrogen exposure on cardiovascular health in humans
30 (e-g-, inflammation, heart rate variability, arrhythmias, vasomotor function, risk of
31 myocardial infarction)?
32 • How do results from recent population-level time-series studies expand our
33 understanding of relationships between exposure to oxides of nitrogen and mortality (all
34 nonaccidental-cause, respiratory, cardiovascular), hospital admissions, or emergency
35 department visits?
36 • To what extent does short-term exposure to oxides of nitrogen contribute to health effects
37 beyond the respiratory and cardiovascular systems?
38 • What is the extent of coherence of findings for changes in lung function, airway
39 hyperresponsiveness, heart rate variability, and vasomotor function and effects such as
February 2014 4-15 Draft - Do Not Quote or Cite
-------
1 hospital admissions, emergency department visits, and mortality? What other biomarkers
2 of early effect may be used in the assessment of health effects?
3 • What evidence is available regarding the shape of concentration-response relationships
4 between short-term exposure to oxides of nitrogen and various health endpoints?
5 o Is there evidence to support the identification of a discernible threshold below which
6 health effects will not occur?
7 • What evidence is available regarding the nature of health effects from interactions
8 between oxides of nitrogen and other ambient air pollutants in comparison to health
9 effects following exposure to oxides of nitrogen alone?
10 o To what extent are the observed epidemiological health effect associations
11 attributable to ambient oxides of nitrogen, another ambient pollutant, or to the
12 pollutant mixtures that oxides of nitrogen may be representing? What information is
13 available specifically from studies conducted near roads or other sources? To what
14 extent do findings from experimental studies provide biological plausibility for the
15 effects observed in epidemiologic studies?
16 • To what extent does information across scientific disciplines on the pattern of exposure to
17 oxides of nitrogen (e.g., peak, repeated peak, average) provide understanding of the time
18 course for changes in health effects? What information is available on time-activity
19 patterns of study subjects such as time spent outdoors or activity levels that can aid in the
20 understanding of nature of exposure or dosimetry of ambient oxides of nitrogen that are
21 associated with health effects?
22 • To what extent do data across scientific disciplines provide information on health effects
23 related to specific oxides of nitrogen (e.g., NC>2, NO) or averaging times of exposure to
24 oxides of nitrogen that are relevant to the 1-hour standard? What data exist comparing
25 associations of health effects among various short-term metrics of exposure to oxides of
26 nitrogen (e.g., 1-hour versus 24-hour)?
27 Long-Term Exposure
28 • How do results of recent studies expand our understanding of the relationships between
29 long-term exposure to oxides of nitrogen and chronic respiratory effects manifested as
30 morphological changes, a reduction in baseline lung function, or a reduction in lung
31 function growth?
32 • To what extent does long-term exposure to oxides of nitrogen promote exacerbation and
33 development of asthma or other chronic lung diseases, cardiovascular diseases, and other
34 conditions?
35 • What is the relationship between long-term exposure to oxides of nitrogen and all-cause
36 mortality and cardiovascular and respiratory mortality?
37 • To what extent does long-term exposure to oxides of nitrogen contribute to other health
38 effects or changes in molecular and cellular processes, e.g., cognitive, behavioral,
39 reproductive, developmental, cancer or epigenetic effects?
40 • How does evidence for health effects associated with oxides of nitrogen compare among
41 healthy individuals, those with pre-existing disease states (e.g., people with asthma or
February 2014 4-16 Draft - Do Not Quote or Cite
-------
1 cardiovascular disease), particular lifestages, or other factors that potentially modify risk
2 (e-g-, genetic, nutritional)?
3 • At what ambient concentrations of oxides of nitrogen are associations observed in
4 epidemiological studies most well characterized?
5 • To what extent does the scientific evidence support the occurrence of health effects from
6 long-term (i.e., more than 1 month to years) exposure to oxides of nitrogen at ambient
7 concentrations that are lower than those previously demonstrated? If so, what
8 uncertainties are related to these associations and are the health effects in question
9 important from a public health perspective?
10 • What evidence is available regarding the shape of concentration-response relationships
11 between long-term exposure to oxides of nitrogen and health effects?
12 o Is there evidence to support the identification of a discernible threshold below which
13 health effects will not occur?
14 • What evidence is available regarding the nature of health effects from interactions
15 between long-term exposure to oxides of nitrogen and other ambient air pollutants in
16 comparison to health effects following exposure to oxides of nitrogen alone?
17 o To what extent are the observed epidemiological health effect associations
18 attributable to ambient oxides of nitrogen, another ambient pollutant, or to the
19 pollutant mixtures that oxides of nitrogen may be representing? What information is
20 available specifically from studies conducted in populations living near roads or other
21 sources? To what extent do findings from experimental studies provide biological
22 plausibility for the effects observed in epidemiological studies?
23 • What information is available regarding the effect of long-term, low-concentration
24 exposure to oxides of nitrogen on an individual's sensitivity to short-term but higher
25 concentration exposures?
26 • What evidence is available regarding health effects related to long-term exposure
27 windows other than annual or lifetime average (e.g., preconception, pregnancy average,
28 pregnancy trimester average)? What data are available comparing associations of health
29 effects among various long-term oxides of nitrogen exposure metrics (e.g., annual,
30 seasonal, pregnancy average)?
31 Causality
32 In the 2008 ISA, the EPA concluded that the findings of epidemiological, controlled
33 human exposure, and animal toxicological studies collectively provided evidence "sufficient to
34 infer a likely causal relationship " between short-term NC>2 exposures and respiratory effects
35 (U.S. EPA, 2008a, sections 3.1.7 and 5.3.2.1). In looking at a broader range of health effects
36 associated with short- or long-term exposures to oxides of nitrogen, the 2008 ISA concluded
37 there was evidence "suggestive but not sufficient to infer a causal relationship" between short-
38 term NC>2 exposures and premature mortality and between long-term NC>2 exposures and
39 respiratory effects (U.S. EPA, 2008a, sections 5.3.2.3 and 5.3.2.4). Furthermore, the 2008 ISA
February 2014 4-17 Draft - Do Not Quote or Cite
-------
1 concluded that the scientific evidence was "inadequate to infer the presence or absence of a
2 causal relationship" between short-term NC>2 exposures and cardiovascular effects as well as
3 between long-term NC>2 exposures and cardiovascular effects, reproductive and developmental
4 effects, premature mortality, and cancer (U.S. EPA, 2008a, sections 5.3.2.2, 5.3.2.5, and 5.3.2.6).
5 The causal determinations, based on the causal framework and integration of available
6 evidence (see sections 4.3.3 and 4.3.4 above), are presented with a summary of the available
7 evidence at the end of the sections for each health effect outcome category and in the integrative
8 synthesis chapter at the beginning of the ISA. In the current review, specific policy-relevant
9 questions related to the causality determinations that will be addressed include:
10 • Does the evidence base from recent studies contain new information to support or re-
11 evaluate the causal determinations made for relationships between NC>2 exposure and
12 various health effects in the 2008 ISA?
13 • What information is available to support a rationale for forming causal determinations for
14 other oxides of nitrogen (e.g., NO, NOx)?
15 • What information is available regarding the health impacts of a decrease in ambient
16 concentrations of oxides of nitrogen?
17 Uncertainties/Limitations
18 The causal determinations described above for the relationships between NC>2 exposure
19 and health effects were informed by uncertainties and limitations in the evidence including the
20 possibility that pollutants other than oxides of nitrogen in broad ambient mixture were
21 responsible for health effects observed in association with NC>2 and/or limited information from
22 experimental studies to provide biological plausibility. In each of the health effects sections and
23 the integrative synthesis chapter, the ISA will evaluate uncertainties and limitations in the
24 scientific data, particularly in relation to observed epidemiological findings and their coherence
25 with human exposure and toxicological studies in terms of observed effects and biological
26 pathways. These uncertainties also will inform causal determinations. In the current review,
27 specific policy-relevant questions related to the evaluation of uncertainties/limitations that will
28 be addressed include:
29 • To what extent are the observed health effect associations attributable specifically to
30 ambient oxides of nitrogen versus other pollutants contained in the broader air pollution
31 mixture? For example, the ISA will consider the possibility that ambient concentrations
32 of NO2 serve not only as an indicator for oxides of nitrogen but as a surrogate for
33 exposure to other vehicle exhaust gaseous and particulate pollutants.
34 o What information about the independent health effects of exposure to oxides of
35 nitrogen can be synthesized from the various lines of available evidence, for example,
36 copollutant models, associations with other traffic-related pollutants, analysis of
37 indoor NC>2, comparisons of results from locations with varying pollutant mixtures,
38 studies of traffic proximity or intensity, and experimental studies?
February 2014 4-18 Draft - Do Not Quote or Cite
-------
1 o How does confounding by co-exposure to other ambient pollutants (e.g., ozone,
2 particulate matter, sulfur dioxide, carbon monoxide) or meteorological factors
3 influence relationships observed between health effects and both short- and long-term
4 exposures to oxides of nitrogen? To what extent do other factors serve as potential
5 confounding factors in epidemiological studies (e.g., demographic and lifestyle
6 attributes, other exposures such as noise)?
7 • What information is available to assess the influence of exposure measurement error on
8 uncertainty in epidemiological study results?
9 o How can the influence of exposure measurement error be assessed through the
10 examination of various study designs, study populations and locations, exposure
11 assessment methods, and analytical models?
12 o What information is available regarding the time-activity patterns of study subjects
13 including time spent outdoors, spatial distribution of study subjects and ambient
14 monitors, exposure assessment methods, potential interference in the measurement of
15 NC>2 from other oxides of nitrogen in low traffic, downwind locations?
16 At-risk Lifestages and Populations
17 The 2008 ISA discussed persons with pre-existing respiratory disease (e.g., people with
18 asthma), children, and older adults as populations and lifestages potentially at greater risk of
19 NO2-related health effects (U.S. EPA, 2008a, section 5.3.2.8). Since completion of the 2008 ISA,
20 the EPA has developed a framework to provide a consistent and transparent basis for classifying
21 the weight of evidence about at-risk lifestages or populations according to one of four levels:
22 adequate evidence, suggestive evidence, inadequate evidence, and evidence of no effect (U.S.
23 EPA, 2013c, Table 8-1). In this framework, key considerations in drawing such conclusions
24 include consistency of findings for a factor within a discipline and coherence of the evidence
25 across disciplines. The ISA will evaluate an array of factors that may characterize potential at-
26 risk lifestages or populations: intrinsic factors (biological factors such as age or genetic variants),
27 acquired factors (e.g., pre-existing disease), extrinsic factors (nonbiological factors such as diet,
28 lower socioeconomic status), and/or factors affecting dose or exposure (sex, age, outdoor activity
29 or work, lower socioeconomic status, physical activity). The ISA will not distinguish among risk
30 due to intrinsic, extrinsic, or other types of factors. The various factors listed above (e.g., age)
31 may influence risk through various mechanisms, including increasing exposure, dose, or
32 increasing biological effect for a given dose, and some factors may contribute to risk via multiple
33 mechanisms. In the current review, with regard to at-risk lifestages and populations, specific
34 policy-relevant questions that will be addressed include:
35 • Based on evidence integrated across studies and disciplines that examine factors that may
36 increase exposure to oxides of nitrogen and/or risk of health effects related to exposure to
37 oxides of nitrogen, what conclusions can be drawn about the presence of at-risk lifestages
38 (e.g., the developing fetus, children, older adults) or populations?
