United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
EPA/600/8-87/014
June 1987
Research and Development
&EPA
EPA Indoor Air Quality
Implementation Plan:
Appendix A.
Preliminary Indoor Air
Pollution Information
Assessment
-------
EPA-600/8-87-014
June 1987
EPA Indoor Air Quality
Implementation Plan
Appendix A: Preliminary Indoor Air Pollution
Information Assessment
U.S. Environmental Protection Agency
Office of Research and Development
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Research Triangle Park NC 27711
Protection
230 South Dearborn Street ^"
Chicago, Illinois 60604 A
':>Ǥ
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental
Protection Agency policy and approved for publication. Mention of trade names
or commercial products does not constitute endorsement or recommendation for
use.
-------
CONTENTS
APPENDIX A
PREFACE xiii
1. INTRODUCTION 1-1
1.1 HISTORICAL BACKGROUND 1-1
1.2 THE EMERGENCE OF AN INDOOR AIR PROGRAM AND POLICY 1-3
1.3 PRELIMINARY INFORMATION ASSESSMENT AND RESEARCH NEEDS
STATEMENT 1-4
1.4 SUMMARY OF WHAT IS KNOWN ABOUT INDOOR AIR POLLUTANTS 1-6
2. POLLUTANT CATEGORIES 2-1
2.1 INTRODUCTION 2-1
2.2 COMBUSTION GASES 2-4
2.2.1 Introduction 2-4
2.2.2 Occurrence and Sources of Combustion Gases 2-6
2.2.2.1 Unvented Kerosene Space Heaters 2-6
2.2.2.2 Gas Appliances 2-8
2.2.2.2.1 Gas Stoves 2-9
2.2.2.2.2 Gas-Fired Space Heaters 2-11
2.2.2.2.3 Gas-Fired Water Heaters and
Dryers 2-13
2.2.2.3 Wood-Burning Stoves and Fireplaces 2-14
2.2.2.4 Attached Garages 2-15
2.2.2.5 Conclusions 2-16
2.2.2.6 Major Knowns and Unknowns 2-17
2.2.3 Carbon Monoxide 2-17
2.2.3.1 Exposure 2-17
2.2.3.2 Monitoring of CO 2-18
2.2.3.3 Health Effects 2-20
2.2.3.4 Conclusions 2-25
2.2.4 Nitrogen Dioxide 2-26
2.2.4.1 Exposure 2-26
2.2.4.2 Monitoring 2-26
2.2.4.2.1 Absorption of N02 2-30
2.2.4.3 Health Effects of N02 2-30
2.2 A A Summary of Knowns and Unknowns for N02 2-38
2.2.5 Sulfur Dioxide 2-41
2.2.5.1 Monitoring 2-41
2.2.5.2 Health Effects 2-42
2.3 PARTICLES AND OTHER COMBUSTION PRODUCTS 2-42
2.3.1 Introduction 2-42
2.3.2 Particles and Organics from Combustion 2-44
2.3.2.1 Occurrence and Sources 2-45
2.3.2.1.1 Unvented Kerosene Space Heaters . 2-45
2.3.2.1.2 Gas Appliances 2-46
2.3.2.1.3 Wood-burning Stoves and
Fireplaces 2-47
2.3.2.1.4 Attached Garages 2-48
111
-------
CONTENTS (continued)
2.3.2.1.5 Conclusions 2-48
2.3.2.2 Exposure 2-49
2.3.2.3 Monitoring of PM 2-51
2.3.2.4 Health Effects 2-53
2.3.2.4.1 Health Effects Associated
With Exposure to Soot 2-58
2.3.3 Polycyclic Aromatic Hydrocarbons 2-59
2.3.4 Other Combustion Organics 2-60
2.3.5 Interaction of Particulate Matter and Organics 2-60
2.3.6 Monitoring 2-61
2.3.7 Woodsmoke 2-64
2.3.7.1 Health Effects 2-64
2.3.8 Major Knowns and Unknowns: Health Effects 2-66
2.3.9 Mitigation and Control Options for Combustion
Sources 2-67
2.3.9.1 Major Knowns and Unknowns 2-71
2.4 ENVIRONMENTAL TOBACCO SMOKE 2-72
2.4.1 Introduction 2-72
2.4.2 Source Characterization 2-74
2.4.3 Exposure Assessment 2-75
2.4.3.1 Exposure Modeling 2-76
2.4.3.2 Monitoring Exposure 2-77
2.4.3.3 Biological Markers 2-78
2.4.4 Health Effects 2-79
2.4.4.1 Introduction 2-79
2.4.4.2 Acute Irritating and Immune Effects 2-79
2.4.4.3 Respiratory Effects 2-79
2.4.4.4 Lung Cancer and Other Cancers 2-87
2.4.4.5 Cardiovascular Disease and Other Effects ... 2-89
2.4.5 IAQ Control Options 2-92
2.4.6 Conclusions 2-93
2.4.6.1 What is Known 2-93
2.4.6.2 What Scientific Information is Missing 2-93
2.4.6.3 Research Needs 2-93
2.5 NONCOMBUSTION PARTICLES 2~94
2.5.1 Asbestos 2-94
2.5.1.1 Sources 2-95
2.5.1.2 Monitoring 2-96
2.5.1.3 Exposure 2-99
2.5.1.4 Health Effects 2-101
2.5.2 Dusts, Sprays, and Cooking Aerosols 2-103
2.5.2.1 Introduction 2-103
2.5.2.2 Monitoring 2-104
2.5.3 IAQ Control Options 2-104
IV
-------
CONTENTS (continued)
2.6 NONCOMBUSTION GAS-PHASE ORGANIC COMPOUNDS 2-105
2.6.1 Gas-Phase Organic Compounds 2-105
2.6.1.1 Introduction 2-105
2.6.1.2 Occurrence and Sources of Gas-Phase
Organic Compounds 2-105
2.6.1.2.1 Outdoor Sources of Volatile
Organic Compounds 2-106
2.6.1.2.2 Sources of Indoor Organic
Vapors 2-106
2.6.1.2.3 Emission Rates 2-107
2.6.1.3 Monitoring of Gas-Phase Organics 2-110
2.6.1.4 Health Effects 2-116
2.6.1.4.1 Neurotoxicity of Volatile
Organic Compounds 2-116
2.6.1.4.2 Genotoxicity of Volatile
Organic Compounds in Relation-
ship to Total Organic Species ... 2-118
2.6.1.5 Mitigation and Control Options 2-121
2.6.1.5.1 Ventilation 2-121
2.6.1.5.2 Air Cleaners 2-121
2.6.1.5.2.1 Absorption 2-121
2.6.1.5.2.2 Catalytic
Oxidation 2-121
2.6.1.5.3 Material/Product Selection 2-122
2.6.1.5.4 Material/Product Use 2-122
2.6.1.5.5 Other Measures 2-122
2.6.2 Formaldehyde 2-123
2.6.2.1 Sources of Formaldehyde 2-123
2.6.2.2 Monitoring of Formaldehyde 2-124
2.6.2.3 Health Effects 2-125
2.7 RADON 2-130
2.7.1 Occurrence and Sources of Radon 2-130
2.7.2 Indoor Concentrations and Exposures 2-133
2.7.3 Health Effects Associated with Radon Exposure 2-135
2.7.4 Risk Estimates 2-137
2.7.5 Estimate of Dosage to People Exposed to Radon 2-137
2.7.6 Indoor Air Quality Control Options 2-138
2.8 BIOLOGICAL CONTAMINANTS 2-143
2.8.1 Introduction 2-143
2.8.2 Sources of Biological Contaminants 2-144
2.8.3 Monitoring of Biological Contaminants 2-146
2.8.4 Health Effects of Biological Contaminants 2-148
2.8.4.1 Infection 2-148
2.8.4.2 Mycointoxication 2-149
2.8.4.3 Allergenic Reactions 2-151
2.8.4.3.1 Allergic Rhinitis 2-151
2.8.4.3.2 Bronchial Asthma 2-151
-------
CONTENTS (continued)
2.8.4.3.3 Hypersensitivity Pneumonitis 2-152
2.8.4.3.4 "Monday Complaints" 2-152
2.8.4.4 Other Allergens 2-152
2.8.5 Indoor Air Quality Control Options 2-152
2.9 PESTICIDES 2-153
2.9.1 Introduction 2-153
2.9.2 Sources of Pesticide Exposure 2-153
2.9.2.1 Emission Rates 2-156
2.9.3 Exposure to Pesticides 2-156
2.9.4 Monitoring 2-157
2.9.5 Health Effects Associated With Pesticides Exposure .. 2-161
2.9.6 Mitigation and Control Options 2-161
2.9.6.1 Ventilation 2-161
2.9.6.2 Air Cleaners 2-161
2.9.6.3 Material/Product Selection and Use 2-161
2.9.6.4 Other Measures 2-162
2.10 NONIONIZING RADIATION: EXTREMELY LOW FREQUENCY ELECTRIC
AND MAGNETIC FIELDS 2-162
2.10.1 Occurrence and Sources of Nonionizing Radiation 2-162
2.10.2 Distribution of Levels and Exposure 2-163
2.10.3 Health Effects of Nonionizing Radiation 2-163
2.10.4 Estimate of Population at Risk 2-166
2.10.5 Mitigation and Control Options 2-166
2.10.6 Conclusions 2-167
3. BUILDING SYSTEMS 3-1
3.1 INTRODUCTION 3-1
3.2 THE BUILDING SYSTEM AS A SOURCE OF INDOOR AIR QUALITY
PROBLEMS 3-2
3.2.1 General 3-2
3.2.2 Ventilation Problems 3-2
3.2.3 Source Effects 3-3
3.2.4 Arrangement of Building Space and Activities 3-4
3.3 MITIGATION OF INDOOR AIR POLLUTION 3-5
3.3.1 General 3-5
3.3.2 Confining Pollution 3-5
3.3.3 Exhausting Pollution 3-5
3.3.4 Dilution with outside air 3-6
3.3.5 Air Cleaners 3-6
3.3.6 Selection of Materials 3-7
3.3.7 Elimination of Entry Routes 3-7
3.4 COMFORT AND OTHER ISSUES 3-7
3.5 MEASUREMENT AND DIAGNOSIS OF BUILDING SYSTEM FACTORS 3-8
3.5.1 General 3-8
3.5.2 Air Circulation/Ventilation 3-8
VI
-------
CONTENTS (continued)
Page
3.5.3 Pollutant Concentration and Identification 3-9
3.5.3.1 Particles 3-9
3.5.3.2 Gases 3-9
3.5.4 Questionnaires 3-11
3.5.5 Diagnosis 3-11
3.6 INDOOR AIR MODELS 3-11
4. WELFARE CONCERNS 4-1
4.1 INTRODUCTION 4-1
4.2 MATERIALS DAMAGE 4-5
4.2.1 Introduction 4-5
4.2.2 Oxides of Nitrogen 4-6
4.2.3 Sulfur Oxides 4-7
4.2.4 Ozone 4-9
4.2.5 Particulate Matter 4-10
4.3 SOILING 4-11
4.4 ODORS 4-17
4.5 ECONOMIC EFFECTS 4-22
5. SOURCE CATEGORIZATION 5-1
5.1 INTRODUCTION 5-1
5.2 ENVIRONMENTAL TOBACCO SMOKE 5-1
5.3 BIOLOGICAL CONTAMINANTS 5-9
5.4 PERSONAL ACTIVITIES 5-10
5.5 BUILDING SYSTEMS 5-11
5.6 MATERIALS AND FURNISHINGS 5-12
5.7 COMBUSTION APPLIANCES 5-13
5.8 OUTDOOR SOURCES 5-16
5.8.1 Radon 5-16
5.8.2 Pesticides 5-18
5.9 CONCLUSIONS 5-19
6. REFERENCES 6-1
ATTACHMENT A: RESPONSE TO ISSUES FROM HARVARD WORKSHOP,
JANUARY 1987 A-l
-------
TABLES
Number
1-1
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
2-11
2-12
2-13
2-14
2-15
2-16
2-17
2-18
Summary: what is known about indoor pollutants
Exposure profile by source prevalence and use patterns
for selected pollutants in the U.S
Indoor air sources and related contaminants
Pollutant emission rates from four portable kerosene-fired
heaters
Emission rates for gas ranges and ovens
Overall comparison of emission factors from IGT and LBL
chamber studi es
Pollutant emission rates
Indoor microenvironments listed in descending order of
weighted mean CO concentration
Indoor residential CO data
Lowest observed effect levels for human health effects
associated with low level carbon monoxide exposure
Controlled human exposure studies of N02 effects
Effects of gas cooking on respiratory illnesses and
symptoms in children
Effects of gas cooking on lung function in children
Effects of gas cooking on pulmonary illness, symptoms,
and f uncti on of adul ts
Two-week average S02 levels by location for homes in six
principal source categories
Particulate emission rates from kerosene space heaters
Nitrated PAH source strengths from well -tuned radiant
and maladjusted convective kerosene space heaters
Gas range and oven particulate emission rates
Particulate emission rates for a gas-fired blue-flame
space heater
Page
1-14
2-5
2-7
2-8
2-10
2-13
2-16
2-18
2-19
2-22
2-33
2-35
2-39
2-40
2-43
2-45
2-46
2-47
2-47
VTM
-------
TABLES (continued)
Number
2-19
2-20
2-21
2-22
2-23
2-24
2-25
2-26
2-27
2-28
2-29
2-30
2-31
2-32
2-33
2-34
2-35
Emission rates for particulate and particulate-bound
materials
Quantile descriptors of personal, indoor, and outdoor RSP
concentrations, by location
Quantity of particles collected in micrograms at a
concentration of 30 ug/m3
Summary of key quantitative conclusions based on newly
available epidemiological studies or analyses relating
health effects to acute exposure to ambient air levels of
S02 and/or PM
Summary of key quantitative conclusions based on newly
available epidemiological studies relating human health
effects to long-term exposures of S02 and/or PM
Methods of IAQ Control for Combustion Sources
Some sidestream to mainstream ratios determined for
various elements and compounds in cigarette smoke
Prevalence of respiratory symptoms in selected investiga-
tions of children, by number of smoking parents
Epidemiological studies of early childhood respiratory
illnesses and passive smoking
Cohort and case-control studies of passive exposure to
tobacco smoke and 1 ung cancer
Summary of environmental asbestos sampling studies
Solvent based organic compounds
Specific indoor sources of organic vapors
Emission rates from particleboard
Comparison of organic emission rates: silicone caulk and
f 1 oor adhes i ve
Weighted 12-hour measurements for breath of residents
for Bayonne and Elizabeth, NJ, combined-TEAM first season ...