February 2014 4-19 Draft - Do Not Quote or Cite
-------
1 • Which disciplines contribute information about particular at-risk lifestages and
2 populations, and to what extent does limited or lack of information from specific
3 disciplines produce uncertainty in conclusions about at-risk lifestages and populations?
4 • How does new information compare with that evaluated in the 2008 ISA regarding
5 people with pre-existing respiratory disease, genetic variants, or low SES as potential at-
6 risk populations and children or older adults as potential at-risk lifestages?
7 • What information is available that provides insight as to whether a potential at-risk
8 lifestage or population has higher exposure or dose of oxides of nitrogen and/or has a
9 greater biological response to a given exposure?
10 • What is the extent of the coherence of evidence regarding potential at-risk lifestages or
11 populations for both short- and long-term exposures to oxides of nitrogen?
12 • What quantitative information is available to characterize the magnitude of greater
13 biological response or risk of health effects associated with exposure to oxides of
14 nitrogen in a particular at-risk lifestage or population?
15 Public Health Impact
16 The integrative synthesis chapter at the beginning of the ISA will present concepts that
17 integrate evidence on health effects and consequent public health significance to aid in the
18 assessment of the public health implications of exposure to short- and long-term exposure to
19 oxides of nitrogen. The discussion will include evaluation of the adversity of the health effects
20 potentially associated with exposure to oxides of nitrogen. Development of these concepts may
21 include, as appropriate, an estimation of the sizes of potential at-risk lifestages and populations
22 and discussion of the public health significance of the magnitudes of change in health outcomes
23 characterized to result from ambient air exposure to oxides of nitrogen. Further, to the extent that
24 evidence is available, the integrative synthesis chapter of the ISA will discuss what evidence is
25 available regarding interrelationships among risk factors (e.g., young age and lower SES, old age
26 and pre-existing cardiovascular disease) in a particular lifestage or population that may add to the
27 understanding of the public health impact of exposure to oxides of nitrogen.
28 4.5 SCIENTIFIC AND PUBLIC REVIEW OF THE ISA
29 Drafts of the ISA will be made available for review by the CAS AC and the public as indicated in
30 Figures 2-1 and 4-1 above. Availability of draft documents will be announced in the Federal
31 Register. The CASAC will review the draft ISA documents and discuss its comments in public
32 meetings that will be announced in the Federal Register. The EPA will take into account
33 comments, advice, and recommendations received from the CASAC and from the public in
34 revising the draft ISA documents. The EPA has established a public docket for the development
February 2014 4-20 Draft - Do Not Quote or Cite
-------
1 of the ISA.43 After appropriate revision based on comments received from the CAS AC and the
2 public, the final document will be made available on an EPA website. A notice announcing the
3 availability of the final ISA will be published in the Federal Register.
43The ISA docket can be accessed at www.regulations.gov using Docket ID number EPA-HQ-ORD-2013-
0232.
February 2014 4-21 Draft - Do Not Quote or Cite
-------
i 5 QUANTITATIVE RISK AND EXPOSURE
2 ASSESSMENT
3 Within the context of NAAQS reviews, a quantitative risk and exposure assessment
4 (REA) is designed to estimate human exposure and health risks associated with existing and
5 potential alternative standards. The appropriate scope of any REA will be informed by the
6 availability of scientific information from the ISA as well as air quality information and
7 information on data and models that may help to address important uncertainties or provide
8 additional insights beyond those provided by previous REAs. As a result, the first step in the
9 REA planning process is an assessment of the appropriate scope of the REA, which includes a
10 determination of whether a distinct REA document is needed. As part of this planning process,
11 we evaluate the 2008 REA in the context of the extent to which important uncertainties may be
12 addressed by new information available since the previous review and the extent to which new
13 information may change results of the 2008 REA in important ways or may allow for additional
14 analyses that can address important gaps in our understanding of the exposures and risks
15 as soci ated with NC>2.
16 This phase of the NAAQS review begins with the preparation of a REA Planning
17 Document and considers the extent to which newly available scientific evidence and
18 tools/methodologies provide support for conducting quantitative risk and exposure assessments.
19 To the extent warranted, the scope and methods for components of exposure/risk assessments
20 will be described. As outlined in Table 2-1 above, the EPA plans to issue this REA Planning
21 Document in September 2014. This document will be the subject of a CASAC consultation and
22 will be made available to the public for review and comment. CASAC advice and public
23 comments on this draft IRP will be considered in developing the REA Planning Document. If
24 warranted, one or more drafts of an REA will then be prepared and released for CASAC review
25 and public comment prior to completion of a final REA.
26 The information newly available in this review will be considered in light of the
27 comprehensive, complex and resource-intensive quantitative assessments of human exposure and
28 health risks documented in the 2008 REA as discussed in section 6.1 below. As discussed in
29 section 6.2 below, the REA Planning Document will consider the available scientific evidence,
30 tools, and methodologies in light of areas of uncertainty identified in the 2008 REA and the
31 potential for new analyses to provide notably different exposure and risk estimates, with lower
32 associated uncertainty. The timeline for collection of ambient NC>2 measurement data within
33 near-road environments under the recently revised monitoring requirements is recognized as an
34 important consideration for the REA Planning Document. CASAC advice and comments from
February 2014 5-1 Draft - Do Not Quote or Cite
-------
1 the public on this draft IRP, as well as the availability of resources, will inform development of
2 the REA Planning Document.
3 5.1 OVERVIEW OF RISK AND EXPOSURE ASSESSMENT FROM
4 PRIOR REVIEW
5 In the last review, as summarized in section 3.1 above, the EPA designed and developed
6 three approaches to estimating exposures and health risks associated with a number of ambient
7 air quality scenarios (i.e., recent air quality unadjusted, air quality adjusted to simulate just
8 meeting the then-existing annual standard (i.e., annual average of 53 ppb), and air quality
9 adjusted to simulate just meeting several potential alternative daily maximum 1-hour standards).
10 Briefly, in the first approach ambient NC>2 concentrations (measured and modeled) were
11 compared to 1-hour health effect benchmark levels derived from the controlled human exposure
12 literature. In the second approach, modeled estimates of human exposures in an urban study area
13 were compared to these same health effect benchmark levels. In the third approach,
14 concentration-response relationships from an epidemiological study were used to estimate health
15 impacts associated with ambient NO2 concentrations in an urban study area. An overview of
16 these approaches used and results generated in characterizing health risks is provided below.
17 5.1.1 Ambient Air Quality Characterization
18 In the first approach, we compared 1-hour ambient NC>2 concentrations (1995 to 2006)
19 with short-term health effect benchmark concentrations of 100, 150, 200, 250, and 300 ppb
20 NC>244 in order to identify the number of days a particular benchmark concentration was
21 exceeded per monitor per year. All U.S. monitoring sites where NC>2 data have been collected
22 were included in this analysis and, as such, the results generated were considered a broad
23 characterization of national air quality and potential human exposures that might be associated
24 with these concentrations.45 The available NC>2 air quality for 18 MS A/CMS A named study
25 areas46 and two aggregate study areas were separated into two six-year groups; one contained
44 The 1-hour NO2 health effect benchmark levels were based on NO2 exposure concentrations associated
with increased airway responsiveness in asthmatics and determined from 1 to 2 hour duration controlled human
exposure studies. These benchmark values were used for the evaluation of both the NO2 air quality concentrations
and estimated NO2 exposures.
45 After applying a 75% data completeness criterion the final analytical data base included 627 monitors
collecting ambient concentrations for 4,177 site-years of data (a valid monitoring day had >18 hourly measurements;
monitors included in the analysis had >75% valid monitoring days in a year).
46 An initial pool of monitors was subset from the total set of monitors based on their belonging to specific
CMSA/MSAs. Then we selected study areas having annual mean NO2 concentrations occurring at a minimum of
one monitor at or above 25.7 ppb (i.e., the 90th percentile concentrations across all study areas and site-years) and/or
had at least one reported 1-hour NO2 level greater than or equal to 200 ppb (i.e., the lowest level of the potential
health effect benchmarks indicated by the ISA at the time these study areas were identified for investigation in the
2008 REA). All remaining not included in this collection of named study areas were aggregated into either one of
two groups: all other CMSA/MSA or all other non-CMSA/MSA.
February 2014 5-2 Draft - Do Not Quote or Cite
-------
1 data from years 1995-2000, representing an historical data set; the other contained the
2 monitoring years 2001-2006, representing recent ambient monitoring (U.S. EPA, 2008b, section
3 7.2.2).
4 Each of these monitoring year-groups and study areas were evaluated considering the
5 ambient NC>2 concentrations as they were reported and representing the conditions at that time
6 (termed in that assessment "as is"). This served as the first air quality scenario. Further, within
7 each year-group and study area we categorized the monitors using their sited distance from a
8 major road: at or within 20 meters (< 20 m), between 20 and 100 meters (> 20 m to < 100 m),
9 and at least 100 meters from a major road (> 100 m).47 These ambient monitor data were
10 categorized in this manner to account for the potential influence of vehicle emissions on NC>2
11 concentrations measured at the monitors within close proximity to roadways (U.S. EPA, 2008b,
12 section 7.2.3). There was potential for different concentration levels measured at each of these
13 locations (i.e., near-road monitors versus those sited away from roadways) and thus potentially
14 different exposure concentrations experienced by those persons spending time in these locations.
15 Then, for each ambient monitor, we summed the number of days per year that monitor recorded
16 a daily maximum 1-hour concentration at or above the health effect benchmark levels and
17 summarized this metric for each study area, year-group, and roadway-distance group, using
18 descriptive statistics (e.g., means, maximums) (U.S. EPA, 2008b, section 7.2.5).
19 A second air quality scenario used the as is ambient monitoring data obtained from
20 monitors sited >100 m from a major road combined with an on-road concentration adjustment
21 factor to estimate on-road7V02 concentrations for each of the year-groups and study areas. This
22 scenario was developed by recognizing that motor vehicles are important emission sources of
23 NOx and NC>2 and that people spend time inside vehicles while travelling on roadways. At that
24 time, a strong relationship had been reported between NO2 concentrations measured on roadways
25 and NC>2 concentrations measured at increasing distance from the road, generally in the form of a
26 first-order exponential decay (e.g., Cape et al., 2004). We derived an empirical distribution of
27 on-road adjustment factors using data from published studies that reported on- and near-road
28 NO2 concentrations and NC>2 concentrations occurring at greater distances of a major road (> 100
AR
29 m). Then, we probabilistically applied these on-road adjustment factors to NO2 concentrations
30 reported at monitors sited > 100 meters from a major road (and generally assumed to be at a
31 background non-roadway influenced concentration) to approximate on-road NC>2 concentrations
47 Major road distances to each monitor were generally determined using a Tele-Atlas roads database in a
GIS application. Major road types were defined as: l=primary limited access or interstate, 2=primary US and State
highways, 3=Secondary State and County, 4=freeway ramp, 5=other ramps. Note only the monitors falling within
the 18 identified study areas had estimated distances to major roads, all other monitors (either characterized as 'other
CMSA/MSA' or 'all other non-CMSA/MSA') were assumed to be > 100 m from a major road.
48 See Table 7-10 of the 2008 REA for the specific values of distributions that were used and Appendix A,
section 8 for the studies used and the derivation methodology (U.S. EPA, 2008b).
February 2014 5-3 Draft - Do Not Quote or Cite
-------
1 for each study area and monitor-year. As described above for the area-wide ambient monitor
2 data, we counted the number of days per year an estimated daily maximum 1 -hour on-road
3 concentration exceeded the benchmark levels and summarized these data using simple
4 descriptive statistics (U.S. EPA, 2008b, section 7.2.4).
5 Additional summaries of benchmark exceedances were generated for the above two air
6 quality scenarios (i.e., using the complete set of ambient monitoring concentrations at each of the
7 three roadway distance categories and the simulated on-road NC>2 concentrations), though they
8 differed in that the ambient concentrations were first adjusted to just meet the then-existing
9 annual standard or alternative daily maximum 1-hour standards. Because the annual average
10 2001 to 2006 ambient concentrations49 were below the level of the existing annual standard (i.e.,
11 < 53 ppb) as well as most of the potential daily maximum 1-hour standards being evaluated,
12 ambient concentrations were primarily adjusted upwards to reflect these additional air quality
13 scenarios. A simple proportional adjustment approach was selected to simulate concentrations to
14 meet a particular standard level, an approach supported by within-monitor comparisons of low
15 and high NC>2 concentration years that largely demonstrated characteristics of a proportional
16 relationship.50 We note also that the as is air quality could be characterized in all study areas as
17 falling within the evaluated alternative standard levels of 50 and 100 ppb (either a 98th or 99th
18 percentile daily maximum 1-hour concentration); thus simulating these particular air quality
19 scenarios required the smallest proportional adjustment. Simulations of just meeting an
20 alternative standard level of 50 ppb required a downward adjustment (U.S. EPA, 2008b, section
21 6.3.1). That said, the number of benchmark exceedances for as is air quality scenarios in each
22 study area also fell within the range of that estimated considering the 50 and 100 ppb daily
23 maximum 1-hour alternative standard scenarios.
24 5.1.1.1 Key Observations
25 • Ambient monitoring NOi concentrations: When considering any of the air quality
26 scenarios, NC>2 concentrations and estimated number of benchmark exceedances are
27 typically higher for monitors that are within 20 meters (m) of a major roadway than when
28 monitors are sited farther from a major roadway (i.e., between 20 m and 100 m or >100
29 m from a major road) (U.S. EPA, 2008b, section 7.3.1). As expected, fewer health effect
30 benchmark exceedances were estimated to occur at highest health effect benchmark
49 Only the 2001 to 2006 ambient concentrations were used to evaluate the existing and alternative standard
levels, the historical air quality data set with measurements from 1995 to 2000 was not used in this part of the
assessment.
50 Linear regressions were performed using the daily maximum 1-hour concentration distributions of each a
low and high concentration year measured at the same ambient monitor and evaluated for model linearity and
presence of statistically significant regression intercepts. Statistically significant linear regression slopes and model
R2 values strongly supported features of linearity .On a few occasions however, the presence of statistically
significant regression intercepts and deviation from linearity at upper percentile concentrations tends to obfuscate a
conclusion of proportionality existing at all monitors (Rizzo, 2008; U.S. EPA, 2008b, section 7.4.5).
February 2014 5-4 Draft - Do Not Quote or Cite
-------
1 levels (e.g., 300 ppb) when compared with the lowest health effect benchmark level (100
2 ppb). While results were generated for health effect benchmark levels ranging from 100
3 to 300 ppb in 50 ppb increments, the discussion in the 2008 REA focused only on the
4 health effect benchmark levels of 100, 200, and 300 ppb.
5 o 100 ppb health effect benchmark lev el: When air quality is adjusted to simulate
6 just meeting the then-existing annual (i.e., 53 ppb, annual average), most study areas
7 were estimated to have, on average, 50 days or more per year with daily maximum 1-
8 hour ambient NC>2 concentrations > 100 ppb, while about 1/3 were estimated to have
9 100 days or more per year with daily maximum 1-hour ambient NC>2 concentrations >
10 100 ppb. When air quality is adjusted to simulate just meeting alternative daily
11 maximum 1-hour standard levels of 50 and 100 ppb (98* or 99* percentile in a year),
12 far fewer days per year were estimated to have, on average, daily maximum 1-hour
13 ambient concentrations > 100 ppb (i.e., <10 days per year, on average) than compared
14 with just meeting higher alternative daily maximum 1-hour standard levels of 150 and
15 200 ppb (generally tens to hundreds of days per year with daily maximum 1 -hour
16 ambient NO2 concentrations > 100 ppb) (U.S. EPA, 2008b, section 7.5).
17 o 200 ppb health effect benchmark level: When air quality was adjusted to simulate
18 just meeting the then-existing annual standard, only two study areas were estimated,
19 on average, to have 10 or more days per year with daily maximum 1-hour ambient
20 NO2 concentrations > 200 ppb. When air quality was adjusted to simulate just
21 meeting alternative daily maximum 1-hour standard levels of 50 and 100 ppb (98* or
22 99* percentile in a year), only four study areas were estimated, on average, to have at
23 least one day per year with daily maximum 1-hour ambient NC>2 concentrations > 200
24 ppb.
25 o 300 ppb health effect benchmark level: When air quality was adjusted to simulate
26 just meeting the then-existing annual standard, only five study areas were estimated,
27 on average, to experience any days with daily maximum 1-hour ambient NC>2
28 concentrations at central site monitors > 300 ppb, and none of those study areas were
29 estimated to experience more than 2 such days per year, on average. When air quality
30 was adjusted to simulate just meeting alternative daily maximum 1-hour standard
31 levels of 50 and 100 ppb (98* or 99 percentile in a year), only three study areas were
32 estimated, on average, to have at least one day per year with daily maximum 1-hour
33 ambient NC>2 concentrations > 300 ppb.
34 • Simulated on-road NOi concentrations: Simulated on-road annual average NO2
35 concentrations, estimated using an on-road adjustment factor, were on average, 80
36 percent higher than the respective ambient concentrations measured at distances >100 m
37 from a road; thus, there were a greater number of days per year where estimated daily
38 maximum 1-hour concentrations on roads exceeded the health effect benchmark levels.
39 o 100 ppb health effect benchmark level: In the majority of study areas, exceedances
40 of the 100 ppb health effect benchmark level were estimated to occur on roadways for
41 most days of the year when air quality was adjusted to simulate just meeting the then-
42 existing standard. Most study areas were estimated, on average, to have between 100
43 and 300 days per year with daily maximum 1-hour on-road NC>2 concentrations > 100
44 ppb. The mean number of days per year where estimated on-road concentrations
February 2014 5-5 Draft - Do Not Quote or Cite
-------
1 were > 100 ppb was always greater than that estimated using concentrations at
2 ambient monitoring locations (e.g., up to 18 days per year for a standard level of 50
3 ppb, 257 for a standard level of 100 ppb, 343 for a standard level of 150 ppb, and 351
4 for a standard level of 200 ppb, based on the 98th percentile of daily maximum 1-hour
5 concentrations, averaged over three years).
6 o 200 and 300 ppb health effect benchmark levels: Even considering the higher
7 health effect benchmark levels, most study areas were estimated, on average, to
8 exceed these benchmark levels on roadways when air quality was adjusted to simulate
9 just meeting the then-existing annual standard. Most study areas were estimated, on
10 average, to have between 25 and 100 days per year with daily maximum 1-hour on-
11 road NC>2 concentrations > 200 ppb. All study areas evaluated, except one, were
12 estimated to have on-road NO2 concentrations > 300 ppb. Four of these study areas
13 were estimated, on average, to experience an average of greater than 20 days per year
14 with on-road NC>2 concentrations > 300 ppb.
15 5.1.1.2 Key Uncertainties
16 An advantage of this approach to estimating potential health risk is its relative simplicity;
17 however, there were a number of important uncertainties identified (US EPA, 2008b, section
18 7.4). One of the most important uncertainties overall regards the spatial representation of the
19 ambient monitors, hence part of the reasoning to revise ambient monitoring networks at the
20 conclusion of the last NC>2 NAAQS review to include monitoring near roadways. To overcome
21 this lack of near-roadway measurement data in the 2008 REA and as briefly described above we
22 developed a simple statistical model using measurement data reported in a limited number of
23 peer-reviewed studies to estimate our on-road NO2 concentrations. In doing so, this statistical
24 model was characterized as having moderate or greater uncertainty in estimating on-road NC>2
25 concentrations, both in potentially under- and over-estimating the number of exceedances of
26 health effect benchmark levels. In addition, the proportional adjustment applied to ambient air
27 quality measurements to simulate just meeting the existing and alternative standards was
28 characterized as another important uncertainty, particularly when adjusting concentrations
29 upwards to meet or approach concentrations reflecting the existing annual standard. Further, the
30 selection of health effect benchmark levels used to characterize risk was based on controlled
31 human exposure studies that used mild asthmatics. In the absence of information regarding the
32 potential health response of persons characterized as having moderate or severe asthma, we
33 characterized the health effect benchmark level selection as an important uncertainty.
34 5.1.2 Human Exposure Assessment
35 In the second approach, we used an inhalation exposure model to generate more realistic
36 estimates of personal NC>2 exposure concentrations and compared those estimates of personal
37 exposure to the health effect benchmark levels. The EPA's Air Pollutants Exposure model
38 (APEX) probabilistically estimated individual exposures considering the time people spend in
February 2014 5-6 Draft - Do Not Quote or Cite
-------
1 different microenvironments and the variable NC>2 concentrations that occur within these
2 microenvironments across time, space, and microenvironment type, including estimation of on-
3 and near-roadway exposure concentrations (U.S. EPA, 2008b, section 8.2). The EPA's air
4 dispersion model (AERMOD) was used to estimate hourly NC>2 concentrations occurring at a
5 census tract level and at roadway receptors, considering emissions from stationary, area-wide,
6 and on-road mobile sources (U.S. EPA 2008b, section 8.4). This approach to assessing
7 exposures at that time was more resource intensive than using ambient measurements as a
8 surrogate for exposure as discussed in section 6.1.1 above; therefore, only one specific study area
9 was selected for analysis (four counties comprising the core Atlanta, GA metropolitan statistical
10 area, or MSA). Although the geographic scope of this analysis is restricted, the approach
11 provides realistic estimates of NC>2 exposures, particularly those exposures associated with
12 important emission sources of NOx and NC>2, and serves to complement the results of the broad
13 NC>2 air quality characterization.
14 For the characterization of risks in the exposure modeling analysis, staff used the same
15 range of short-term potential health effect benchmark levels described above in the air quality
16 characterization summarized in section 6.1.1 (i.e., 1-hour NC>2 concentrations of 100, 150, 200,
17 250, and 300 ppb) and considered the same air quality scenarios (recent "as is" ambient
18 concentrations and ambient concentrations adjusted to just meet the existing and potential
19 alternative standards, though using only years 2001-2003).51 Asthmatic school-age children (5
20 to 17 years of age) and all asthmatics (0 to 99 years of age) were considered the most important
21 exposure study groups in this assessment based on their having potentially increased health risk
22 to NC>2 exposure concentrations (U.S. EPA, 2008a, section 4.3). Exposure estimates for
23 asthmatic school-age children were segregated from that estimated for the broader asthmatic
24 population group because of their potential to have greater participation rate and time engaged in
25 outdoor activities, thus possibly increasing their NC>2 exposures. When personal exposures for
26 either exposure study group were simulated, the output of the analysis was an estimate of the
27 number of individuals at risk for experiencing daily maximum 1-hour levels of NC>2
28 concentrations of ambient origin that exceeded particular health effect benchmark levels. An
29 advantage of using potential health effect benchmark levels based on evidence from controlled
30 human exposure studies to characterize health risks in this exposure modeling approach was that
31 the effects observed in these studies clearly resulted from NC>2 exposure. This was in contrast to
32 using health effects associated with ambient NO2 concentrations in epidemiological studies (as
51 This three-year period was selected to bound the most recent year of NOX emissions data available (i.e.,
2002) and used to model ambient concentrations at the time of our exposure assessment was conducted (U.S. EPA,
2008b)
February 2014 5-7 Draft - Do Not Quote or Cite
-------
1 discussed in the third approach described in section 5.1.3 below), which can also be associated
2 with pollutants that co-occur with NC>2 in the ambient air.