Neurotoxic effects of volatile organic compounds commonly
found i n i ndoor envi ronments
Page
2-49
2-51
2-54
2-56
2-57
2-67
2-73
2-81
2-82
2-85
2-100
2-107
2-108
2-109
2-111
2-113
2-117
IX
-------
TABLES (continued)
Number
2-36
2-37
2-38
2-39
2-40
2-41
2-42
2-43
2-44
2-45
3-1
4-1
4-2
4-3
4-4
5-1
Partitioning of indoor air compounds as indicated by
reviewed 1 iterature
Chemicals detected in indoor air sampling: their
occurrence and bioassay
Acute human health effects of formaldehyde at various
concentrati ons
Surveys of occupants living or working in mobile homes or
homes wi th UFFI
Studies of formaldehyde exposed cohorts and cancer
Impaction samples useful for bioaerosol sampling
Source of indoor pesticide exposure
Emission rates of paradichlorobenzene from moth crystal
cakes
Summary of pesticides found in air from NOPES
Summary of monitoring data for the five most prevalent
pesticides
Methods of air infiltration measurement
Air pol 1 uti on effects on materi al s
Selected physical damage functions related to S02
exposure
Summary of literature addressing factors contributing
to soiling effects
Major odorous air pollutants, olfactory thresholds, and
rel ated data
Relationship between pollutant and source categories
Page
2-119
2-120
2-125
2-127
2-129
2-147
2-154
2-157
2-160
2-160
3-10
4-3
4-8
4-13
4-19
5-20
-------
FIGURES
Number Page
2-1 Mean indoor/outdoor difference in nitrogen dioxide
concentrations by cooking fuel and kitchen ventilation,
average across all indoor/outdoor sites (May 1977-April
1978) 2-27
4-1 Factors contributing to soiling effects of deposited
particles 4-12
4-2 Relationship among emissions, air quality, damages and
benefits, and policy decisions 4-23
-------
AUTHORS AND CONTRIBUTORS
Harriet Ammann
Michael Berry
Carl Blackman
Vandy Bradow
Darcy Campbell
Larry Claxton
Beverly Comfort
Robert Dyer
Joseph Elder
Rob Eli as
Judith Graham
Dave Holland
Donald Horstman
Dennis Kotchmar
Joel 1 en Lewtas
Dave Mage
Harriet Burge -
Jonathan Samet -
ECAO/RTP
ECAO/RTP
HERL
ECAO/RTP
ECAO/RTP
HERL
ECAO/RTP
HERL
HERL
ECAO/RTP
HERL
EMSL/RTP
HERL
ECAO/RTP
HERL
Thomas McMullen
Frederick Miller
Ronald Mosley
Judy Mumford
John O'Neil
Dave Otto
Jim Raub
Charles Rodes
Max Samfield
Leslie Sparks
Bruce Tichenor
W. Eugene Tucker
Lance Wallace
James White
Sam Windham
EMSL/RTP
University of Michigan, Ann Arbor
New Mexico Tumor Registry, U.N.M. Medical Center
ECAO/RTP
HERL
AEERL/RTP
HERL
HERL
HERL
ECAO/RTP
EMSL/RTP
AEERL/RTP
AEERL/RTP
AEERL/RTP
AEERL/RTP
EMSD
AEERL/RTP
OAR
The project officer for this document was Harriet Ammann, MD-52, U.S. EPA
ECAO, Research Triangle Park, NC 27711
Technical assistance within the Environmental Criteria and Assessment Office
was provided by: Norm Chi Ids, Doug Fennel 1, Allen Hoyt, and Diane Ray.
Technical assistance was also provided by Systems Research and Development:
Deborah Staves, Mary Williamson, and Susan McDonald; and by Northrop Services
Inc., Environmental Sciences: Miriam Gattis and Lorrie Godley.
XI1
-------
PRELIMINARY INDOOR AIR INFORMATION ASSESSMENT
PREFACE
This document has been prepared by the U.S. Environmental Protection
Agency's Office of Research and Development (ORD) to serve as the scientific
basis for the development of a coordinated multi-disciplinary research program
which focuses on risk reduction from indoor air pollution. The Agency's
Science Advisory Board (SAB), in reviewing the Agency's research program in
August, 1986, concurred that such an assessment would be useful in identifying
gaps in knowledge, methodology or analysis, which can be filled by research.
The purpose of the document is to assemble and assess the world's scientific
literature on indoor air pollution, and to summarize this information in a
manner that permits identification of knowledge gaps that need to be addressed
through research. A further purpose is to contribute to an analysis of indoor
air problems which will assist in formulating an effective Agency policy aimed
at reducing public health risk by reducing exposure to indoor pollutants.
This iteration of the preliminary assessment has been produced in a narrow
time-frame, from October, 1986 through April, 1987. Contributors from offices
and laboratories throughout ORD have participated in its writing and review.
An earlier iteration was reviewed by a group of twenty expert scientists in
indoor air assembled by Harvard in January, 1986, and was reformatted and
refined, in the light of expert opinion, and the availability of new informa-
tion into the present document. One of the issues specifically remaining to
be addressed is the problem of determining exposure to indoor pollutants, which
would permit a clearer estimation of the risk that various pollutants, or pol-
lutant mixtures, pose to the public. A workshop on the problem of estimating
exposure to indoor air contaminants is planned as part of this process. The
assessment will also be available to public scrutiny and review. A public
information document which draws on the information in this assessment, is
also projected.
xm
-------
The present preliminary assessment is divided into an introductory chap-
ter, which includes a history of the indoor air efforts of the Agency and a
profile of the ongoing evolution of policy, and four chapters, summarized
below.
Pollutant category chapter
Summarizes scientific informa-
tion on a chemical-specific
basis
Building systems chapter
Addresses indoor problems
across building-related param-
eters
Welfare effects chapter
Addresses economic aesthetic
and other effects not directly
related to health
Source category chapter
Summarizes pollutant informa-
tion by source, and leads into
research strategy and plan,
which for practical reasons is
organized by source, as well as
by generic concerns spanning
several sources.
xiv
-------
1. INTRODUCTION
1.1 HISTORICAL BACKGROUND
When the Clean Air Act was passed in 1970, the air pollution problems of
greatest concern to the nation were out-of-doors. The "mounting dangers to the
public health and welfare," as described by the Congress in Title I, were per-
ceived to be caused by "urbanization, industrial development, and the increasing
use of motor vehicles...." Consequently, the law that was intended to protect
and enhance the quality of the nation's air resources gave the U.S. Environ-
mental Protection Agency (EPA) authority to control a wide variety of air emis-
sions sources and air pollutants that contributed to the degradation of ambient
air. EPA interpreted the term "ambient" to apply to outdoor air only.
The quality of indoor air was not mentioned in the law. At that time,
except for studies of specialized environments like submarines, space cabins,
and the industrial workplace, virtually no scientific research had been done on
indoor air quality. Indoor air pollution and its associated health effects
were considered neither serious enough nor pervasive enough to merit national
attention.
However, in the early 1970s, indoor air pollution received increasing
public attention when the Government instituted energy conservation measures.
During this time, formaldehyde was identified as the cause of acute irritant
reactions, primarily eye and nose irritation and respiratory distress, in
individuals living in homes insulated with urea formaldehyde foam insulation
(UFFI), and mobile homes constructed with large quantities of particleboard and
plywood. This led to additional research to assess the types and quantities of
air pollutants found in various indoor environments, all of which came to the
same conclusion: pollutant concentrations were often much higher indoors than
they were outdoors. Furthermore, when high exposure levels were coupled with
the fact that most people spend more of their time indoors than outdoors, the
risk to human health from indoor air pollution was shown to be potentially
greater than the risk posed outdoors.
1-1
-------
As the general problem of indoor air pollution was drawing more and more
public attention as a potential health hazard nationwide, a particular type of
indoor air pollutant, radon, was causing immediate concern in certain parts of
the country. Epidemiological studies of underground miners had established a
link between exposure to elevated levels of radon and the development of lung
cancer. In the late 1960s and early 1970s, EPA had investigated homes in Grand
Junction, Colorado, contaminated by uranium mill tailings, a by-product of
uranium mining. The elevated radon levels found in those homes led to the
issuance of the Surgeon General's guidelines regarding remedial action in
houses built on or with uranium mill tailings.
During the 1970s, EPA also investigated instances of elevated radon levels
in houses built on reclaimed phosphate mines in central Florida. In 1979, EPA
issued guidelines to the State of Florida for remedial action in existing homes
and for new home construction. In 1983, the Agency began to clean up, under
the Superfund program, a number of homes in New Jersey that were built on
industrial radium wastes sites.
National attention was focused on the problem of indoor radon in 1984 when
a worker at a nuclear power plant in Pennsylvania was found to be living in a
house that was contaminated by extremely high levels of radon. In this case
the radon was being emitted by the natural soil on which the house was built.
Subsequent investigations revealed that thousands of homes in the Reading
Prong, a geological formation that runs from Pennsylvania through New Jersey
and into New York, were contaminated by naturally occurring radon. Public
concern over the potential health effects of radon exposure, and the realiza-
tion that such exposures could be occurring over wide areas, led to the estab-
lishment of EPA's Radon Action Program, directed specifically at the indoor air
pollution problem caused by radon.
In fiscal years (FY) 1982 and 1983, the Congress appropriated $500 thou-
sand each year specifically for EPA to conduct research on indoor air quality.
Eventually, a total of approximately $7 million was appropriated—roughly $2
million per year in FY 1984, FY 1985, and FY 1986. Of the monies spent in FY
1984 and FY 1985, approximately $300,000 was spent to conduct research on miti-
gation technologies for radon. In FY 1986, $1.5 million, in addition to the
$2 million for research on indoor air pollution, was appropriated for radon
research. The resources set aside for radon research were supplemented by
$200,000 in FY 1984, $200,000 in FY 1985, and $3 million in FY 1986 for program
activities associated with radon.
1-2
-------
EPA's Indoor air research program has recently been reviewed by the
Agency's Science Advisory Board (SAB). In August 1986, EPA asked the Indoor
Air Quality Research Review Panel (IAQRRP) of the SAB to review the Agency pro-
posal for assessing the current state of knowledge of indoor air pollution, and
to review the relevance of ongoing and proposed Agency research projects
related to indoor air pollution.
The SAB concluded that the indoor air research program being conducted by
EPA was of high quality and had contributed to the overall understanding of
indoor air pollution. However, the SAB felt that EPA's program at the time of
the review was lacking a clearly stated policy objective and a clearly focused
program for research support. The SAB recommended that as a preliminary plan-
ning step, EPA rapidly conduct an assessment of available information and pre-
pare a research strategy and plan based on that assessment.