3 5.1.2.1 Key Observations
4 • Estimated daily maximum 1-hour exposures at or above potential health effect
5 benchmark levels using APEX were largely a function of roadway-related exposure
6 concentrations (greater than 99 percent). Of these exposures, approximately 70 percent
7 resulted from in-vehicle exposures, with the remainder associated with outdoor near-road
8 exposures. Overall, when simulating air quality that just meets the then-existing annual
9 standard, virtually all asthmatics in Atlanta were estimated to experience six or more
10 daily maximum 1-hour exposures per year to NO2 concentrations above the highest
11 health effect benchmark level evaluated (i.e., >300 ppb), indicating the extremely limited
12 ability of the then-existing annual standard at that time to protect against daily maximum
13 1-hour exposures at or above any of the selected health effect benchmark levels (U.S.
14 EPA, 2008b, section 8.10).
15 o 100 ppb health effect benchmark level: For all air quality scenarios considered,
16 more than 90 percent of all asthmatics in Atlanta were estimated to be exposed to
17 concentration > 100 ppb at least one time per year. Of the daily maximum 1-hour
18 alternative standard levels evaluated, 50 ppb was the only standard level estimated to
19 reduce repeat daily maximum 1-hour NO2 exposures above 100 ppb compared to
20 recent air quality concentrations.
21 o 200 ppb health effect benchmark level: Of all the air quality scenarios considered,
22 only the 98 and 99* percentile daily maximum 1-hour alternative standards set at 50
23 ppb were estimated to reduce the percent of asthmatics exposed at least one time per
24 year to concentrations > 200 ppb (by approximately 40 to 50 percent) relative to
25 recent air quality concentrations.
26 o 300 ppb health effect benchmark level: Of all the air quality scenarios considered,
27 only alternative standard levels set at 50 ppb or 100 ppb were estimated to reduce the
28 percent of asthmatics exposed at least one time per year to concentrations > 300 ppb
29 (by approximately 80 percent and 15 percent, respectively) relative to recent air
30 quality concentrations.
31 5.1.2.2 Key Uncertainties
32 The same important uncertainties exist for the exposure modeling results as described in
33 section 6.1.1.2 above for the air quality characterization where similar approaches were used
34 (i.e., proportional air quality adjustment approach and selection of health effect benchmark
35 levels). One important uncertainty identified as specific to the exposure assessment was the
36 AERMOD estimated concentrations used to represent the air quality surface across the Atlanta
37 study area (U.S. EPA, 2008b, section 8.12). An evaluation with limited ambient monitor
38 measurement data suggested a potential bias in overestimating ambient concentrations,
39 potentially associated with uncertainty in mobile source emissions and/or diurnal profiles used as
40 input (among other sources of uncertainty) (U.S. EPA, 2008b, section 8.4.8). Given the few
February 2014 5-8 Draft - Do Not Quote or Cite
-------
1 monitors available and overall confidence in the AERMOD system and other input data, it was
2 difficult to reasonably justify adjusting all estimated concentrations across the entire 4-county
3 modeling domain based on the differing concentrations observed at the few monitor locations,
4 thus they were used without adjusting for this observed difference. An additional uncertainty,
5 though not specifically identified as an exposure uncertainty in the prior assessment, was the
6 similar factors approach used to adjust 1-hour AERMOD ambient concentrations to estimate on-
7 and near-roadways concentrations. While AERMOD estimated 1 -hour NO2 concentrations
8 occurring on roadway link-based receptors, these on-road concentrations could not be used
9 directly as an input to APEX based on its existing configuration. Thus an on-road adjustment
10 factor was developed from the AERMOD estimated on-road and census tract level
11 concentrations (U.S. EPA, 2008b, section 8.7.2.5). These new on-road adjustment factor
12 distributions used by APEX, along with the number of estimated on-road peak concentrations,
13 were compared with those used for the air quality characterization (U.S. EPA, 2008b, section
14 8.4.8.3). The two similar, though independently developed, concentration adjustment
15 approaches were found to be comparable across a wide range of estimated values, though they
CO
16 diverge at the upper percentiles of the distribution. And finally, in a limited set of targeted
17 exposure analyses, exposures were also modeled considering indoor source emissions (U.S.
18 EPA, 2008b, section 8.7.2.1).53 The characterization of indoor source emissions of NO2 and
19 estimated air exchange rates used to simulate indoor microenvironments were considered an
20 important uncertainty.
21 5.1.3 Epidemiological-based Human Health Risk Assessment
22 In the third approach, respiratory-related hospital emergency department (ED) visits were
23 estimated as a function of short-term ambient NO2 concentrations measured at a fixed-site
24 monitor representing ambient air quality for an urban area (U.S. EPA, 2008b, chapter 9). This
25 health endpoint was selected because the 2008 ISA reported several epidemiological studies with
26 observed positive associations between ambient NO2 concentrations and ED visits and
27 hospitalizations for all respiratory diseases and asthma (U.S. EPA, 2008a, section 3.1.6). In this
52 See Figure 8-8 and Table 8-7 of the 2008 REA (U.S. EPA, 2008b, pp!75 to 176). While the divergence
between the two on-road concentration estimations was noteworthy, particularly at the upper percentiles of the
distribution, important differences in the data used to develop the two approaches were likely a highly influential
contributing factor. For example, the on-road adjustment factor used in the exposure modeling approach used 1 -
hour concentrations, while the on-road adjustment factor used in air quality characterization approach was
developed from measurement data time averaged over 7 to 14 days. It is likely that the latter approach smoothes the
distribution of possible concentration ratios by using the long-term time-averaged concentrations, though we believe
it is reasonable to still highlight this issue here as an important uncertainty in the exposure assessment conducted for
the last review.
53 While potentially important in understanding health effects and the total exposure/health risk from NO2,
exposures resultant from indoor sources of NO2 have limited relevance in understanding health risk associated with
ambient concentrations.
February 2014 5-9 Draft - Do Not Quote or Cite
-------
1 type of risk estimation approach, concentration-response functions were based on findings from
2 NC>2 epidemiological studies that relied on fixed-site, population-oriented, ambient monitors as a
3 surrogate for actual ambient NC>2 exposures and were used to estimate the impact of daily
4 maximum 1-hour ambient NC>2 concentrations, as measured at a single fixed-site monitor, on ED
5 visits (U.S. EPA, 2008b, section 9.3). By focusing on a different health endpoint from the first
6 two approaches described above, this epidemiological-based approach provided additional
7 perspective on the potential public health impacts resultant from NC>2 exposures.
8 Because of the limited number of epidemiological studies reporting concentration-
9 response (C-R) functions and the limited availability of other data needed for a quantitative risk
10 assessment (e.g., baseline incidence rates), this type of health risk assessment was only
11 conducted in one study area (Atlanta, GA) using C-R functions extracted from one study
12 (Tolbert et al., 2007). The same general air quality scenarios described in section 5.1.1 above
13 were evaluated (i.e., ambient concentrations as is and those adjusted to just meeting the then-
14 existing annual and potential alternative daily maximum 1-hour standard levels), using ambient
15 air quality data from 2005 to 2007, the most recent ambient monitoring data available at the time
16 the assessment was conducted and similar to the years of air quality used in the epidemiological
17 study in determining C-R functions (1993 through 2004) (Tolbert et al., 2007).
18 5.1.3.1 Key Observations
19 • Health risks associated with just meeting the existing annual standard: Central estimates
20 of annual NC>2-related respiratory ED visits associated with air quality adjusted upward to
21 simulate just meeting the then-existing annual standard (based on 2006 to 2007 air
22 quality data) range from 8.1 to 9.0 percent of total incidence (or 9,800 to 10,900 NO2-
23 related incidences per year) based on single-pollutant models and from 1.7 to 7.7 percent
24 (or 3,100 to 9,400 NC>2-related incidences per year) based on co-pollutant models.
25 • Health risks associated with just meeting alternative daily maximum 1-hour standards:
26 Central estimates of annual NCVrelated respiratory ED visits associated with air quality
27 adjusted to simulate just meeting a 100 ppb, 1-h daily maximum, 98th percentile standard
28 (based on 2005 to 2007 air quality data) ranged from 3.9 to 4.3 percent of total incidence
29 based on single-pollutant models and from 0.8 to 3.7 percent based on co-pollutant
30 models. Central estimates of annual NCVrelated respiratory ED visits associated with air
31 quality adjusted to simulate just meeting a 50 ppb, 1-h daily maximum, 98th percentile
32 standard (based on 2005 to 2007 air quality data) ranged from 2.0 to 2.2 percent based
33 on single-pollutant models and from 0.4 to 1.9 percent based on co-pollutant models.
34 5.1.3.2 Key Uncertainties
35 A few of the same important uncertainties exist for the health risk modeling results as
36 described in sections 5.1.1.2 and 5.1.2.2 above for the air quality characterization and exposure
37 modeling assessments where similar approaches were used (i.e., uncertainties related to the
38 proportional air quality adjustment approach and the spatial representativeness of air quality
February 2014 5-10 Draft - Do Not Quote or Cite
-------
1 data, in general). In addition, two uncertainties unique to the epidemiological-based health risk
2 assessment approach recognized as important included 1) the risk model specification (i.e.,
3 overall form for concentration-response functions and the presence or not of a threshold) and 2)
4 the adequacy of the ambient NO2 monitors serving as a surrogate for population exposure (U.S.
5 EPA, 2008b, section 9.6).
6 5.2 CONSIDERATION OF QUANTITATIVE ASSESSMENTS FOR THIS
7 REVIEW
8 This discussion is focused particularly on considering the extent to which newly available
9 scientific evidence and tools/methodologies are available to inform our understanding of the key
10 areas of uncertainty identified in the 2008 REA as discussed in sections 5.1.1.2, 5.1.2.2 and
11 5.1.3.2 above. As outlined in Table 2-1 above, the EPA plans to release an REA Planning
12 Document for consultation with CAS AC and for public comments in September 2014 that will
13 consider the extent to which new quantitative risk and exposure assessments would be
14 appropriate to conduct in the current review. CASAC review and public comments on this draft
15 IRP will be considered in developing the REA Planning Document.
16 Some key areas being considered by staff, including types of data, methodologies and
17 tools, are identified and summarized below. Building upon each of the three approaches used to
18 estimate exposure or health risk in the previous review, we summarize the potential areas where
19 additional information, if available, would provide reasonable substance to address key
20 uncertainties identified in the previous review. We then discuss the potential utility and impacts
21 of this new information to improve upon the assessments performed in the prior review.
22 5.2.1 Air Quality Characterization
23 Table 5-1 summarizes the potentially important uncertainties where additional
24 information, if available, would provide reasonable substance to the discussion of improving the
25 air quality characterization performed in the prior review (U.S. EPA, 2008b, section 7.4). The
26 major uncertainties identified in section 5.1 above based on the 2008 REA were related to 1)
27 ambient monitoring representativeness, 2) the approach used to estimate on-road NO2
28 concentrations, 3) the approach used to estimate the existing and alternative air quality standard
29 scenarios, and 4) the selection of health effect benchmark levels.
30 5.2.2 Human Exposure Assessment
31 In addition to some of the uncertainties identified and described above in section 5.2.1,
32 three additional uncertainties were identified as specific to the exposure assessment conducted
33 for the 2008 REA. The major uncertainties identified in section 5.1.2.2 above that warrant
34 additional discussion here include 1) the use of unadjusted AERMOD estimated ambient NC>2
February 2014 5-11 Draft - Do Not Quote or Cite
-------
1 concentrations as input to APEX, 2) the factors approach used to estimate in-vehicle and near-
2 road NC>2 concentrations, and 3) the limited input data used to estimate the contribution of a
3 single source emission (indoor gas stoves) to a simulated person's total NC>2 exposure (Table 5-
4 2).
February 2014 5-12 Draft - Do Not Quote or Cite
-------
1 Table 5-1. Information (data, methods, models, etc.) identified as potentially important and/or newly available to inform the
2 air quality characterization for the current review.
Major
Uncertainty or
Limitation
Uncertainty/Limitation Remaining from 2008 REA
Sub-group
Description
Consideration of Potential Utility and Impact of
Information Newly Available in This Review
Could Have on Assessment
Near-Road
Ambient
Monitoring Data
There is a general lack of existing ambient monitoring data
near-roads in most U.S. urban areas. This, combined with the
potential for short-term high NO2 concentrations occurring in
these locations creates a significant uncertainty regarding
how often on- and near-road NO2 concentrations may exceed
exposure levels of concern. This uncertainty served as a
driver for revising ambient monitor siting.
Ambient Monitor
Spatial and
Temporal
Representativeness
The near-road monitoring network is to be developed
in three phases: the first set of 52 monitors to begin
measuring near-road NO2 concentrations starting
January 1, 2014, followed by the addition of 22
monitors by January 1, 2015, and phase III adding 52
monitoring sites (see section 5.2 of this IRP). Given
this schedule, it is possible that there will be a lag in
the availability of the newest near-road ambient
monitor concentration data due to quality assurance
reviews, potentially delaying the utilization of this
new and important data in this NO2 review.
Long-term
Exposures
The annual average standard of 53 ppb was retained in the
last review due to some evidence suggesting NO2
concentrations had a causal relationship with respiratory-
related health effects. Newly identified for this review
would be the consideration of how long-term, though
spatially variable, ambient NO2 concentrations could
adversely affect health.
Model and data fusion techniques could improve the
estimation of the spatial and temporal distribution of
NO2 concentrations across urban areas (not simply
area-wide and on-road distinctions), generally
accounting for increased emission sources and other
influential factors within a defined spatial sector.
There could be utility in estimating screening level
long-term NO2 concentrations if there were a newly
defined long-term health effect benchmark level of
interest.
Approach Used to
Estimate On-Road
NO2
Concentrations
Exponential
Relationship Used
to Characterize
Concentration
Decline with
Increasing
Distance from
Roads
Based on a literature review of studies that measured both
near- and away from road NO2 concentrations conducted by
OAQPS staff at the time of the last review and our analysis
of the NO2 concentration decline with distance from a
roadway, an exponential relationship was used to derive our
on-road adjustment factors. Variability in the form of the
relationship could result in the derivation of different factors
and potentially influencing estimated on-road NO2
concentration levels, though of course is dependent on the
form and parameters describing the relationship.
Distribution of
Two distributions of NO2 on-road adjustment factors were
We could use air quality models, e.g., AERMOD
dispersion model, to characterize near-source
gradients (major roadways and combustion sources).
In addition, we could evaluate alternative
relationships (e.g., linear, biphasic, etc.), better
characterize the distributions of on-road adjustment
factors, and consider other factors used to define
ambient monitors if there are studies newly available
(modeling and/or measurement) that indicate
alternative relationships exist outside of the range
already considered in the 2008 REA.
February 2014
5-13
Draft - Do Not Quote or Cite
-------
Major
Uncertainty or
Limitation
Uncertainty/Limitation Remaining from 2008 REA
Sub-group
On-road
Adjustment
Factors
Effect of Using
Longer-term
(weekly) Data
versus Hourly
Data to Estimate
On-road
Adjustment
Factors
Definition of
monitors
minimally
influenced by
roadway
emissions to use
in the calculation
of on-road
concentrations
Description
derived and used in their empirical form, one for summer
months ranging from 1.5 to 3.7 (median 1.9), the other non-
summer months ranging from 1.2 to 2.5 (median 1.75).
Data used to derive on-road adjustment factors were all from
data collected over 1-week or longer term measurements. It
is possible that there is variability in the derived relationship
over different time-averaging periods and concentration
levels. Thus there is uncertainty in application of the time-
averaged derived distribution to accurately estimate
variability in hourly NO2 concentrations.
We used any ambient monitor that was > 100 meters from a
major road to estimate on-road concentrations. At that time,
this distance alone was considered a reasonable criterion
given the literature reviewed and patterns noted in the then-
existing ambient monitoring data. As stated in the prior
REA section 7.4.6, it is possible that some of the ambient
monitors used to estimate on-road concentrations could have
been influenced by emissions from a non-road NOx source.
Consideration of Potential Utility and Impact of
Information Newly Available in This Review
Could Have on Assessment
Approach Used to
Simulate Just
Meeting Potential
Air Quality
Standard Scenarios
N/A
Six study areas (Los Angeles, Atlanta, New York,
Philadelphia, Denver, and Chicago) were used to evaluate
existence of a proportional relationship between high
concentration and low concentration years. The proportional
adjustment factors derived from the area design monitor was
similarly applied to adjust all ambient monitors within a
given study area. Deviation from proportionality (where
exists) at any monitor could result in either over or under-
estimation of concentrations.
New, adjusted ambient air quality data sets could be
developed if there are studies newly available that
indicate alternative approaches to adjusting air quality
exist that would generate demonstrably different data
sets outside of those already considered in the 2008
REA. This could include additional analysis of
ambient monitor data trends and air quality model
based approaches.
Selection of Health
Effect Benchmark
Levels
N/A
A generally common and important uncertainty in controlled
human exposure studies is the limited number of study
subjects as well as limits to the type of pre-existing health
conditions subjects may have, particularly if the health
condition affords the subject with heightened effects
New estimates of benchmark exceedances could be
developed if there are studies newly available that
indicate alternative benchmark levels exist outside of
the range already considered in the 2008 REA. This
would also apply where any new health endpoints are
February 2014
5-14
Draft - Do Not Quote or Cite
-------
Major
Uncertainty or
Limitation
Uncertainty /Limitation Remaining from 2008 REA
Sub-group
Description
sensitivity to the pollutant exposure. Further, there is a lack
of exposure data from study groups at potentially sensitive
lifestages (e.g., pregnant women, children).
Consideration of Potential Utility and Impact of
Information Newly Available in This Review
Could Have on Assessment
identified beyond those associated with respiratory
effects due to short-term exposures. Further,
additional analysis of existing human exposure study
data sets and an improved characterization of
adversity could be possible using newly identified
approaches or information.
February 2014
5-15
Draft - Do Not Quote or Cite
-------
1 Table 5-2. Information (data, methods, models, etc.) identified as potentially important and/or newly available to inform the
2 exposure assessment for the current review.
Major Uncertainty
or Limitation
Uncertainty/Limitation Remaining from 2008 REA
Description
Consideration of Potential Utility and Impact of
Information Newly Available in This Review
Could Have on Assessment
Use of Unadjusted
AERMOD
Estimated
Concentrations as
Input to APEX
AERMOD was used to reasonably represent spatial variability in ambient NO2
concentrations across an urban area where using limited monitor data alone cannot.
Even though comparisons made with limited ambient monitoring data suggest
AERMOD estimated ambient concentrations may have been systematically over-
estimated, concentrations were not adjusted for this potential upward bias.
We could further evaluate existing data generated
for the 2008 REA (both the modeled exposures
and ambient concentrations) to approximate the
potential impact of developing a newly modeled
alternative ambient concentration data set. In
addition, we could use AERMOD, with recent
improvements by EPA, to improve the quality and
characterization of the ambient concentration data
set for input to APEX.
Approach Used to
Estimate In-Vehicle
and Near-Road
Microenvironmental
(ME)
Concentrations in
APEX
In-vehicle and near-road NO2 concentrations were estimated using a similar
concentration adjustment approach described above for the air quality
characterization, only differing in that the relationship between on-road and away-
from-road receptor concentrations were estimated using AERMOD and that
penetration/decay inside motor vehicles was accounted for by APEX. While the
distribution of adjustment factors was stratified by time-of-day (11PM-6AM, 6AM-
7PM, 7PM-11PM) and seasons (summer and not-summer) (U.S. EPA, 2008b, Table
B-42), it is possible that use of a factors approach and randomly sampling from
distributions occasionally leads to a mismatching of on-road adjustment factors and
the away from road census block concentrations, leading to either over or under
estimated on-road NO2 concentrations. In addition, near-road NO2 concentrations
were considered at the same level as the estimated on-road concentrations without
additional adjustment for their occurring at a given distance from the road.
APEX has been recently modified to allow for a
time series of on-road concentrations, and can be
stratified by a geographic identifier, as an input to
estimating in-vehicle exposure concentrations.
Thus, APEX is capable of utilizing the AERMOD
on-road concentrations themselves rather than
using a factors-based approach. In the absence of
having a time-series of on-road concentrations for
potential study areas of interest, it is possible that
new factors or concentration distributions could be
developed and used to estimate near-road
microenvironmental concentrations, where newly
published data are identified.
Limited Input Data
Used to Estimate
Contribution of
Indoor Source
Emissions of NO2 to
Total Exposures
For a few APEX simulations, we estimated exposures associated with ambient
concentrations along with those associated with a single indoor emission source (gas
stoves). The estimation was based on limited input data readily available to
represent variability in the source emissions, population prevalence of gas stoves,
frequency of source use per cooking event and times of occurrence, and indoor
removal rates.
The role and relevance of understanding the
contribution of indoor source emissions to
exposures when setting ambient air quality
standards could be further evaluated.
February 2014
5-16
Draft - Do Not Quote or Cite
-------
1 5.2.3 Controlled Exposure-based Human Health Risk Assessment
2 One question to be raised for this review is whether or not there are newly identified
3 controlled human exposure studies on top of what was previously considered in the 2008 REA
4 that substantially expand our understanding of respiratory-related (or other) adverse health
5 effects. If new studies are identified and adequate data are available to develop exposure
6 concentration response relationships, these functions could be combined with NC>2 exposure
7 concentrations (e.g., output from a population-based exposure model) and thus, warrant
8 conducting a quantitative risk assessment based on the controlled human exposure evidence. In
9 addition, having new human exposure study data would not necessarily preclude a comparison of
10 earlier approaches (e.g., meta-analyses) used to develop the health effect benchmarks with any
11 newly identified or alternative approaches identified in this review.