1.2 THE EMERGENCE OF AN INDOOR AIR PROGRAM AND POLICY
Subsequent to the SAB review of EPA's Indoor Air Program, Title IV of the
recently passed Superfund Bill (PL-99-499), the Radon Gas and Indoor Air
Quality Research Act (RGIAQRA), came to provide clearer Congressional direction
for EPA's indoor air program.
Section 403 of the Superfund legislation directs in part that the Adminis-
trator of the EPA establish an indoor air quality research program designed to
contribute to the understanding of health problems associated with indoor air
pollutants. Also, the statute directs that EPA coordinate with Federal, State,
local, and private sector research and development efforts related to improve-
ment of indoor air quality and assess appropriate Federal actions to mitigate
environmental and health risks associated with indoor air quality problems.
Section 403 of the statute encourages EPA to disseminate information regarding
indoor pollutant sources, strengths, and concentrations, high-risk building
types, measurement instruments, health effects, as well as recommended methods
for the prevention and abatement of indoor air pollution.
As recently stated by EPA to Congress, in response to the new legislation,
the goals of EPA's indoor air quality program will be to:
identify the nature and magnitude of the health and welfare problems
posed by indoor air pollution; and
1-3
-------
reduce the risk to human health and productivity from exposure to
indoor air pollution.
EPA's program to achieve these goals is based on the belief that non-regulatory
programs (i.e., research and development, surveys and assessments, information
dissemination, technical assistance, and training) are the most appropriate
mechanisms for responding to the problem.
Due to the complexity of the issue, EPA is beginning to adopt a dual-
program approach. Efforts will focus both on the sources of individual
pollutants and products found indoors, and on the various building types in
which pollutants are found. Indoor air pollution can be prevented or mitigated
by reductions in levels of specific pollutants emitted by specific products.
Traditionally, improvements in environmental quality have been achieved by
controlling these specific pollutants or sources of pollutants. However, there
is another facet to indoor air quality. Indoor air pollution also occurs as a
function of the ways in which buildings are designed, operated, and used.
Thus, a second way to reduce indoor air pollution is to approach it as a
buildings problem and to change the ways in which buildings function. EPA's
indoor air quality program will attempt to address the problem from both
perspectives.
The Agency's indoor air research program will be run by EPA's Office of
Research and Development (ORD). The Office of Air and Radiation (OAR) will
implement a program to develop and disseminate information and guidance to
State and local governments, private sector organizations, and the public.
1.3 PRELIMINARY INFORMATION ASSESSMENT AND RESEARCH NEEDS STATEMENT
The EPA's ORD produced a draft document which was a preliminary assessment
of available indoor air pollution literature. This draft comprised a synthesis
of available information in the form of a summary document, which responded to
the concensus of both Congress and the SAB. At EPA's request, Harvard Univer-
sity organized a workshop with 20 of the world's leading researchers in indoor
air pollution, rapporteurs from Harvard's faculty, and twelve EPA researchers
as observers. The purpose of this workshop was to examine the draft assessment
critically for gaps in information, and for the methodology of approach.
Thoughtful and constructive suggestions from the workshop participants initi-
ated a change in the format of the presentation of information. The new format
1-4
-------
presents pollutant categories and examines each from the standpoint of charac-
terization, occurrence and sources, monitoring, exposure, health effects, and
mitigation. A summary delineates the important knowns and unknowns for each
pollutant category, which then point to research needs. Participants in the
Harvard workshop also suggested that while specific exposure assessments were
at present not possible for most pollutants, crude exposure estimates by source
prevalence and use patterns were an important first step at risk estimation
from indoor pollutants. Since building design and ventilation encompass many
pollutant categories, as well as mitigation of exposure, workshop participants
suggested that it should be discussed in a separate chapter (see Summary of
the Harvard Review Attachment A of this document).
The purpose of this document is to assemble and synthesize information on
indoor air pollution from work being done by the research community within EPA
and other agencies, State and local governments, universities, research insti-
tutions, and individual scientists throughout the world. The objective of this
volume is to give an overview of available information and to provide an
assessment that is intended to serve a number of purposes. Primary among these
purposes are 1) recognition of the multidisciplinary nature of the indoor
problem, and 2) an identification of research needs for responsible Government
agencies and the scientific community at large. With the integration of
exposure and health effects data, indoor air problems can be prioritized.
Integration of effort among the engineering, monitoring, and health effects
laboratories can then provide a research strategy to address specific concerns,
while redundancies of effort become evident.
Through this assessment EPA can begin to identify those questions which,
when answered by research results, will allow EPA and other Federal agencies
to better educate the public about available options regarding indoor air qua-
lity. Individual homeowners as well as designers, builders, manufacturers, and
maintenance professionals will be made aware of the sources of contamination,
as well as the relative hazard that indoor pollution may represent. Other
benefits of this assessment, and the resulting research strategy, include new
methods of assessing problems, the rethinking and refinement of monitoring
techniques, the definition of health parameters affected, and the identifica-
tion of specific populations at risk. Public awareness of potential hazards
associated with improper use and storage of chemicals (which can pollute indoor
spaces) for example, must be increased. Means to avoid or mitigate potential
1-5
-------
problems can be described, so that affected persons can make intelligent
choices regarding the contaminants to which they are exposed. State and local
health organizations can play a vital role in disseminating information .
packages developed by the Agency which will be the end product of the new
policy on indoor air quality.
This Volume is organized so as to begin answering the following questions:
1. What pollutants exist in indoor environments? (Pollutant
Characterization; Monitoring)
2. Where are they? (Characterization of Indoor Spaces)
3. Where do they come from? (Sources)
4. What is the human exposure to indoor pollutants? (Exposure)
5. What are the effects on the human receptor? (Health Effects)
6. Who is affected? (Health Effects)
7. What are the health hazards? (Health Effects)
8. What are the welfare concerns? (Welfare Concerns)
9. How can exposure be mitigated? (Mitigation)
The preliminary assessment presented here will undergo review by experts
in indoor air quality and technical experts within EPA and other agencies of
the Federal Government. It will continue to be revised in accordance with
these reviews and, after completion, will be made available to the general
public for examination, by Federal Register notice, prior to a public review
workshop. In that this information on indoor air pollution is of interest and
importance to a broad audience, EPA plans to ensure that it will continue to be
reviewed and revised in an open public forum as new information and analyses on
indoor air pollution become available.
1.4 SUMMARY OF WHAT IS KNOWN ABOUT INDOOR AIR POLLUTANTS
Indoor pollutants can be grouped into categories and the information
regarding them summarized in relatively brief terms. For a number of indoor
air contaminants the source is known, while for some it is surmised. Health
effects are also known for some specific pollutants, from human clinical, epide-
miologic or animal studies focused on occupational or ambient air situations.
For certain classes of compounds only general effects are known and dose-
response parameters have not been determined. Specific monitoring information
1-6
-------
information for most indoor pollutants is still relatively sparse, although
active and some passive monitoring methods have been developed for some specific
pollutants or pollutant classes. Development of risk estimates for specific
pollutants is not feasible presently, and may not be of practical value. This
is especially true for high risk populations, whose responses to indoor pol-
lutants must be extrapolated from human clinical, epidemiologic or chamber
studies. Almost no information is available about the interactions of the
various indoor pollutants, either chemically, in the mixture that individuals
breathe, or their additive or synergistic effects within the body. Estimation
of health hazards from specific indoor situations or sources may be a more use-
ful approach than trying to derive risk estimates since it can give the inhabi-
tant of the indoor space options to choose in mitigation of hazard and promotion
of healthful environments.
Combustion products can be divided into combustion gases, principally
nitrogen dioxide, carbon monoxide and sulfur dioxide, combustion particles, and
combustion organics, including PAH and formaldehyde. The indoor sources for
all of these contaminants are either combustion appliances such as gas stoves,
water heaters, dryers or gas or kerosene space heaters or environmental tobacco
smoke (ETS). Fairly specific information is available concerning the three
major combustion gases, less is known regarding particles, especially about
their interaction with combustion organics.
While NO,, is usually monitored passively over a one to two week period, it
can be monitored by this mode for as little an 8-hour day. Active monitoring is
typically done hourly, but can be done in real time. This is important since
health effects appear to be more dependent on short-term exposure to high levels
of N02> rather than by chronic, low level exposures. Personal exposure monitor-
ing is also possible, which together with diaries detailing personal activity,
can pinpoint high-level exposure peaks more accurately. Chamber studies indi-
cate that exposure to concentrations greater than 0.3 ppm N02 for 15 minutes
can decrease pulmonary function in asthmatics, while levels greater than 2.5
ppm for longer periods of time increase susceptibility to infection in animals.
Exposure to between 1 to 2.5 ppm has been shown to decrease pulmonary function
in children and perhaps adults. Evidence from animal studies, and some epide-
miological studies, indicate an interactive response with other pollutants,
resulting in decreased immune response and changes in anatomy and function of
the lung. Actual measurements of NO, during cooking with gas or during use of
3
kerosene heaters have shown levels exceeding 1 mg/m (0.53 ppm). Census data
1-7
-------
show that 95,802,000 people cook with gas stoves averaging 4 hours per day,
while 7,022,000 individuals are exposed to kerosene heater fumes averaging
2 hours per day. These numbers allow a generalized estimate of the potential
hazard from N02 to those people exposed, but lack of dose-response and specific
concentrations and exposure times, as well as variation in susceptibility do
not permit a risk estimate for any of the health effects.
Carbon monoxide is monitored presently by use of personal exposure moni-
tors which together with activity diaries can yield good estimates of personal
exposure. Relatively few studies exist which have monitored CO concentrations
and exposure times, however. The existing studies indicate CO exposure in
vehicles (cars) in commuter traffic, as well as from ETS and faulty or unvented
combustion appliances. The extent of population exposure to the latter is not
known, since symptoms of moderate to high level concentrations mimic many
illnesses, and CO poisoning symptoms are largely not attributed to CO, but to
illness such as flu, or food poisoning.
CO attaches avidly to hemoglobin in the human bloodstream, forming com-
pound carboxyhemoglobin (COHb); exposure to CO can be measured as the percent
of hemoglobin which is in the COHb form. COHb cannot transport oxygen to tis-
sues, and tissues with high oxygen demand, such as the brain and heart, are most
rapidly affected by the resulting oxygen deprivation. At 2.9 to 4.5 percent
COHb patients with angina pectoris have an aggravation of their attacks in fre-
quency and duration. At levels above 5 percent COHb, healthy males show a
decrease in work capacity, while at 10 percent COHb patients whose cardiopul-
monary function is compromised show an exacerbation of their dysfunction.
Between 10 to 30 percent COHb, healthy adults show decreased mental alertness,
suffer headaches and exibit flu-like symptoms. At COHb levels greater than
60 percent, asphyxiation occurs. If death does not follow, permanent brain
damage from anoxia can result.
Monitored concentrations of CO relate to percent COHb through a relation-
ship developed by Coburn et al. (1965), which permits calculation of blood COHb
as a function of time, considering appropriate physiologic and physical fac-
tors. Up to 60 ppm CO have been measured inside cars stopped in traffic jams,
while up to 18 ppm were measured in public garages. With faulty combustion
appliances concentrations in excess of 100 to 200 ppm have been measured.
There is presently no way of determining the number of people living in
"sick" homes with faulty or unvented combustion appliances. Even when properly
1-8
-------
vented, homes that are very weather-tight may have down-drafts through the
chimney which can cause dangerous levels of CO (and NC^).
Sulfur dioxide is also a combustion gas, but is derived from burning fuels
with high sulfur content such as some natural gas and kerosene or oil. It is
monitored passively, usually measured over 12 hours, and not actively monitored
indoors.
Chamber studies have shown that concentrations of less than 0.75 ppm
administered for less than one minute have caused a decrease in lung function
in infants and the elderly, in synergism with inhaled particles. Lung airway
resistance doubled in the affected individuals at this exposure. Monitored
concentrations with use of low-sulfur fuel in unvented kerosene heaters has
determined concentrations of 0.1 to 2.0 ppm/12 hour average. Acid aerosol
emissions from kerosene heater use can be extremely high. The number of people
exposed to S02 indoors and the extent of their exposure is not known at present.
Inhalable particles are produced in cigarette smoke and by combustion
appliances. At present there are no indoor passive monitors for particles, but
there are active fixed and personal monitors which may be used indoors. Obtain-
ing adequate sample sizes with such monitors is difficult, but larger samplers
or larger-volume samplers are often too bulky and obtrusive for indoor use.