12 5.2.4 Epidemiological-based Human Health Risk Assessment
13 In addition to a few relevant uncertainties identified above in section 5.2.1 above, two
14 additional uncertainties were identified as unique to the epidemiological-based health risk
15 conducted for the 2008 REA. The major uncertainties identified in section 5.1.3.2 above that
16 warrant additional discussion here include 1) the selection of the C-R function and 2) the ability
17 of the ambient NO2 monitors to serve as a surrogate for population exposure. The risk
18 assessment conducted in the last review focused on one health endpoint (i.e., respiratory-related
19 ED visits) in one urban study area (Atlanta). An important issue in this review is whether or not
20 additional information is available to consider conducting a quantitative risk assessment that
21 would include additional health effect endpoints and/or additional urban study areas. The EPA's
22 decisions regarding the conduct of and associated scope of an REA for this review can be
23 informed by the ISA causality determinations,54 in addition to the availability of appropriate data
24 for quantitative analyses (e.g., availability of C-R functions, baseline incidence data, etc.).
25 General criteria to be evaluated in identifying potential candidate studies to inform a quantitative
26 risk assessment include the following:
27 • the study was a published, peer-reviewed study that had been evaluated in the ISA for the
28 pollutant of interest and judged adequate by the EPA staff for purposes of inclusion in the
29 risk assessment based on that evaluation;
30 • it directly measured, rather than estimated, the pollutant of interest on a reasonable
31 proportion of the days in the study;
32 • it preferably included both single- and co-pollutant models.
54 The strength of the ISA causal determinations serves as a preliminary screen in identifying health
endpoints to consider in conducting our health assessments; historically we have considered those health endpoints
having either 'causal' or 'likely to be causal' determinations.
February 2014 5-17 Draft - Do Not Quote or Cite
-------
1 Table 5-3 summarizes the potentially important uncertainties where additional
2 information, if available, would provide reasonable substance to the discussion of improving the
3 epidemiological-based human health risk assessment performed in the prior review.
4 Table 5-3. Information (data, methods, models, etc.) identified as potentially important
5 and/or newly available to inform the epidemiological-based risk assessment for
6 the current review.
Major
Uncertainty or
Limitation
Uncertainty/Limitation Remaining
From 2008 REA
Description
Consideration of Potential Utility and Impact
that Information Newly Available in this Review
Could Have on Assessment
Form of
Concentration -
Response
Functions
In the epidemiological-based health
risk assessment, a single short-term
NO2 epidemiological study was used
to estimate respiratory-related health
risk (ED visits) in Atlanta (Tolbert et
al., 2007). A log-linear form
assuming no threshold was selected,
and both single and co-pollutant
models were employed. An
additional study by Ito et al. (2007)
conducted in New York City was also
identified but due to time and
resource constraints, was not included
in the 2008 REA.
It is possible that new epidemiological-based health
risk assessments in additional study areas could be
performed using new or alternative C-R functions
and potentially assuming alternative model
specifications if such published studies are available.
Adequacy of the
Ambient NO2
Monitors to
Serve as a
Surrogate for
Population
Exposure
For the epidemiological-based health
risk assessment developed from data
reported in Tolbert et al. (2007),
concentrations from a single ambient
monitor were used to represent area-
wide exposures.
While this is a common approach used in these types
of assessments, it is possible that the effect of using
a single monitor could be further evaluated where
new studies have employed additional or alternative
monitors in estimating health risk.
February 2014
5-18
Draft - Do Not Quote or Cite
-------
1 5.3 SCIENTIFIC AND PUBLIC REVIEW
2 The REA Planning Document will be distributed to the CASAC for their consideration
3 and provided to the public for review and comment. The document will be the subject of a
4 consultation with the CASAC at a public meeting or teleconference that will be announced in the
5 Federal Register.
6 If, upon consideration of CASAC recommendations and public comments, the EPA
7 concludes that development of a new REA, or updating or expanding the last assessment, is
8 warranted, staff will take into account comments received from CASAC and the public in
9 designing and conducting the assessment. In such a case, staff would prepare at least one draft
10 of the assessment for CASAC review and public comment. Review would be conducted by
11 CASAC and discussed at a public meeting that would be announced in the Federal Register.
12 Based on past practice by CASAC, the EPA expects that key advice and recommendations for
13 revision of the document would be summarized in a letter to the EPA Administrator. In revising
14 the draft REA document, the EPA would take into account any such recommendations, and also
15 consider comments received from the public, at the meeting itself and any written comments
16 received. A final document would then be made available on an EPA website, with its public
17 availability announced in the Federal Register.
18 If upon consideration of CASAC and public comments on the REA planning document,
19 the EPA concludes that development of a new REA is not warranted, a REA will not be
20 developed and the Policy Assessment for this review will draw from the REA developed in the
21 last review in light of analyses or assessments made in the REA planning document with regard
22 to the current evidence pertaining to exposure and risk, as well as the evidence presented in the
23 ISA and other documents prepared for the review. Review steps for the PA are described in
24 section 7.1 below.
25
February 2014 5-19 Draft - Do Not Quote or Cite
-------
i 6 AMBIENT AIR MONITORING
2 In the course of NAAQS reviews, aspects of the methods for sampling and analysis of the
3 NAAQS pollutant, and the current network of monitors, including their physical locations and
4 monitoring objectives, are reviewed. The methods for sampling and analysis of each NAAQS
5 pollutant are generally reviewed in conjunction with consideration of the indicator element for
6 each NAAQS. Consideration of the ambient air monitoring network generally informs the
7 interpretation of current data on ambient air concentrations, and helps identify if the monitoring
8 network is adequate to determine compliance with the existing or, as appropriate, a potentially
9 revised NAAQS. This chapter describes plans for considering these aspects of the ambient air
10 monitoring program for oxides of nitrogen which includes the indicator NC>2.
11 6.1 CONSIDERATION OF SAMPLING AND ANALYSIS METHODS
12 Generally, in order to be used for regulatory purposes, ambient NC>2 concentration data
13 must be obtained using Federal Reference Methods (FRMs) or Federal Equivalent Methods
14 (FEMs) which are designated by the Agency in accordance with 40 CFR part 53.55 As described
15 earlier, NC>2 is the indicator for the oxides of nitrogen NAAQS, and has been routinely measured
16 by chemiluminescent FRMs since the early 1980s.56 However, in 2012 a photolytic
17 chemiluminescent method became commercially available and was approved by the Agency as
18 an FEM (Federal Register: Vol 77, page 32632, 06/01/2012;
19 https://www.federalregi ster.gov/articles/2012/06/01/2012-13350/ambient-air-monitoring-
20 reference-and-equivalent-methods-designation-of-three-new-equivalent-methods). This new
21 FEM is expected to be used to some degree into the ambient NO2 monitoring network, but not
22 displace a majority of traditional chemiluminescence FRMs. Both the chemiluminescent FRM
23 and the photolytic chemiluminescent FEM are indirect measurement techniques for NC>2. In the
24 chemiluminescent FRM, the analyzer can only detect NO in the sample stream and therefore
25 utilizes a two-step process in determining the amount of NC>2 in ambient air. First, the analyzer
26 determines the amount of NO in the sample air. Second, the analyzer re-routes air flow so that
27 the sample air stream passes over a heated molybdenum oxide catalytic converter reducing all
28 oxidized nitrogen species in the sample to NO, before again measuring the amount of NO in the
29 sample stream. The analyzer then subtracts the measured, actual ambient NO from the amount
30 measured in the second step, allowing for the determination of NO, NO2, and NOx (where NOx
55 A list of designated FRMs and FEMs is available on EPA's website:
http://www.epa.gov/ttn/amtic/criteria.html.
56 See 40 CFR part 50, Appendix F.
February 2014 6-1 Draft - Do Not Quote or Cite
-------
1 = NO + NC>2). Similarly, the photolytic-chemiluminescence FEM follows the same two step
2 process as the FRM except that the reduction of NC>2 to NO is carried out in a photolytic
3 converter with a known converter efficiency rate. Data produced by FRM and FEM analyzers
4 include NO, NO2, and NOx, which are all routinely logged by state and local agencies whom
5 typically report the hourly average values to EPA's Air Quality System (AQS).
6 The Agency is aware of a number of recent technological advances for direct
7 measurements of NO2 which are now or will soon become commercially available as FEMs, e.g.,
8 cavity attenuated phase shift spectrometry and cavity ring-down spectroscopy. The first of these
9 new methods was approved in November of 2013
10 (http://www.epa.gov/ttnamtil/files/ambient/criteria/reference-equivalent-methods-list.pdf.)
11 These new direct measurement techniques are anticipated to be specific to NO2. This would
12 create a notable, anticipated difference with the traditional chemiluminescent FRMs in that these
13 direct measurement methods will only provide NO2 data, and will not provide NO or NOx data.
14 Sampling and analysis issues to be considered during the current review include the
15 following:
16 • To what extent are additional direct NO2 measurements available for consideration as an
17 FEM?
18 • If new, direct NO2 measurement methods become available and integrated into the
19 ambient network, what would be the anticipated impact to subsequent air quality data
20 analyses by potentially losing NO and NOx measurements?
21 6.2 CONSIDERATION OF AIR MONITORING NETWORK
22 REQUIREMENTS
23 The majority of data used to determine compliance with the NO2 NAAQS are obtained
24 from monitors operated by state, local, and tribal air monitoring agencies. These monitors are
25 either required due to federal regulation contained in 40 CFR Part 58, Appendix D, State
26 Implementation Plans, industrial permits, or other state or local based requirements or voluntary
27 actions. The monitoring networks support three major objectives: (1) to provide air pollution data
28 to the general public in a timely manner; (2) to support compliance with NAAQS and emissions
29 strategy development; and (3) to support air pollution research studies.
30 A review of the available NO2 monitoring network and data was performed as part of the
31 primary NO2 NAAQS review completed in 2010. In conjunction with revising the primary
32 standards in that review, the Agency promulgated minimum monitoring requirements to support
33 the implementation of a new primary 1-hour NO2 standard. The minimum requirements
34 consisted of: (1) near-road monitors which would be placed in locations of expected maximum
35 1-hour NO2 concentrations near heavily trafficked roads in urban areas and (2) monitors located
36 to characterize areas with the highest expected NO2 concentrations at the neighborhood and
February 2014 6-2 Draft - Do Not Quote or Cite
-------
1 larger spatial scales (also referred to as "area-wide" monitors) (75 FR 6505 to 6506, February 9,
2 2010), and (3) a specific set of monitors to maintain in areas with susceptible and vulnerable
3 communities exposed to NC>2 concentrations that have the potential to approach or exceed the
4 standards (75 FR 6509). The near-road NC>2 monitoring requirements were novel at the time of
5 promulgation and stemmed from findings that roadway associated exposures account for a
6 majority of exposures to peak NC>2 concentrations (75 FR 6514). The area-wide and the
7 susceptible and vulnerable communities monitoring requirements each minimally required
8 approximately 52 and 40 monitors, respectively, and were consistent with traditional monitoring
9 approaches, meaning the existing network did not require significant modification in order to
10 satisfy these two requirements. Sites satisfying these two monitoring network requirements were
11 identified and documented, or became fully operational, on or before January 1, 2013.
12 Conversely, the near-road NO2 monitoring network did not exist at the time of promulgation, and
13 the Agency acknowledged that it would have to be designed, funded, and installed in its entirety.