Cancer is the most serious health consequence that has been associated
with inhalable particles. Soot itself is carcinogenic, and pro-, co- and
frankly carcinogenic polycyclic aromatic hydrocarbons are adsorbed to fine par-
ticles which are inhaled deeply into the lungs where they can deliver their
dose. Particles themselves may play a synergistic or additive effect in the
carcinogenesis of these adsorbed compounds. Composition of these entities
varies with fuel and burning conditions and no dose-response information is
known. Another effect is irritation of respiratory tissue and the eyes,
at variable concentration, depending in part on particle composition. Decrease
in lung function occurs at concentrations of respirable particles greater than
300 ug/m , alone, or in synergism with S02 or other gases.
Monitored peak concentrations from combustion appliances indoors have
reached 119 ug/m3. Concentrations of particles in ETS can increase this value
several-fold. While the aggregate exposure to particles from combustion appli-
ances is not known, it has been calculated that 86 percent of nonsmokers exposed
to ETS (124,700,000 people) experience a peak exposure of 1.43 mg/day.
1-9
-------
Many organic products of incomplete combustion enter the indoor environment
from cigarette smoke and from unvented combustion appliances. Only some of
these organics have been identified, fewer have been quantified. A large number
of PAH have been measured by active 8 hour-average monitoring. One to 3 ng/m
large molecules and 20 to 50 ng small molecules have been collected over this
time period from ETS. However, dose-response information is not known. Among
the health effects determined in epidemiologic studies are cancers, cardiovas-
cular effects and irritation of mucous membranes. Animal data show a decrease
in immune function with exposure, and a role in the etiology of atherosclero-
sis. Exposure to combustion organics from unvented or faulty appliances is not
known, and exposure to ETS is included in the discussion on particles.
Formaldehyde is a combustion organic that is also a compound released from
sources such as building material and furnishings. It will be discussed under
the gas-phase organics. Its primary source under combustion organics is ETS and
its primary effect is cancer, with irritation of mucous membranes and allergic
reactions less lethal but more common.
Among the noncombustion particles of concern in indoor air are asbestos,
dusts, sprays, and cooking aerosols. Biologic particles will be discussed
separately.
Asbestos fibers derive from asbestos cement and insulation used in some
residences, schools and public buildings built prior to 1970. It was used to
insulate steam-boilers and steam pipes. As this insulation ages it can become
friable, and fibers can be entrained into breathable air. Vehicles can also be
a source of asbestos from wear on brake linings. Since asbestos fibers fall
into the particle classification, they can be collected by active samplers and
by PEM, but require special electron microscope analysis for fiber counts.
Little measurement has been done to determine extent of exposure, except in a
few public buildings, and in schools. Monitoring performed in schools throughr
3
out the U.S. can give ranges of concentrations found (20 to 4500 ng/m ; approxi-
mately 0.0006 to 0.15 fibers/mL, ±4 f/mL). The number of children potentially
exposed for 6 hours per day in older schools is 19,783,000. Approximately
33,387,000 people are potentially exposed to asbestos in steam-heated homes
built prior to 1970 for 10 to 24 hours per day. Commuters in cars and buses,
who number about 217,500,000 and at least an additional 38,100,000 truck
drivers, are exposed to asbestos fibers from brake linings.
Dusts, sprays and cooking aerosols result from personal activities and
vary in their concentrations depending on the duration and extent of such
1-10
-------
activity. Practically no monitoring efforts have been made to distinguish and
quantify these entities and methods development is still needed. Since the
identity of the components of such mixtures is not known, neither exposure nor
health effects data are available.
Noncombustion gas-phase organics (also known as VOCs or volatile organic
compounds. This is a misnomer since many have rather high boiling points and
are not very volatile; many semi-volatile organic compounds are included in
the designation) constitute a broad class of compounds of varying reactivities
and physical properties. Their sources are paints, stains, adhesives, dyes,
solvents, caulks, cleaners, pesticides, and building materials and furnishings
found in homes, offices, public buildings, and vehicles. No passive monitoring
is feasible for these gas-phase organics but active monitors for indoor sam-
pling, usually for 24-hour averages, are available, as are PEMs. More than 900
different compounds have been identified in indoor environments and many more
remain unidentified. Health effects for some individual compounds are known,
but concentrations at which such identified health effects have been noted have
been much greater than those generally measured in indoor air. VOCs occur in
mixtures, however, whose aggregate effect is not known. However, many gas-phase
organics have been measured indoors at concentrations greater than that in the
ambient air, where the concentrations are regulated. Effects range from sensory
irritation to behavioral and neurotoxic effects, hepatotoxic effects, and
cancer. No epidemiologic work has been performed to relate concentrations with
such health effects. Concentration-response relationships are known for
individual compounds, and cancer risk numbers have been calculated for some
compounds by the Carcinogen Assessment Group of the U.S. EPA. Concentration-
response effects for aggregate mixtures of commonly found gas-phase organics
in office buildings in Denmark have been determined in chamber studies, with
5 to 25 mg/m causing synergistic behavioral changes in sensitive adults. How-
ever, no health effects data is available for assessing the effect on neuro-
logic, liver functions, respiratory system changes or for the risk of various
cancers from the effects of aggregate mixtures of gas-phase organic compounds.
Estimates of exposure may be possible for individual compounds such as formal-
dehyde, in specific indoor settings, but because of the multiple sources, the
ubiquitousness of the compounds, and the lack of monitoring data generally, no
real estimate of exposure is possible at this time.
1-11
-------
Formaldehyde (HCHO) is a gas-phase organic compound that is a combustion
product but also is emitted from urea formaldehyde foam insulation (UFFI),
resins from particle board and plywood, furnishings, carpets and upholstery.
Because of very high concentrations emitted from UFFI, considerable attention
has been focused on this contaminant, and it has been identified as a problem
in certain environments, especially in manufactured homes, due to the prevalence
of sources there. Passive monitors are available. These required 5 to 7 days
monitoring time. Both active stationary samplers requiring two hours averaging
times, and PEM are available. Health effects such as irritation of mucous
membranes have been shown to occur in chamber studies at 0.1 to 0.2 ppm. Indi-
viduals sensitized to HCHO react allergically at concentrations of less than
0.1 ppm, however. EPA has recently reclassified HCHO as a possible human
carcinogen, based largely on data from animal studies. Concentrations measured
in mobile homes range from 0.03 to 8 ppm, with 9,279,000 people potentially
exposed in these environments for 10 to 24 hours per day. Concentrations in
offices, public buildings, schools and homes can also range into irritatory
levels, especially after remodeling or after installation of new furnishings and
carpets. Formaldehyde is a suspected actor in sick building syndrome. The
extent of exposure in environments other than mobile homes has not been
estimated.
Radon emanates from radium-containing soil and enters residences through
cracks or other openings in basements, or through crawl spaces. It is passive-
ly monitored. Concentrations measured in various parts of the U.S. have
typically ranged from 0.5 to 2 x 10 pCi/L. The U.S. EPA considers a level of
4 pCi/L to be the action level at which mitigation should occur. An estimated
93,760,000 people in the U.S. live in states with an average radon concentration
greater than 4 pCi/L and are exposed to levels at or greater than this. The
U.S. EPA has published risk estimates which predict 5,000 to 20,000 excess
cancers/year for exposures presently projected.
Biological contaminants represent a very diverse group of substances in
indoor environments. They include infective agents such as viruses, bacteria
and molds, capable of causing infectious diseases. Some bacteria and molds
produce highly toxic substances which have effects on many systems in the human
body, and thereby constitute a threat aside from virulence. In addition these
viable agents can cause allergic reactions, as can mold and bacterial spores,
pollen, insect, acarid, and arachnid body parts and excreta, and animal and
human dander. While a variety of passive and active methods for monitoring
1-12
-------
biological contaminants exist, passive monitors do not collect representative
samples, and active monitoring methods have not been standardized. Without
standardized methodology, results from different researchers are not compara-
ble. Baseline data to distinguish normal levels from contamination are not
available. Since a clear distinction of what constitutes contamination by the
various biologic agents has not been made, the extent of exposure cannot be
estimated.
Pesticides are organic chemicals applied in and around buildings, primarily
to control insects. They can enter indoor spaces through cracks and other
openings in foundations, or be directly introduced. They are by definition
poisonous substances, affecting the nervous system, the liver, or the reproduc-
tive systems. Allergic reactions have been documented. Dose-response rela-
tionships for toxicities for individual pesticides are known in some cases from
industrial and animal studies. Since they are organic substances, primarily in
vapor-phase, they are monitored actively by the same methodology as other gas-
phase organic compounds, specifically semi-volatile organic compounds. Concen-
trations measured through the NOPES program have detected aggregate concentra-
o
tions in the range of 1.7 to 15 ug/m . Cumulative effects of different
pesticides and interactions with other organic compounds have not been deter-
mined. Monitoring has not been extensive so that exposure estimates for pesti-
cides are not possible at this time.
Nonionizing radiation is produced by all electric conductors which gener-
ate electric and magnetic fields around themselves whenever a current flows
through them. Such conductors are ubiquitous. Animal and epidemic!ogic studies
have shown behavioral and reproductive effects, and possible cancers resulting
from the influence of low level electric and magnetic fields. The dose-response
is not known. Such fields can be detected by passive monitors. Field strengths
2
measured are in the range of 1 to 2 m gauss for magnetic fields and 10 mV/m
for electric fields. All individuals who use electricity or come near electric
conductors are exposed to such nonionizing radiation, but the effects of such
exposure on the human population is unknown.
Environmental tobacco smoke is a complex mixture of pollutants whose
source is primarily cigarette smoking. Various components of this mixture
have been actively monitored and include respirable suspended particles, CO,
nicotine, nitrogen oxides, acrolein, nitroso-compounds and benzo(a)pyrene.
Health effects from ETS have been extensively discussed in both the 1987
1-13
-------
Surgeon General's report and the 1986 report from the National Academy of
Sciences. They include cancer, cardiovascular effects, increased suscepti-
bility to infectious diseases, chronic and acute pulmonary effects in children,
mucous membrane irritation and allergic response. Approximately 124,000,000
people are exposed to up to 1.43 ug/day of ETS per day. Dose-response effects
are not known.
Since dose-response effects for indoor pollutants and even concentration-
response effects are generally not known, it should be emphasized that risk
estimates for populations are not presently possible. This is espeically true
since individuals are exposed to a wide variety of complex mixtures of
pollutants whose interactive risks are unknown. It is possible to state
numbers of people potentially exposed to an indoor situation or a source that
poses a hazard, and to suggest ways to mitigate the source or the potential
exposure for people. Individuals can then be presented with options for
avoiding hazards or choosing or avoiding activities or sources that impinge on
their health.
Table 1-1 summarizes the information described in this narrative in
abbreviated form.
1-14
-------
TABLE 1-1. SUMMARY: WHAT IS KNOWN ABOUT INDOOR POLLUTANTS
Pollutant
Source
Health Effect
Concentration
for Effect
Monitoring Indoors
Passive/
Time
,_
Active/
Time
PEN
Concentration
Measured/
Activity
Estimated
Number
of People
Exposed
Estimated
Length of
Exposure/
Day
M02
Combustion
Appliances
ETS
Decreased pulmonary func- 1 ppffl N02 1 day (8 hr) 1 hr
tion in asthmatics 15 min (minimum)
i
en
CO
Yes
Increased susceptibility
to infection
>2.5 ppm N02 1-2 weeks Real time
1-2.5 ppm N02
Combustion
Appliances
ETS
Infiltrated
auto exhaust
Effect on pulmonary func-
tion in children, perhaps
adults
Synergistic detrimental Varies
effects with other
pollutants
Animal studies indicate
decreased immune capa-
bility, changes in
anatomy and function
of lung
Aggravation of angina in 2.9-45.% COHb No
patients
Decreased work capacity 5% COHb
in healthy adult males
Headaches, decreased 10-30% COHb
alertness, flu-like
symptoms
Exacerbation of cardio- 10% COHb
pulmonary dysfunction
in compromised patients
Asphyxiation >60% COHb
Yes, for
outdoors;
too bulky
indoors
Yes
Up to 1 mg/rn3
during cook-
inf with gas
Up to 1 •g/m3
use of kero-
sene heater
Up to 18 ppn
in public
garages
Up to 60 ppm
in vehicles
in traffic
jams
100-200 ppm
with faulty
appliances,
no venting
95,802,000
7,022,000
4 hrs
2 hrs
Unknown
Unknown
Unknown
Unknown
(continued on following page)
-------
TABLE 1-1. (continued)
i
01
Pollutant
S02
Combustion
particles
Combustion
organics:
PAH's
Source
Combustion
of fuels
containing
sulfur
(e.g..