14 One near-road NC>2 monitor is required in each Core Based Statistical Area (CBSA)
15 having 500,000 or more persons, per 40 CFR part 58, Appendix D, section 4.3.2. A second near-
16 road NO2 monitor is required in those CBSAs having either (a) 2,500,000 persons or more, or (b)
17 any CBSA having 500,000 or more persons that also has one or more road segments carrying
18 250,000 or greater Annual Average Daily Traffic (AADT) counts. At each of the near-road NC>2
19 sites, the monitors are subject to all requirements in 40 CFR part 58 and its appendices, which
20 include specific siting criteria such as having the monitor probe placed "... as near as practicable
21 to the outside nearest edge of the traffic lanes of the target segment; but shall not be located at a
22 distance greater than 50 meters, in the horizontal, from the outside nearest edge of the traffic
23 lanes of the target road segment" and having the monitor probe placed between 2 and 7 meters
24 above the ground.
25 The near-road NO2 monitors are required to be installed in three phases (78 FR 16184,
26 March 14, 2013). The first phase required one monitor in any CBSA of 1,000,000 or more
27 persons to be operational by January 1, 2014. We anticipated that 52 near-road sites would be
28 added in Phase 1. The second phase is for any second monitor required in a CBSA (those having
29 2,500,000 or more persons or those CBSAs having 500,000 or more persons that also has one
30 more road segments carrying 250,000 or greater AADT counts) to be operational by January 1,
31 2015. We anticipate 22 near-road sites will be added in Phase II. The third phase is for those
32 monitors in CBSAs having between 500,000 and 1,000,000 persons, which are to be operational
33 by January 1, 2017. We anticipate 52 near-road sites will be added in Phase III.
34 By the end of 2013, the ambient NC>2 monitoring network was estimated to have 391 NC>2
35 monitors in operation nationwide. This estimate does not reflect the impending additions of the
36 three-phased implementation of the near-road NO2 monitors. We anticipate 126 near-road sites
February 2014 6-3 Draft - Do Not Quote or Cite
-------
1 will be operational on or before January 1, 2017, potentially resulting in more than 500 NC>2
2 monitors nationwide when the NC>2 network is fully operational.
Legend
• Active NO2 Monitors - Sept 2013 (-391)
Note: Hawaii has two NO2 sites in operation (not shown).
These active monitors do not include the anticipated near-road NO2 monitors
which are expected to phase into operation at the start of 2014.
Figure 6-1. NOi Monitoring Network: Active monitors as of September 2013.
4
5 In the last review, there was limited near-road ambient NC>2 monitoring data. Analyses
6 conducted as part of the 2008 REA (U.S. EPA, 2008b,) along with public comment on the
7 proposed rule were considered in reaching final decisions on how many and where near-road
8 NC>2 monitors would be required. In particular, the 2008 REA considered estimates of on- or
9 near-roadway exceedances in 17 urban areas associated with CBS A populations ranging from
10 approximately 540,000 to 19,000,000 persons. Those analyses indicated that the areas under
11 explicit consideration were estimated to experience NO2 concentrations on- or near-roads that
12 exceeded health benchmark levels. In this review, the EPA will have the benefit of monitored
13 near-road data for consideration in analyzing potential exposures and exceedances. Although the
14 new near-road NC>2 monitoring network is not yet fully installed, data from newly operational
15 monitors, plus data from more recent research efforts, may provide a clearer picture of what NC>2
16 concentrations are in the near-road environment, with the continued understanding that factors
February 2014
6-4
Draft - Do Not Quote or Cite
-------
1 including traffic counts, fleet mix, congestion patterns, roadway design, terrain, and meteorology
2 all play a major role in measured roadside NC>2 concentrations.
3 Considering the availability of new near-road NC>2 monitoring data, the EPA may be in a
4 position to re-evaluate the analyses underlying the minimum monitoring requirements
5 promulgated in the 2010 revisions in this review.
February 2014 6-5 Draft - Do Not Quote or Cite
-------
i 7 POLICY ASSESSMENT AND RULEMAKING
2 As outlined in section 1.2 above, the fourth and final stage of the NAAQS review is the
3 preparation of a Policy Assessment (PA) and rulemaking notices. These two steps are described,
4 respectively, in sections 7.1 and 7.2 below.
5 7.1 POLICY ASSESSMENT
6 The PA provides a transparent OAQPS staff analysis and staff conclusions regarding the
7 adequacy of the current standards and potential alternatives that are appropriate to consider prior
8 to the issuance of proposed and final rules. The PA integrates and interprets the information from
9 the ISA and, if available, REA(s) to frame policy options for consideration by the Administrator.
10 The PA is also intended to facilitate CAS AC's advice to the Agency and recommendations to the
11 Administrator on the adequacy of the existing standards or revisions that may be appropriate to
12 consider, as provided for in the CAA. Staff conclusions in the PA are based on the information
13 contained in the ISA, and, as available, the REA, and any additional staff evaluations and
14 assessments discussed in the PA. In so doing, the discussion in the PA is framed by consideration
15 of a series of policy-relevant questions drawn from those outlined in section 3.2 above, including
16 the fundamental questions associated with the adequacy of the current standards and, as
17 appropriate, consideration of alternative standards in terms of the specific elements of the
18 standards: indicator, averaging time, level, and form.
19 The PA for the current review will identify conceptual evidence-based and risk/exposure-
20 based approaches for reaching public health policy judgments. It will discuss the implications of
21 the science and quantitative assessments for the adequacy of the current primary standards and
22 for any alternative standards under consideration. The PA will also describe a broad range of
23 policy options for standard setting, identifying the range for which the staff identifies support
24 within the available information. In so doing, the PA will describe the underlying interpretations
25 of the scientific evidence and risk/exposure information that might support such alternative
26 policy options that could be considered by the Administrator in making decisions for the primary
27 NC>2 standards. Additionally, the PA will identify key uncertainties and limitations in the
28 underlying scientific information and in our assessments. The PA will also highlight areas for
29 future health-related research, model development, and data collection.
30 In identifying a range of primary standard options for the Administrator to consider, it is
31 recognized that the final decision will be largely a public health policy judgment. A final
32 decision must draw upon scientific information and analyses about health effects and risks, as
33 well as judgments about how to deal with the range of uncertainties that are inherent in the
34 scientific evidence and analyses. Staffs approach to informing these judgments recognizes that
February 2014 7-1 Draft - Do Not Quote or Cite
-------
1 the available health effects evidence generally reflects a continuum consisting of ambient
2 concentrations at which scientists generally agree that health effects are likely to occur, through
3 lower concentrations at which the likelihood and magnitude of the response become increasingly
4 uncertain. This approach is consistent with the requirements of the NAAQS provisions of the
5 CAA and with how the EPA and the courts have historically interpreted the Act. These
6 provisions require the Administrator to establish primary standards that are requisite to protect
7 public health and are neither more nor less stringent than necessary for this purpose. As
8 discussed in section 1.1 above, the provisions do not require that primary standards be set at a
9 zero-risk level, but rather at a level that avoids unacceptable risks to public health, including the
10 health of at-risk populations.57
11 Staff will prepare at least one draft of the PA document for CAS AC review and public
12 comment. The draft PA document will be distributed to the CASAC Oxides of Nitrogen Primary
13 NAAQS Review Panel for their consideration and provided to the public for review and
14 comment. Review by the CASAC Panel will be discussed at public meetings that will be
15 announced in the Federal Register. Based on past practice by CASAC, the EPA expects that
16 CASAC will summarize key advice and recommendations for revision of the document in a
17 letter to the EPA Administrator. In revising the draft PA document, OAQPS will take into
18 account any such recommendations and also consider comments received, from CASAC and
19 from the public, at the meeting itself, and any written comments received. The final document
20 will be made available on an EPA website, with its public availability announced in the Federal
21 Register.
22 7.2 RULEMAKING
23 Following issuance of the final PA and the EPA management consideration of staff
24 analyses and conclusions presented therein, and taking into consideration CASAC advice and
25 recommendations, the Agency will develop a notice of proposed rulemaking. The proposed
26 rulemaking notice conveys the Administrator's proposed conclusions regarding the adequacy of
27 the current standards and any revision that may be appropriate. The EPA will submit a draft
28 notice of proposed rulemaking to the Office of Management and Budget (OMB) for interagency
29 review, to provide OMB and other federal agencies the opportunity for review and comment.
30 After the completion of interagency review, the EPA will publish the notice of proposed
57 The at-risk population groups identified in a NAAQS review may include low income or minority
groups. Where low income/minority groups are among the at-risk populations, the rulemaking decision will be
based on providing protection for these and other at-risk populations and lifestages (e.g., children, older adults,
persons with pre-existing heart and lung disease). To the extent that low income/minority groups are not among the
at-risk populations identified in the ISA, a decision based on providing protection of the at-risk lifestages and
populations would be expected to provide protection for the low income/minority groups.
February 2014 7-2 Draft - Do Not Quote or Cite
-------
1 rulemaking in the Federal Register. Monitoring rule changes associated with review of the
2 primary NC>2 standards, and drawing from considerations outlined in Chapter 6 above, will be
3 developed and proposed, as appropriate, in conjunction with this NAAQS rulemaking.
4 At the time of publication of the notice of proposed rulemaking, all materials on which
CO
5 the proposal is based are made available in the public docket for the rulemaking. Publication
6 of the proposal notice is followed by a public comment period, generally lasting 60 to 90 days,
7 during which the public is invited to submit comments on the proposal to the rulemaking docket.
8 Taking into account comments received on the proposed rule, the Agency will then develop a
9 notice of final rulemaking, which again undergoes OMB-coordinated interagency review prior to
10 issuance by the EPA of the final rule. At the time of final rulemaking, the Agency responds to all
11 significant comments on the proposed rule.59 Publication of the final rule in the Federal Register
12 completes the rulemaking process.
58 The rulemaking docket for the current primary NO2 NAAQS review is identified as EPA-HQ-OAR-
2013-0146. This docket has incorporated the ISA docket (EPA-HQ-ORD-2013-0232) by reference. Both dockets
are publicly accessible at www.regulations.gov.
59 For example, Agency responses to all substantive comments on the 2009 notice of proposed rulemaking
in the last review were provided in the preamble to the final rule and in a document titled Responses to Significant
Comments on the 2009 Proposed Rule on the Primary National Ambient Air Quality Standards for Nitrogen Dioxide
(July 15, 2009; 74 FR 34404) (U.S. EPA, 2010).
February 2014 7-3 Draft - Do Not Quote or Cite
-------
8 REFERENCES
2 Cape, JN, Tang YS, van Dijk N, Love L, Sutton MA, Palmer SCF. (2004). Concentrations of ammonia and nitrogen
3 dioxide at roadside verges, and their contribution to nitrogen deposition. Environ Pollut. 132:469-478.
4 Folinsbee L (1992). Does nitrogen dioxide exposure increase airways responsiveness? Toxicol. Ind. Health. 8(5):
5 273-283.
6 Ito, K; Mathes, R; Ross, Z; Nadas, A; Thurston, G; Matte, T. (2011). Fine paniculate matter constituents associated
7 with cardiovascular hospitalizations and mortality in New York City. Environ Health Perspect. 119: 467-
8 473.
9 Jackson L (2009). Memo from Administrator Lisa P. Jackson to Elizabeth Craig, Acting Assistant Administrator for
10 OAR and Lek Kadeli, Acting Assistant Administrator for ORD. Process for Reviewing the National
11 Ambient Air Quality Standards. May 21, 2009. Available at:
12 http://www.epa.gov/ttn/naaqs/pdfs/NAAQSReviewProcessMemo52109.pdf.