kerosene,
oil
Combustion
Appliances
ETS
Combustion
Appliances
ETS
Kerosene
Heaters
Concentration
Health Effect for Effect
Decreased lung function (0.75 ppm/
in infants, elderly (in 1 min)
synergism with particles)
increased (doubled) air-
way resistance
Animal studies show 1 ppm
decreased lung
function
Cancer (soot, PAH absorbed Dose-response
to particles) unknown
Irritation or respira- Varies
tory tissues, eyes
Decreased lung function, >300 ug/m3
alone and synergistically
with S02, other gases
Cancer, irritation, car- Dose-response
diovascular effects unknown
Animal data show
decreased immune
function, athero-
sclerosis etiology
Monitoring Indoors Concentration
Passive/ Active/ Measured/
Time Time PEM Activity
Yes Yes No 0.1 to 2.0 ppm/
12 hr average
with unvented
kerosene heater
use, low sulfur
fuel
No Yes Yes 119 ug/m3
No Yes Yes 1-3 ng/m3,
8 hr large
molecules
20-50 ng/m3,
small
molecules
Estimated
Number
of People
Exposed
Unknown
Aggregate
combustion
appliance
exposure
unknown
ETS: 86%
of non-
smokers
exposed
(124,700,000
at 1.43
mg/day,
peak
Aggregate com-
bustion
appliance
exposure
unknown
124,000,000
exposed to
1.43 mg/day
ETS residue
Estimated
Length of
Exposure/
Day
Unknown
1-10 hrs
Unknown
2-10 hrs
(continued on following page)
-------
TABLE 1-1. (continued)
Pollutant
Combustion
organics:
Formaldehyde
(see below)
Noncombustion
particles
Asbestos
Ousts
Sprays
Cooking
aerosols
Noncombustion
gas-phase
organics (VOC)
Formaldehyde
Source
Combustion
Appliances
ETS
Asbestos
cement,
insulation
Floor tiles
Brake linings
Personal
activity
Paints
Stains
Adhesives
Dyes
Solvents
Caulks
Cleaners
Pesticides
Building
materials
UFFI
Particle
board
Plywood
Furnishings
Carpets
Upholstery
Health Effect
Irritation
Cancer
2 fibers/m3/50 yrs
Unknown; can range from
irritation to cancer
Irritation
Neurotoxic/Behavior
effects
Hepatotoxic effects
Cancer
Irritation
Allergy
Cancer
Estimated
Monitoring Indoors Concentration Number
Concentration Passive/ Active/ Measured/ of People
for Effect Time Time PEM Activity Exposed
No Yes Yes
Unknown No No No Unknown Unknown
Varies No Yes Yes Individual
Dose- response 24 hrs compounds
to aggregate
VOCs: 5-25
mg/m3
Dose- response
known for in-
dividual
organics
Risk estimated
for individual
compounds
0.1-2.0 ppm Yes Yes Yes 0.03-8 ppm in 9,279,000
5-7 days 2 ppb-hrs mobile homes in mobile
>0.1 ppm homes
Dose- response
unknown
Estimated
Length of
Exposure/
Day
Unknown
10-24 hrs
(continued on following page)
-------
TABLE 1-1. (continued)
Pollutant
Radon
Biological
contaminants:
Viruses
Bacteria
Molds
Insect and
arachnid
excreta
Pollen
Animal and
human
dander
Pesticides
Nonionizing
radiation
Electric and
magnetic
fields
ETS
Source
Soil
Well Water
Some building
materials
Outdoors
Humans
Animals
Moist
building
areas
Outdoors
Indoor
spraying
Pets
Moth
control
Electric
conductors
Cigarette
smoking
Health Effect
Cancer
Infectious diseases
Allergy
Intoxication
Neurotoxicity
Hepatotoxicity
Reproductive effects
Possible cancer,
behavioral and
reproductive
effects
Cancer
Irritation to mucous
membranes
Chronic and acute
pulmonary effects in
children
Cardiovascular effects
Monitoring Indoors
Concentration Passive/ Active/
for Effect Time Time PEM
Indoor dose- Yes No No
response
unknown
Varies; mostly Yes Yes No
unknown (useless) (not
standard-
ized)
Dose-response No Yes Yes
for individual
pesticides is
known
Dose-response Yes No No
unknown
Dose-response Some components: particles,
unknown CO, nicotine, NO acrolein
Varies nitro-compounds Benzo-a-pyrene
quantified
Dose- response
unknown
Dose- response
unknown
Estimated
Concentration Number
Measured/ of People
Activity Exposed
0.5-2 x 106 2,500,000
PCi/liter
Varies Unknown
Varies Unknown
1.7-15 ug/m3
1-2 mgauss Unknown
10 mV/m3
124,000,000 2-10 hr
exposed to
1.43 mg/day
Estimated
Length of
Exposure/
Day
10-24 hrs
10-24 hrs
10-24 hrs
10-24 hrs
-------
2. POLLUTANT CATEGORIES
2.1 INTRODUCTION
Individuals breathing the air within indoor spaces are exposed to complex
mixtures of pollutants whose total impact on health and well-being is presently
not known. The interactions that occur chemically and physically among the
various pollutants are not well characterized; their combined interactions on
and with materials found indoors are also not well understood. Their syner-
gistic, antagonistic, and additive effects on health parameters are not known.
Pollutants are often identified with a specific source, that is, a gas stove,
and individual pollutants from such a source are measured and correlated with
source presence or use, but other emission components are not measured. Most
of the information relating to exposure and health, especially that coming from
animal or clinical studies, derives from investigation of single pollutants.
For these reasons the information known about indoor pollutants is here organ-
ized into single pollutant categories; when a complex mixture such as environ-
mental tobacco smoke (ETS) has been studied as a source entity, it is presented
as such.
The chapter is organized to give an overview of published information
about selected pollutant categories as follows: combustion gases, including
carbon monoxide (CO), nitrogen dioxide (NOp), and sulfur dioxide ($02); parti-
cles and combustion organics, including gas phase organics and polycyclic
aromatic hydrocarbons (PAH), among others; ETS; noncombustion particles such
as asbestos fibers, dust, spray and cooking aerosols; noncombustion gas-phase
organic compounds (also called volatile organic compounds or VOCs); radon; bio-
logical contaminants; pesticides; and nonionizing radiation. For each of these
pollutant categories, if such information is available, the chapter discusses
occurrence or sources, estimates of exposure, including exposure levels and
distributions among the population, and a brief discussion of monitoring tech-
niques and the factors that affect monitoring, such as averaging times. Health
effects, which are the driving force for a consideration of indoor pollutants,
2-1
-------
are summarized next, including results from in vitro, animal and human studies,
concentration-response relationships where known, and the length of exposure
known to produce effects.
Since the goal of the Indoor Air Quality program is to reduce exposure
for the total population, this analysis is followed with a discussion of miti-
gation and control options for each pollutant category. Finally, the informa-
tion for each pollutant category is summarized to give an indication of the
major known and unknown areas of information, which summary leads to an indica-
tion of what the needs for future research will be.
Accurate estimation of human exposure is essential for determination of
risk posed by pollutants, as well as for the design and implementation of
effective exposure reduction methods. In order to determine exposure in a
useful way, a clear understanding of the meaning of the term, and related
terms, is needed. A distinction must be made among the terms concentration,
exposure, and dose. Concentration of an air pollutant is the mass or amount
of that material per unit volume of air (e.g., milligram per cubic meter).
Monitoring devices measure pollutant concentrations, which may allow accurate
exposure estimates to be projected. Nominal exposure is defined here as the
contact between the surface of the body (skin, respiratory tract, gastro-
intestinal tract) and the pollutant. Two events must occur simultaneously to
constitute a nominal exposure:
1. A pollutant of a certain concentration is present at a specific
location for a measurable time, and
2. A person is present at that location during the given time
period. (Duan, 1982; Ott, 1985; Sexton and Ryan, 1987).
The key distinction between a concentration and an exposure is that expo-
sure is defined by the presence of persons in the contaminated environment.
Monitored concentrations can act as a surrogate for exposure, on the assumption
that they represent concentrations experienced by people. It is an assumption
whose utility is especially important for source mitigation considerations,
since reduction in concentration of pollutant implies a reduction in exposure.
The distinction between exposure and dose is also important. While
exposure is defined as the pollutant concentration at the interface between the
body and the external environment, dose is defined as the amount of pollutant
2-2
-------
that crosses this interface and reaches the target tissue. Factors that affect
the size of the dose delivered through the respiratory system are respiration
rate, volume, mode (mouth or nose breathing), uptake, metabolism, and clear-
ance. The relationship between exposure and dose has not been established for
most air pollutants. The distinction between the terms can be made clear by
considering two individuals in the same environment, one at rest, the other
exercising vigorously. They both have the same nominal exposure, but clearly
the second is receiving a higher dose (Sexton and Ryan, 1987).
With regard to the health consequences of exposure to indoor air pollu-
tants, three parameters of exposure are especially important:
1. Magnitude or pollutant concentration
2. Duration or length of exposure
3. Frequency of exposure.
Magnitude is assumed to be directly proportional to dose, and is an
important aspect of exposure, but duration is likewise important. Just as high
level or peak exposures are likely to impact the human system differently from
low level exposures, the length of exposure (5 minutes versus an hour or a day)
is likely to be a determinant of tissue injury. Frequency, too, has an impact,
since repeated insult to tissues is likely to do more damage than a single
encounter.
Measurement of exposure most often depends on the capabilities of monitor-
ing instruments. Real-time exposure monitors, together with activity diaries,
can provide a continuous record of exposure over a time period, and indicate
periods of no exposure to a pollutant, as well as frequency, duration and
magnitude of exposures. These instruments and the results they provide are
often complex, and the data is difficult to analyze. Summaries of data, which
average concentrations over time (e.g., average concentration per hour or per
day) are more commonly used. Exposure profiles obtained from personal exposure
monitors combined with activity pattern diaries give the most complete picture
of exposure, including frequency and extent of peaks. Integrated exposures, on
the other hand, give a single value obtained from integration of the function
(C-t) over a specified time period. This method does not provide informa-
tion about the pattern and severity of short-term, peak exposures. "Average
2-3
-------
exposure" is a specialized form of integrated exposure defined as the integral
of C-t over the averaging time t , divided by t .
i a a
Individual or personal exposure as measured by time-activity patterns and
personal exposure monitors (PEM) can vary substantially from person to person.
Estimates of group exposure may require many such individual exposure measure-
ments. From a public health perspective, it is important to know the number
and characteristics of the group of people at risk. Such determinations give
estimates of "population exposure" or the aggregate exposure for a specified
group of people. Since exposures are likely to vary substantially between
individuals in a group, the distinction of personal exposures within the group,
including the average values, and the variance, should be specified. The upper
tail of distribution, which delineates those individuals who experience the
highest pollutant concentrations, can be essential to health risk assessment.
The lower tail of distribution can identify especially sensitive populations.
At the present time, monitored information which would allow exposure
calculations for many indoor pollutants is still too sparse to give good
predictions of population exposures. Risk estimates for populations exposed to
indoor air pollutants are therefore premature. Information does exist, however,
regarding source frequency and usage patterns for many source categories of
indoor pollutants. It is therefore possible, at present, to give a crude
estimate of the number of people exposed to a range of pollutant concentra-
tions, over differing averaging times. While this does not permit the calcula-
tion of risk numbers, it does give a sketch of possible risk that would be
helpful towards ranking pollutants by the severity of their impact on the
population as a whole. Table 2-1 describes such a profile for selected
pollutant sources.
2.2 COMBUSTION GASES
2.2.1 Introduction
Combustion research has demonstrated that the combustion of fossil fuels
(e.g., oil, kerosene, coal, natural gas) and vegetative sources (e.g., wood,
plant material) can result in the production, and usually the emission, of a
very complex mixture of organic and inorganic gaseous and particulate pol-
lutants. In addition, the complex mixture that results from incomplete combus-
tion can contain a variety of other volatile and semivolatile, polar and
nonpolar chemical products.
2-4
-------
TABLE 2-1. EXPOSURE PROFILE BY SOURCE PREVALENCE AND USE PATTERNS FOR SELECTED POLLUTANTS IN THE U.S.'
ro
Source
Combustion Appliances
Environmental
Asbestos
Gas-Phase
Organic Compounds
Pollutants
Emitted
CO, N02) S02) PM, HCHO
Gas Stoves
Gas Water Heaters
Gas Dryers
Gas Furnaces
Gas Space Heaters
Kerosene Space Heaters
Woodstoves
Fireplaces
CO, N02, PM, PAHs, HCHO
Tobacco Smoke
Asbestos Fibers
(in school buildings)
Pesticides, HCHO, Solvents,
Cleaners
Number of People
Potentially
Exposed
95,802,000
100,317,000
47,023,000
115,866,000
20,063,000
7,022,000
13,543,000
45,143,000
85,269,000e
19,783,0009
9,279,000h
Percent of
Total
U.S. Population
41
43
20
49
8
3
6
19
36
8
4
Estimated
Average Number
of Hours
Exposed Per Day
4
4
1
5
2
2
5
3
10f
6
10
a = Base year: 1983
b _
c _
= Average number of residents per home: 2.5
= 1983 estimated total U.S. population = 234,500,000
= Winter months
e _
f _
= Including smokers
h _
= Estimated average number of hours at home, does not include occupational settings.
= Average number of children per school x number of older schools.
= Estimated number of people living in mobile homes; extent of exposure not known for households, office buildings.
-------
The primary emphasis in this section is on three products of incomplete
combustion: carbon monoxide, nitrogen dioxide and sulfur dioxide. The first
of these is given off from any incompletely burned carbon compound. Nitrogen
dioxide is formed from nitrogen and oxygen in air at high combustion tempera-
tures, and sulfur dioxide is formed during the combustion of sulfur-containing
fuels (such as gas or oil). A large body of literature is available on these
pollutants since they are classed as regulated criteria pollutants for ambient
air. Other gas-phase products of combustion, such as some of the PAHs, are
also sources of risk due to their carcinogenicity, but specific information
such as dose-response relationships are not known. These are described further
in the particle section of this document, since they also adsorb to particle
surfaces.
2.2.2 Occurrence and Sources of Combustion Gases
Some typical indoor sources of combustion products and related contami-
nants are contained in Table 2-2. This table will serve as a basis for the
subsequent discussion of indoor combustion sources.
2.2.2.1 Unvented Kerosene Space Heaters. Unvented kerosene heaters can be
divided on the basis of flame type into two categories: blue-flame and white
flame. The color of the flame is determined by the burner design, the manner
in which air enters the burner area, and the flame temperature. Within these
two categories there are five basic types of kerosene heaters on the market.
They are convective, radiant, radiant/radiant, radiant/convective, and wickless.
The primary pollutants that have been measured from unvented kerosene
space heaters are NO , CO, carbon dioxide (C0?), S0?, and formaldehyde (HCHO)
/\ c- £.
(Consumer Products Safety Commission, 1983). In general, higher flames than
normal will increase the emission of NO, while lower flames increase CO, HCHO,
and particulate matter (U.S. Department of Energy, 1985). The S02 emission
rates depend on the sulfur content of the fuel and are not significantly
affected by changes in the flame or heater type or age of the unit (U.S.
Department of Energy, 1985).
Unvented kerosene space heaters vary widely in their production of indoor
pollutants. The variation can be attributed to the wick height, fuel type,
adjustment of primary air/fuel ratio, length of time the burner has been oper-
ating during a single use period, either over or under firing of the burner,
and the overall design of the unit (Consumer Products Safety Commission, 1983;
Lionel et al., 1986).
2-6
-------
TABLE 2-2. INDOOR AIR SOURCES AND RELATED CONTAMINANTS
Sources
Gases and Vapors
Particles
Unvented appliances
gas-fired
kerosene-fired
Ventless heaters
gas-fired
oil-fired
wood-fired
Automobile garages
H20, C02, CO, NOX, N02
(+)S02
H20, C02, CO
(+)NO , mercaptans
A.
large particles
fine particles
Aitken nuclei
(+)NOX, S02
(+)NOX, S02, HC
CO, HC, nitrogen compounds
Aitken nuclei
fine particles
large particles
B(a)P
fine particles
Aitken nuclei
Source: Woods (1983).
Traynor et al. (1983), reported on pollutant emissions from four portable
3
kerosene-fired space heaters. The tests were conducted in a 27 m environ-
mental test chamber operating at an air infiltration rate of approximately 0.4
air changes per hour. The concentrations reported were based on one-hour burn
times. The results are summarized in Table 2-3. All four heaters tested were
found to emit CO, NO, N02> and HCHO. There were no significant differences
found in emission rates with or without a warm-up period.
In a subsequent work, Traynor et al. (1984) reported on two kerosene
heaters (1 white-flame convective, I blue-flame radiant) that were operated in
the master bedroom of an unoccupied house. The test attempted to identify the
pollutants and to determine interroom transport of those pollutants over a
range of test conditions. The test conditions included doors and windows
closed; door closed and the window open 2.5 cm; door open 2.5 cm and the window
closed; door wide open (74 cm) and the window closed.
In each case the heater was operated until an 8°C increase in temperature
was achieved in the room and the increase in CO, C02, NO, and ML was recorded.
Carbon dioxide concentrations were determined to range from 2440 to 5440 ppm;
NOp levels ranged from 0.12 to 0.60 ppm. Traynor and co-workers reported
interroom transport for those test conditions with the window closed. They
2-7
-------
TABLE 2-3. POLLUTANT EMISSION RATES FROM FOUR PORTABLE KEROSENE-FIRED HEATERS
Heater/Fuel
Consumption
(kJ/hr)
New Convective
7830
7980b
7840°
Old Convective
5480,
5780°
New Radiant
8180,
8250°
Old Radiant
6640D
CO
14.5
10.3
9.1
115.1
110.5
60.2
71.7
54.0
Emission Rates
(ug/kJ)
C02 NO N02
70100
72500
70400
69000
63600
70300
68500
66200
23.7
25.2
25.3
11.1
10.9
1.4
1.2
2.1
14.1
12.5
12.2
33.6
29.7
5.2
4.1
5.1
HCHO
0.01
0.08
0.18
1.22
0.98
0.63
0.49
0.10
Parti culatec
<0.004
<0.004
<0.004
a
0.006
0.019
0.022
0.019
440 ug emitted at ignition.
10-minute warm-up outside chamber.
cParticulate values: Mass of particles from 0.005 to 0.4 urn diameter.
Source: Traynor et al. (1983)
found that interroom transport was less than 10 m /hr for the pollutant gases
o
with the door closed, 30 ± 10 m /hr with the door open 2.5 cm, and ranged from
190 to 3400 m3/hr with the door fully open (Traynor et al., 1984).
Traynor et al. (1986) have also reported finding that other compounds,
including aliphatic hydrocarbons, alcohols, ketones, phthalates, and alkyl
benzenes are emitted from kerosene heaters. In a chamber study of convective
and radiant kerosene heaters they concluded that the reactivity of semivolatile
organics (SVOCs) implies that reactivity rate for SVOCs is more important in
determining their indoor concentration than ventilation or air exchange rates.
This indicated that further studies are needed to quantify the indoor
reactivity processes for individual SVOCs in order to get accurate insight into
the exposure (Traynor et al., 1986).
2.2.2.2 Gas Appliances. It has been estimated that 45 percent of American
homes use natural gas, and that most do not vent combustion products outside
the house (U.S. Department of Energy, 1985). Therefore it is not surprising
that the house unvented gas combustion appliances can be a significant source
2-8
-------
of indoor pollutants. They have been reported to be a source of particulate
matter, NO , S09, HCHO, and CO (Gas Research Institute, 1985; Girman et al.,
X £.
1982). The factors that appear to affect the pollutant emission rates are
usage pattern, burner design and manufacture, age of burner, fuel consumption
rate, and combustion efficiency (U.S. Department of Energy, 1985).
2.2.2.2.1 Gas stoves. In homes where gas stoves are used, kitchen pollutant
concentrations have been shown to respond rapidly to stove use. For a given
house during a given season, there is a rough correlation between average NO,,
concentration and the average stove use. It has been reported that normal
stove operation frequently results in N09 concentrations in the kitchen that
3
are 100 ug/m (0.053 ppm) over a 2 week sampling period, with NO and N02 being
produced in roughly equal amounts (Wade et al., 1975).
Pollutant emissions from newer stoves have not consistently been reported
to be higher or lower than those from older stoves. The different designs of
the burners do not appear to have a consistent or reproducible affect upon
pollutant emissions. An evaluation of the number of burners in use and their
flame intensity has shown that emissions are similar for all conditions when
adjusted to equivalent heat output (GEOMET Technologies, Inc., 1976).
Oven and broiler emissions were reported to be somewhat less than those of
burners on the heat output basis. The total pollutant emissions (per unit of
time) from a gas stove are roughly proportional to the number and types of
burners and the period of use. Pilot lights appear to contribute quantities of
pollutants comparable to those generated during cooking activity over a typical
24-hour period (GEOMET Technologies, Inc., 1976).
Table 2-4 presents emissions rates for gas range and oven (U.S. Department
of Energy, 1985).
Two major field studies dealing with the indoor air impact of gas stoves
have focused on N0?. They are the Southern California Gas Corporation study
(Colome et al., 1982) of 500 hours (400 with gas stoves, 100 with electric
ranges) in the Los Angeles, CA, area and the Harvard-Gas Research Institute
study of 600 hours (450 gas, 150 electric) in the Boston, MA area (Soczek
et al., 1986).
Both studies included the main components of random selection of a cluster
of homes; two-week measurements using Palmes tubes; multiple indoor sampling
locations (i.e., kitchen, bedroom, and living room); air exchange measurement;
water vapor measurements; and outdoor air measurements.
2-9
-------
TABLE 2-4. EMISSION RATES FOR GAS RANGES AND OVENS (mg/hr)
Nonstanding pilot, fuel input rate per burner 8680 to 9060 Btu/hr
Rich gas (1022 Btu/SFC) Lean gas (983 Btu/SFC)
NO 165-176
N02 87-104
CO 195-1769
NO 163-182
N02 84-97
CO 257-1456
Nonstanding pilot, fuel input rate per burner 7950 to 8870 Btu/hr
Rich Gas (1022 Btu/SFC) Lean Gas (983 Btu/SFC)
NO 131-151
N02 75-105
CO 213-675
NO 135-164
N02 66-76
CO 213-675
Standing pilot, fuel input rate per burner 9130 to 9890 Btu/hr
Rich Gas (1022 Btu/SFC) Lean Gas(983 Btu/SFC)
NO 163-206
N02 105-117
CO 180-563
NO 155-177
N02 101-110
CO 191-670
Gas-fired oven operated at 180°C; Fuel input rate 7970 Btu/hr
Average
NO 56
N02 85
CO 1898
The studies compared the bedroom median N02 values (2-week averaging time)
in Los Angeles and in Boston. Homes with gas stoves showed approximately
double the N09 concentrations reported for those of homes with electric stoves
3 3 3 3
(40 ug/m compared to 20 pg/m in Boston; 80 pg/m compared to 40 (jg/m in Los
Angeles).
In Boston in winter, about 5 percent of the gas-stove homes had bedroom
NO/, averages that exceeded the annual average National Ambient Air Quality
3
Standard (NAAQS) of 100 ug/m . This value fell to less than one percent in
summer. In Los Angeles in winter more than 30 percent of homes with gas-stoves
exceeded this standard and more than 20 percent exceeded the standard in
summer. In both cities less than one percent of the electric-range homes
exceeded the standard.
2-10
-------
Both studies found that gas stoves, both the pilot light and cooking
operation, were a major influence on indoor air quality, adding about 30 to 40
pg/m of N0? on the average. In addition, wall- and floor-vented gas furnaces
in the Los Angeles area added roughly equivalent amounts of l^.
A study of 152 homes in the Netherlands (Remijn et al., 1985) showed bed-
o
room NOp levels ranging between 8 and 66 pg/m , with a geometric mean of 23
pg/m3 and a geometric standard deviation of 1.5. (This was reported to be
similar to the values observed in Los Angeles and in Boston.) Kitchen and
3
living room levels were higher, with geometric means of 79 and 39 pg/m ,
respectively, and with a higher geometric standard deviation of 2.2 and 2.0,
respectively.
Davidson reviewed the pollutant emission factors for gas stoves presented
in the literature. Using these published emission factors, he attempted to
determine the important parameters influencing emissions from gas stoves by
statistical analysis (Davidson, 1986). His results showed that roughly one-
half of the observed variance in the base 10 log of the emission factors for CO
could be explained by either poorly adjusted or well adjusted combustion. For
N0?, approximately one-third of the variance could be explained by poor/well
adjustments. For NO and NO the resulting variance depended on the data subset
/\
and ranged from 0.088 to 0.56.
Davidson concluded that burner position and method of sampling were rela-
tively unimportant in explaining variance for CO, NO, N02 and NOX. However,
the critical gaps that merit further study were the influence of stove design,
gas flow rate, and characteristics of the stove (Davidson, 1986).
Several investigators have indicated that kitchen gas ranges are a major
source of indoor air pollution. Such studies, however, address only normal
operation of gas ranges for cooking. Sterling and Kobayashi (1981) reports
that 50 to 55 percent of New York gas ranges are used for both cooking and
heating.
2.2.2.2.2 Gas-fired space heaters. Unvented gas-fired space heaters are
reported to contribute to indoor concentrations of NO, NOp, CO, C0£, respirable
suspended particles (RSP), and HCHO. The factors that influence the emissions
include usage pattern, brand of heater, burner design, size of heater (rated
input), and the tuning of the fuel to air mixture (U.S. Department of Energy,
1985).