13 Rizzo (2008). Investigation of how distributions of hourly nitrogen dioxide concentrations have changed over time
14 in six cities. Nitrogen Dioxide NAAQS Review Docket (EPA-HQ-OAR-2006-0922 ). Available at
15 http://www.epa.gOv/ttn/naaqs/standards/nox/s noxcrrea.html.
16 Samet J (2008a). Letter form Dr. Jonathan M. Samet, Chair, Clean Air Scientific Advisory Committee to EPA
17 Administrator Stephen Johnson. Clean Air Scientific Advisory Committee's (CASAC) Peer Review of
18 Draft Chapter 8 of EPA's Risk and Exposure Assessment to Support the Review of the NO2 Primary
19 National Ambient Air Quality Standard. EPA-CASAC-09-001. October 28, 2008. Available at:
20 http://vosemite.epa.gov/sab/sabproduct.nsf/264cbl227d55e02c85257402007446a4/87D38275673D66B885
21 2574F00069D45E/$File/EPA-CAS AC-09-001-unsigned.pdf.
22 Samet J (2008b). Letter form Dr. Jonathan M. Samet, Chair, Clean Air Scientific Advisory Committee to EPA
23 Administrator Stephen Johnson. Clean Air Scientific Advisory Committee's (CASAC) Review comments
24 on EPA's Risk and Exposure Assessment to Support the Review of the NO2 Primary National Ambient Air
25 Quality Standard. EPA-CASAC-09-003. December 16, 2008. Available at:
26 http://vosemite.epa.gov/sab/sabproduct.nsf/264cbl227d55e02c85257402007446a4/9C4A540D86BFB67A8
27 52575210074 A7AE/$File/EP A-CASAC-09-003-unsigned.pdf.
28 Samet J (2009). Letter form Dr. Jonathan M. Samet, Chair, Clean Air Scientific Advisory Committee to EPA
29 Administrator Lisa P. Jackson. Comments and Recommendations Concerning EPA's Proposed Rule for the
3 0 Revision of the National Ambient Air Quality Standards (NAAQS) for Nitrogen Dioxide. EPA-CAS AC-
31 09-014. September 9, 2009. Available at:
32 http://vosemite.epa.gov/sab/sabproduct.nsf/264cbl227d55e02c85257402007446a4/0067573718EDA17F85
33 25762C0074059E/$File/EPA-CASAC-09-014-unsigned.pdf.
34 Thompson R (2008). Nitrogen Dioxide (NO2) Descriptive Statistics Tables. Memo to the NO2 NAAQS Review
3 5 Docket. November 5, 2008. Available at:
36 http://www.epa.gov/ttn/naaqs/standards/nox/data/20081124 thompson.pdf.
3 7 Tolbert, PE, Klein M, Peel JL, Sarnat SE, Sarnat JA. (2007). Multipollutant modeling issues in a study of ambient
38 air quality and emergency department visits in Atlanta. J Expos Sci Environ Epidemiol. 17S2:S29-35.
39 U.S. EPA (1971). Air Quality Criteria for Nitrogen Oxides. U.S. Environmental Protection Agency. Air Pollution
40 Control Office, Washington, D.C. January 1971. Air Pollution Control Office Publication No. AP-84.
41 U.S. EPA (1993). Air Quality Criteria for Oxides of Nitrogen. Office of Health and Environmental Assessment,
42 Environmental Criteria and Assessment Office. Research Triangle Park, NC. EPA-600/8-9!-049aF-cF,
43 August 1993. Available at: http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179.
44 U.S. EPA (1995a). Review of the National Ambient Air Quality Standards for Nitrogen Oxides: Assessment of
45 Scientific and Technical Information, OAQPS Staff Paper. US EPA, Office of Air Quality Planning and
46 Standards, Research Triangle Park, NC. EPA-452/R-95-005, September 1995. Available at:
47 http://www.epa. gov/ttn/naaqs/standards/nox/data/noxsp 1995 .pdf.
February 2014 8-1 Draft - Do Note Quote or Cite
-------
1 U.S. EPA (2007a). Integrated Review Plan for the Primary National Ambient Air Quality Standard for Nitrogen
2 Dioxide. U.S. EPA. National Center for Environmental Assessment and Office of Air Quality Planning and
3 Standards. Research Triangle Park, NC. August 2007. Available at:
4 http://www.epa.gov/ttn/naaqs/standards/nox/data/20070823 nox review_plan_final.pdf.
5 U.S. EPA (2007b). Nitrogen Dioxide Health Assessment Plan: Scope and Methods for Exposure and Risk
6 Assessment. U.S. EPA, Office of Air Quality Planning and Standards, Research Triangle Park, NC. Draft
7 September 2007. Available at:
8 http://www.epa.gov/ttn/naaqs/standards/nox/data/20070927 risk exposure scope.pdf.
9 U.S. EPA (2008a). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria. U.S. EPA, National
10 Center for Environmental Assessment and Office, Research Triangle Park, NC. EPA/600/R-08/071. July
11 2008. Available at: http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid= 194645.
12 U.S. EPA (2008b). Risk and Exposure Assessment to Support the Review of the NO2 Primary National Ambient Air
13 Quality Standard. U.S. EPA, Office of Air Quality Planning and Standards. Research Triangle Park, NC.
14 EPA 452/R-08-008a/b. November 2008. Available at:
15 http://www.epa.gOv/ttn/naaqs/standards/nox/s nox crrea.html.
16 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter. U.S. EPA, National Center for
17 Environmental Assessment. Research Triangle Park, NC. EPA 600/R-08/139F. December 2009. Available
18 at: http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
19 U.S. EPA (2010). Responses to Significant Comments on the 2009 Proposed Rule on the Primary National Ambient
20 Air Quality Standards for Nitrogen Dioxide (July 15, 2009; 74 FR 34404). U.S. Environmental Protection
21 Agency. January 2010. Available at: http://www.epa.gov/ttn/naaqs/standards/nox/data/20100122rtc.pdf.
22 U.S. EPA (2013a). Draft Plan for Development of the Integrated Science Assessment (ISA) for Nitrogen Oxides —
23 Health Criteria. U.S EPA, National Center for Environmental Assessment. Research Triangle Park, NC.
24 May 2013. Available at: http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=254570#Download.
25 U.S. EPA (2013b). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (First External Review
26 Draft). U.S. EPA, National Center for Environmental Assessment and Office, Research Triangle Park.
27 EPA/600/R-13/202. November 2013. Available at:
28 http://cfpub.epa. gov/ncea/isa/recordisplav.cfm?deid=259167.
29 U.S. EPA (2013c). Integrated Science Assessment for Ozone and Related Photochemical Oxidants (Final Report).
30 U.S. EPA, Washington, DC. February 2013. EPA/600/R-10/076F. Available at:
31 http://www.epa.gOv/ttn/naaqs/standards/ozone/s o3 2008 isa.html.
32 Zeger, SL; Thomas, D; Dominici, F; Samet, JM; Schwartz, J; Dockery, D; Cohen, A. (2000). Exposure
3 3 Measurement Error in Time-series Studies of Air Pollution: Concepts and consequences. Environ Health
34 Perspect. 108(5): 419-426.
February 2014 8-2 Draft - Do Note Quote or Cite
-------
APPENDIX A
DRAFT OUTLINE FOR INTEGRATED SCIENCE ASSESSMENT FOR OXIDES OF
NITROGEN - HEALTH CRITERIA
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43 1.7
Preamble
(will be available online)
Preface
Executive Summary
Chapter 1
1.1
1.2
1.3
1.4
1.5
traffic proximity
1.6
Process of ISA Development
Literature Search
Study Selection
Evaluation of Individual Study Quality
Evaluation, Synthesis, and Integration across Disciplines and
Development of Scientific Conclusions and Causal
Determinations
EPA Framework for Causal Determinations
Public Health Impact
Approach to Classifying At-risk Factors
Concepts in Evaluating Adversity of Health Effects
Legislative Requirements for the Primary NAAQS Review
History of the Review of the Air Quality Criteria for the Oxides
of Nitrogen and NAAQS for Nitrogen Dioxide
Integrative Summary
ISA Development and Scope
Organization of the ISA
Sources of Oxides of Nitrogen to Human Exposure
Health Effects of Oxides of Nitrogen
Dosimetry and Modes of Action
Causal Determinations and Key Evidence for Evaluated
Health Effects
Evaluation of the Independent Effects of Nitrogen Dioxide
Potential confounding by time-varying and individual- or
population-level characteristics
Potential confounding by copollutant exposures -
multivariate models, indoor NO2,
and intensity
Policy-Relevant Considerations
NO2 Exposure Metrics
Lag Structure of NO2-related Morbidity and Mortality
Associations
Concentration-Response Relationships and Thresholds
Public Health Significance - Adversity of Effects, At-risk
Lifestages and Populations
Conclusions
February 2014
A-l
Draft - Do Not Quote or Cite
-------
1
2 Chapter 2 Atmospheric Chemistry and Exposure to Nitrogen Oxides
3 2.1 Introduction
4 2.2 Atmospheric Chemistry and Fate
5 2.3 Sources
6 2.4 Measurement Methods
7 2.5 Ambient Concentrations of Oxides of Nitrogen
8 2.6 Exposure Assessment
9 2.7 Summary and Conclusions
10
11 Chapter 3 Dosimetry and Modes of Action of Inhaled Nitrogen Oxides
12 3.1 Introduction
13 3.2 Dosimetry of Inhaled Oxides of Nitrogen
14 Dosimetry of Nitrogen Dioxide
15 Dosimetry of Nitric Oxide
16 Metabolism, Distribution, and Elimination of
17 Products Derived from Inhaled Oxides of Nitrogen
18 Summary
19 3.3 Modes of Action for Inhaled Oxides of Nitrogen
20 Introduction
21 Nitrogen Di oxi de
22 Nitric Oxide
23 Metabolites of Nitric Oxide and Nitrogen
24 Dioxide
25 Summary
26 3.4 Summary
27
28 Chapter 4 Integrated Health Effects of Short-term Exposure to Oxides
29 of Nitrogen60
30 4.1 Introduction
31 4.2 Respiratory Effects
32 4.3 Cardiovascular Effects
33 4.4 Total Mortality
34
35 Chapters Integrated Health Effects of Long-term Exposure to Oxides
36 of Nitrogen35
37 5.1 Introduction
38 5.2 Respiratory Effects
39 5.3 Cardiovascular Effects
40 5.4 Reproductive and Developmental Effects
41 Fertility, Reproduction, and Pregnancy
42 Birth Outcomes
43 Postnatal Development
60Sections for each of the major health effect outcome categories will include a review of the available evidence and
conclude with the causal determination and summary of contributing evidence.
February 2014 A-2 Draft - Do Not Quote or Cite
-------
1 5.5 Total Mortality
2 5.6 Cancer
3
4 Chapter 6 Lifestages and Populations Potentially at Increased Risk for
5 Health Effects Related to Exposure to Oxides of Nitrogen
6 6.1 Introduction
7 6.2 Genetic Factors
8 6.3 Pre-existing Disease/Conditions
9 6.4 Sociodemographic Factors (lifestage, socioeconomic status,
10 race/ethnicity, sex)
11 6.5 Behavioral and Other Factors (diet, obesity, smoking, residential
12 location)
13 6.6 Summary
February 2014 A-3 Draft - Do Not Quote or Cite
-------
[This page intentionally left blank]
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/P-14-001
Environmental Protection Health and Environmental Impacts Division February 2014
Agency Research Triangle Park, NC
------- |