2-11
-------
Zawacki et al. (1985) reported emission factors for NO,,, NO, CO, and
unburned hydrocarbons from gas-fired space heaters. The factors were experi-
mentally determined for ten different unvented gas space heaters by three
different measurement methods (probe, hood, and chamber). The heaters were of
various designs (blue-flame with and without radiating tiles, and infrared) and
were fueled by either natural gas or liquified propane gas (LPG) and fired over
a range of input rates (10,000 to 40,000 Btu/hr).
Zawacki and co-workers initially concluded that emissions were primarily
dependent on the heater design (blue-flame versus infrared) and the method of
measurement. The probe and hood methods gave almost identical results, which
were different from the chamber method, especially for NO and CO. The differ-
ence was apparently caused by the reduced oxygen content of the chamber which
3
was held at approximately 0.5 mg/m each (Zawacki et al., 1985). In the final
report of this work, Zawacki and co-workers concluded that natural gas-fired
infrared heaters averaged NOp emissions levels of one-fourth to one-half of
that of blue-flame heaters and that these levels appeared to depend on heater
design and method of measurement (Zawacki et al., 1986). For blue-flame
burners, the average N09 emissions levels were essentially the same, about 0.02
6
Ib/lOE Btu, irrespective of burner input rate and type, fuel type, method of
measurement, and existence or absence of suspended radiating tiles (Zawacki et
al., 1986).
The propane fired heaters exhibited the highest NO emission levels, which
was consistent with expectations that propane heaters developed the highest
adiabatic flame temperatures. Natural gas fueled blue-flame burners exhibited
about one-half equivalent NO emissions, while natural gas infrared burners pro-
duced virtually none. For blue-flame burners, the probe and hood methods of
measurement resulted in approximately identical NO emissions (0.1 Ib/lOE Btu)
from propane fuel and about one-half as much for natural gas-fired (Zawacki et
al., 1986).
Natural gas-fired infrared heaters exhibited the highest average CO emis-
sions levels, 0.03 to 0.17 Ib/lOE Btu, followed by natural gas-fired blue-
flame heaters, 0.02 to 0.4 Ib/lOE Btu, and propane-fired blue-flame heaters
(0.01 to 0.015 Ib/lOE Btu). The burner design and test method was demon-
strated to have a modest to large effect within a category of heaters (Zawacki
et al., 1986).
Blue-flame propane heaters exhibited virtually no unburned hydrocarbon
(UBHC) emissions, while natural gas blue-flame heaters produced about 0.05 to
2-12
-------
0.2 Ib/lOE Btu, and the infrared tile heaters produced a wide range of UBHC
emissions from nearly zero up to 0.9 Ib/lOE Btu (Zawacki et al., 1986).
The comparison of 8 heaters using the chamber method at Lawrence Berkeley
Laboratory (LBL) and the Institute of Gas Technology (IGT) showed no substan-
tial difference in emissions of N02, NO and CO (Table 2-5). NO emission were
found to be dependent on the quantity of air used for combustion and emissions
factors decreased as the chamber humidity increased and as air infiltration
decreased. The decreasing air infiltration apparently reduced the chamber
oxygen content, inhibiting NO formation and stimulating increased production of
CO. The overall effect of water vapor, whether chemical or thermal, is
unknown. No effect of either humidity or air on N0« emission rates could be
detected (Gas Research Institute, 1985).
TABLE 2-5. OVERALL COMPARISON OF EMISSION FACTORS
FROM IGT AND LBL CHAMBER STUDIES
Contaminants
N02
NO
CO
N02
NO
CO
Number of
Heaters
2
2
2
5
5
5
Emissions Factors (lb/10E6 Btu)
IGT LBL
Blue-Flame
Heaters
0.0215 ± 0.0025 0.0260 ± 0.0028
0.0699 ± 0.0042 0.0663 ± 0.0008
0.0260 ± 0.0132 0.0447 ± 0.0194
Infrared
Heaters
0.0106 ± 0.0028 0.0119 ± 0.0025
0.0005 ± 0.0005 0.0007 ± 0.0004
0.1000 ± 0.070 0.0007 ± 0.0004
2.2.2.2.3 Gas-fired water heaters and dryers. Gas water heaters and clothes
dryers have not received much attention to date. Although both are designed to
be vented, there is the possibility that faulty systems can produce an indoor
air impact. The U.S. DOE reports a personal communication with Lawrence
Berkeley Laboratory in which it is reported that homes with vented gas water
heaters were measured to have indoor concentrations of N0? greater than outdoor
concentrations. The source in each was traced to flue collars atop the gas
water heater (U.S. Department of Energy, 1985).
2-13
-------
2.2.2.3 Wood-burning Stoves and Fireplaces. Laboratory and field measurements
show that pollutant emission rates from wood combustion in general can vary
widely: CO, 4 to 400 g/kg of wood; particles, 0.5 to 63.5 g/kg; total HC, 0.2
to 48.5 g/kg; NO , 0.2 to 7.3 g/kg, and particulate organic matter (POM), 0.004
/\
to 0.37 g/kg (Fisk et al. , 1985). Consequently, wood-burning stoves, wood-
burning furnaces, and fireplaces, even though vented to the outside, can
represent a significant source of indoor CO, NO , HC, HCHO, and RSP, including
X
carcinogenic POM.
Factors that can affect the indoor emission rates are improper installa-
tion, cracks or leaks in stove pipes, negative air pressure within the dwell-
ing, downdrafts, refueling, and accidents. Emissions are also dependent upon
the use pattern, manufacturer and type of appliance, and type of wood burned.
There are little data available on the emission rates indoors from such
devices. However, they probably emit relatively small quantities of pollutants
over long periods of time as a consequence of leaks, and relatively high
emissions over short periods such as during reloading. Consequently, the
emissions can be expected to vary considerably in terms of mix and frequency.
The Tennesse Valley Authority (TVA) and the Bonneville Power Authority
(BPA) undertook to develop data on Indoor Air Quality (IAQ) in relationship to
the use of conventional and new technology wood heaters in airtight homes.
Four heaters were used in the study: two conventional non-airtight, and two
high-technology airtight. Sequential indoor and outdoor measurements of CO,
C09, NO, and NO were taken in a modular test home, evaluated and compared to
L. X
the National Ambient Air Quality Standards for the same pollutants (Tennessee
Valley Authority, 1985).
The non-airtight (NAT) heaters were reported in a TVA/BPA study to repre-
sent a significantly larger source of indoor C0? than the airtight heaters
(AT). The highest hourly indoor CO^ concentration reported during the test of
a NAT wood heater was 1240 ppm, and the highest instantaneous indoor level was
1358 ppm (Tennessee Valley Authority, 1985).
The highest 12-hour indoor S02 concentration was 42.4 ppb, which was 30
percent of the 24-hour NAAQS and 8 percent of the 3-hour NAAQS. A statistical
analysis of the data showed that the NAT operated with significantly higher
indoor SOp source strength than the AT unit (Tennessee Valley Authority, 1985).
In a final report, TVA/BPA reported that with respect to NO, the NAT wood
heater operated as a much higher indoor NO source strength during closed
2-14
-------
stack-damper operations when the burn rate was being reduced. Indoor N0?
levels and source strength were not found to be significantly related to AT or
NAT designs. The indoor NCL levels were essentially driven by outdoor levels
and were generally less than 60 percent of those outdoors (Tennessee Valley
Authority, 1986).
Both AT and NAT wood heaters were found to present a statistically signif-
icant (at the 0.05 level) source of CO. The source strengths and Indoor/
Outdoor Ratio (I/O) for CO were greater for NAT than for AT heaters. The NAT
generated much higher indoor CO during closed stack-damper operations. In the
final report for this work, TVA/BPA concluded that the maximum 1-hour concen-
tration of CO was 9.1 ppm, and the maximum 8-hour CO concentration was 5.7 ppm
(Tennessee Valley Authority, 1986).
In general, the indoor methane hydrocarbons were slightly higher than out-
door levels for both non-burning and active testing, consequently, neither type
of wood heater was determined to be a significant source for non-methane hydro-
carbons (NMHC) (Tennessee Valley Authority, 1985). However, both were deter-
mined to be a source of indoor PAH and B(a)P.
2.2.2.4 Attached Garages. Automobiles or trucks operating in confined small
spaces, such as parking garages, can cause extremely high CO concentrations.
Wallace reported elevated CO levels in the air and lungs of workers in an
underground office that was connected to a parking garage (Wallace, 1983).
Even so, there is little in the literature on the affect of automotive exhaust
on indoor air quality via an attached garage.
Flachsbart and co-workers employed miniature personal exposure monitors to
measure CO in 588 different commercial settings, for example, retail stores,
office buildings, hotels, and restaurants in five California cities. They
determined that the CO levels indoors were similar to those measured outdoors,
usually greater than 0 but less than the NAAQS for 8-hour exposure, unless an
indoor source was present. Office buildings with indoor garages were measured
to be greater than the NAAQS of 9 ppm (Flachsbart et al., 1984).
Attached garages also present a source of pollutants in addition to the
conventional combustion gases. Gammage et al. measured volatile organics
inside 40 homes; volatiles being defined as those organics with boiling points
of less than 110°C. He determined that the concentration of volatiles was
usually ten times higher inside than outside homes. The highly volatile
organics were usually dominated by gasoline fumes. Attached garages containing
2-15
-------
automobiles, stored gasoline, and motor oil were considered to be responsible
for this ubiquity. Gammage and co-workers concluded from this work that air
drawn through air conditioning ducts located in the attached garages provided
an effective means of funnel ing gasoline and engine exhaust into living spaces
(Gammage et al., 1984).
2.2.2.5 Conclusions. A summary table of emission rates for combustion related
gases from indoor sources has been published by DOE and is presented in Table
2-6.
TABLE 2-6. POLLUTANT EMISSION RATES (MG/HR)
Source
Kerosene
Space
Heaters
Gas Space
Heaters
Wood Heaters
Gas Appliances
Appliance
Type
Radiant
Convective
Range
(1 burner)
Oven
NO
0.54-11
2-195
80-4578
1.2-3.9
9.5-455
30-581
N02
16-38
33-530
3-1225
1.3-7.0
18-430
67-270
CO
281-542
35-635
12-5004
70-375
191-2700
195-3564
S02
31-109
37-94
--
—
1.29-1.66
0.67-1.09
Source: U.S. Department of Energy (1985).
The absence of reliable emission factors from the indoor sources makes it
impossible to establish a relative ranking between reported sources and their
contribution to indoor air pollution at this time. In general, the absence of
standardized test procedures, including sampling and analysis methods, and the
lack of uniformity in test conditions precludes the use of much of the existing
data for the predictive modeling which facilitates risk assessment calculations.
Results from the testing of gas-fired appliances are a good example of the
difficulty in making such comparisons. The comparison of various studies of
NO, N0~, NO , and CO emissions and concentrations from such sources must be
L- f\
made with great care, since each experiment uses a different sampling point
location, appliance, and air-fuel ratio (GEOMET Technologies, Inc., 1976).
2-16
-------
Major sources of uncertainty related to indoor air combustion products are
the source, identity, and emission rates of unburned benzene hydrocarbons
(UBHC); the reactivity of the various organic and inorganic gases; and the
applicability of a given control option to a specific pollutant source.
2.2.2.6 Major Knowns and Unknowns. The most important research need is accu-
rate source characterization, including measurements of emissions from combus-
tion sources, pollutant removal by indoor "sinks," and the factors that affect
their emission or removal rates. The ultimate objective of source character-
ization is to determine the most important sources of indoor air pollutants,
and to characterize them in such a way that the most cost-effective methods of
controlling them can be determined.
Indoor air source characterization research needs related to combustion
sources include the following:
Expand the characterization of organics from combustion devices,
Survey indoor combustion sources to gather statistics on age,
condition, and operational parameters,
Investigate the factors affecting leakage rates of pollutants
from vented combustion sources,
Study sink rates of pollutants, especially N02 and S02, on
indoor materials, and
Measure source strengths of automobile exhausts in attached
garages.
2.2.3 Carbon Monoxide
Carbon monoxide is a colorless, odorless, tasteless gas that is slightly
soluble in water. It is slightly lighter than air (specific gravity = 0.967).
It is an asphyxiant for which the mechanism of action is an avid binding to
hemoglobin. Its affinity for hemoglobin is about 210 to 250 times that of
oxygen, thereby interfering with oxygen transport to tissues and resulting in
tissue hypoxia.
2.2.3.1 Exposure. Human exposures to CO in the residential indoor microenvi-
ronment have been investigated in studies in which subjects have carried per-
sonal monitors that recorded an average CO exposure during intervals of up to
one hour while a subject is indoors at home, or at work in a public building
2-17
-------
(Akland et al., 1985). The results are reported in several publications and
EPA reports (Holland, 1983; Akland et al., 1986). Table 2-7 shows the mean
and standard deviations of CO in the several indoor microenvironments defined
in the study. These highly skewed results indicate that people may be exposed
to CO values that approach the NAAQS for both the 8-hour and 1-hour averaging
times. However, these data were collected during the winter and do not repre-
sent an annual average for these cities. Table 2-8 gives the CO distribution
found in the indoor residential microenvironment. The 8-hour NAAQS concentra-
tion for CO of 9 ppm was exceeded in 4.3 percent of the homes.
TABLE 2-7. INDOOR MICROENVIRONMENTS LISTED IN DESCENDING ORDER
OF WEIGHTED MEAN CO CONCENTRATION
Indoor Microenvironment
Category Measured
Public garage
Service station or motor vehicle
repair facility
Other location
Other repair shop
Shopping mall
Residential garage
Restaurant
Office
Auditorium, sports arena,
concert hall , etc.
Store
Health care facility
Other public buildings
Manufacturing facility
Residence
School
Church
Number
of subjects
116
125
427
55
58
66
524
2287
100
734
351
115
42
21543
426
179
CO Concentration (ppm)
Mean Std. Dev.
13.46
9.17
7.40
5.64
4.90
4.35
3.71
3.59
3.37
3.23
2.22
2.15
2.04
2.04
1.64
1.56
18.14
9.33
17.97
7.67
6.50
7.06
4.35
4.18
4.76
5.56
4.25
3.26
2.55
4.06
2.76
3.35
2.2.3.2 Monitoring of CO. From a health effects assessment perspective, there
is a need to measure CO levels in the air which people breathe on a continuous,
or real-time, basis. The most common method for continuous CO monitoring in
ambient air is based on nondispersive infrared (NDIR) spectroscopic detection.
While sensitive enough for the purposes, NDIR instruments are much too bulky
and complicated for personal exposure monitoring. To address this need,
therefore, a variety of small, portable personal monitors were developed during
2-18
-------
TABLE 2-8. INDOOR RESIDENTIAL CO DATA FOR
DENVER, COLORADO - WINTER 1982-83
21,229 OBSERVATIONS UNWEIGHTED
Percent!le Concentration (ppm)
10% 0.5
20 0.8
30 1.0
40 1.3
50 1.8
60 2.3
70 3.0
80 4.2
90 6.2
95 8.5
99 14.5
Maximum 26.7
Mean 2.76
Standard Deviation 2.92
the past decade (Ott et al., 1986b). Most of these CO personal exposure moni-
tors (PEMs) were developed for occupational safety and health uses, especially
by the mining industry. Most of the devices are based on electrochemical
detection; that is, they employ a liquid or solid electrolyte in which CO is
converted to COp, thereby generating an electrical signal. The signal is
proportional to the quantity of CO present in the gas stream, and the con-
tinuous electrical signal is either recorded internally or displayed on a
digital readout system. A small pump operates continuously to send air into
the sensing cell, and chemical filters in the intake stream remove interference.
The most promising of the electrochemical CO PEMs was one developed by
General Electric Company and called "COED" (Ott et al., 1986a). In the winter
of 1982-83, the EPA successfully demonstrated the utility of two versions of
this new monitor in large-scale pilot studies of statistically representative
samples of the populations in two cities, Denver, CO, and Washington, DC
(Akland et al., 1985). The CO PEM used in Denver contained a sensing cell with
a solid polymer electrolyte, and an internal microprocessor data logger which
recorded and stored up to 100 readings from the PEM (Ott et al., 1986b). This
version of the CO PEM, designated the COED-I, weighed less than 4.4 Ig (2 Ib.)
and included a digital readout display. The second version of the CO PEM
utilized in the Washington study (Mack et al., 1987) incorporated the same CO
2-19
-------
detection system, but used an external data logging system built around a user-
programmable pocket calculator. This prototype system, designated the COED-II,
was much more flexible and stored more than 400 data entries. The COED-II was
capable of operating continuously for 40 hours and provided accurate measure-
ments at 2 ppm sensitivity. During the Washington study, more than 1200
exposure profiles were constructed using the COED-II, and, considering the
newness and complexity of the technique, the problems encountered were rela-
tively minor.
While the COED-II monitor functioned well, it required daily servicing. A
self-controlled data-logging PEM that can be operated unattended for two weeks
or more is needed for further Total Human Exposure (THE) and IAQ surveys. For
large surveys, use of the COED-II or a similar active (pumped) PEM would be
cost prohibitive. Therefore, a sensitive and reliable passive device needs to
be developed.
2.2.3.3 Health Effects. The effects of CO on oxygen transport are well estab-
lished (National Research Council, 1977; U.S. Environmental Protection Agency,
1979, 1984a). CO interferes with oxygen transport by avidly binding to hemo-
globin to form carboxyhemoglobin (COHb) and by shifting the oxyhemoglobin dis-
sociation curve to the left. As a result of this shift, oxygen that is
attached to hemoglobin is released to the tissues less readily, and additional
tissue hypoxia results. The health effects of low levels of CO exposure (COHb
<3 percent) are controversial, but the problem of CO poisoning by indoor
combustion sources has been well described. The clinical manifestations of
poisoning by CO primarily represent the effects of reduced oxygen transport to
tissue and organs with high oxygen demand, such as the heart and the brain.
The neurological manifestations range from impaired mentation and behavioral
alterations to coma (Dolan, 1985; Ginsberg, 1985). Cardiac effects include
arrhythmias and myocardial infarction (Dolan, 1985). Even low levels of carbon
monoxide exposure affect the heart and brain.
Research on the health effects of lower levels of carbon monoxide has
primarily been conducted with experimental exposures of subjects, rather than
with epidemiological methods. Studies of carbon monoxide exposures in indoor
environments have been reported (Wallace, 1983; Cox et al., 1985), and health
effects of indoor exposure to carbon monoxide at levels lower than those
causing poisoning have occasionally been reported.
Tissue hypoxia resulting from the reduced oxygen carrying capacity of the
blood is generally thought to be the main mechanism of action underlying the
2-20
-------
induction of toxic effects of low-level CO exposures (5 to 20 percent COHb).
However, the precise mechanisms by which toxic effects are induced via COHb
formation are not yet fully understood. Alternative mechanisms of CO-induced
toxicity (besides COHb) have been hypothesized (Coburn, 1979; Agostoni et al.,
1980), but none has yet been demonstrated to operate at relatively low CO expo-
sure levels. Blood COHb levels, then, are currently accepted as representing
a useful physiological marker by which to estimate internal CO burdens due to
the combined contribution of (1) endogenously derived CO and (2) exogenously
derived CO resulting from exposure to external sources of CO. COHb levels
likely to result from particular patterns (concentrations, durations, etc.) of
external CO exposure can be reasonably well estimated from equations developed
by Coburn et al. (1965).
Evaluation of human CO exposure situations indicates that occupational
exposures in some workplace situations can regularly exceed 100 ppm CO, often
leading to COHb levels of 10 percent or more. In contrast, such high exposure
levels are much less commonly encountered by the nonoccupationally exposed
general public. More frequently, exposures to less than 25 ppm CO for extended
periods of time occur among the general population and, at the low exercise
levels usually engaged in under such circumstances, resulting COHb levels most
typically remain 2 to 3 percent among nonsmokers. Those levels can be compared
to the physiologic norm for nonsmokers, which is estimated to be in the range
of 0.3 to 0.7 percent COHb. Baseline COHb concentrations in smokers, however,
often greatly exceed 3 percent, reflecting absorption of CO from inhaled smoke.
Some health effects associated with CO exposure occur at rather low COHb
levels (See Table 2-9). Four types of health effects reported or hypothesized
to be associated with CO exposures (especially those producing COHb levels
below 10 percent) were evaluated in the addendum to the 1979 Air Quality
Criteria Document for Carbon Monoxide (U.S. Environmental Protection Agency,
1984a): (1) cardiovascular effects; (2) neurobehavioral effects; (3) fibrino-
lysis effects; and (4) perinatal effects. Available data demonstrate an
association between cardiovascular and neurobehavioral effects at relatively
low-level CO exposures. Evidence that is less clear suggests that other types
of health effects are associated with low-level CO exposures. In regard to
cardiovascular effects, decreased oxygen uptake capacity and resultant decreased
work capacity under maximal exercise conditions have been clearly shown to
occur in healthy young adults starting at 5.0 percent COHb; and several studies
2-21
-------
TABLE 2-9. LOWEST OBSERVED EFFECT LEVELS FOR HUMAN HEALTH EFFECTS
ASSOCIATED WITH LOW LEVEL CARBON MONOXIDE EXPOSURE
Effects
COHb
concentration
(Percent)3
Reference
Statistically significant
decreased (~3 to 7 percent)
work time to exhaustion in
exercising young healthy men
Statistically significant
decreased exercise capacity
(i.e., shortened duration
of exercise before onset of
pain) in patients with angina
pectoris and increased
duration of angina attacks
Statistically significant
decreased maximal oxygen
consumption and exercise
time during strenuous
exercise in young healthy men
No statistically significant
vigilance decrements after
exposure to CO
Statistically significant
diminution of visual
perception, manual dexterity
ability to learn, or
performance in complex
sensorimotor tasks (such
as driving)
Statistically significant
decreased maximal oxygen
consumption during strenuous
exercise in young healthy men
2.3 to 4.3 Horvath et al. (1975)
Drinkwater et al. (1974)
2.9 to 4.5 Anderson et al. (1973)
5 to 5.5
Less than
5
5 to 17
7 to 20
Klein et al. (1980)
Stewart et al. (1978)
Weiser et al. (1978)
Haider et al. (1976)
Winneke (1973)
Christensen et al. (1977)
Benignus et al. (1977)
Putz et al. (1976)
Bender et al. (1971)
Schulte (1973)
O'Donnell et al. (1971)
McFarland et al. (1944)
McFarland (1973)
Putz et al. (1976)
Salvatore (1974)
Wright et al. (1973)
Rockwell and Weir (1975)
Rummo and Sarlanis (1974)
Putz et al. (1979)
Putz (1979)
Ekblom and Huot (1972)
Pi may et al. (1971)
The physiologic norm (i.e., COHb levels resulting from the normal catabolism
of hemoglobin and other heme-containing materials) has been estimated to be
in the range of 0.3 to 0.7 percent (Coburn et al., 1963).
Source: U.S. Environmental Protection Agency (1984a).
2-22
-------
revealed small decreases in work capacity at COHb levels as low as 2.3 to 4.3
percent. These cardiovascular effects may have health implications for the
general population in terms of potential curtailment of certain physically
demanding occupational or recreational activities under circumstances of
sufficiently high CO exposure. However, of greater concern at more typical
ambient CO exposure levels are certain cardiovascular effects (i.e., aggrava-
tion of angina symptoms during exercise) likely to occur in a smaller, but
sizeable, segment of the general population. This group, chronic angina
patients, is presently viewed as the most sensitive risk group for CO exposure
effects, based on evidence for aggravation of angina occurring in patients at
COHb levels of 2.9 to 4.5 percent COHb. Such aggravation of angina is thought
to represent an adverse health effect for several reasons articulated in the
1980 proposal preamble (Federal Register, 1980); the Clean Air Scientific
Advisory Committee (CASAC) concurred with EPA's judgment on this matter.
Dose-response relationships for cardiovascular effects in coronary artery
disease patients remain to be more conclusively defined, and the possibility
cannot be ruled out at this time that such effects may occur at levels less
than 2.9 percent COHb (as hinted at by the results of the now questioned Aronow
studies). New studies are, therefore, currently in progress to investigate the
effects of CO on aggravation of angina at levels in the range of 2 to 4 percent
COHb.
No reliable evidence demonstrating decrements in neurobehavioral function
in healthy young adults has been reported at COHb levels less than 5 percent.
Much of the research at 5 percent COHb did not show any effect even when beha-
viors similar to those affected in other studies were involved. However, if
any CO effects on neurobehavioral functions in fact occur less than 5 percent
COHb, then none of the significant-effects studies would have found such decre-
ments, because none of them used COHb levels less than 5 percent. Other
workers who failed to find CO decrements at 5 percent or higher COHb levels may
have employed tests not sufficiently sensitive to reliably detect small effects
of CO. From the empirical evidence, then, it can be said that the COHb levels
in the 5 percent range and greater produce decrements in neurobehavioral
function. However, it cannot be said confidently that COHb levels less than
than 5 percent would be without effect. One important point made in the 1979
document should be reiterated here. Only young, healthy adults have been
studied using demonstrably sensitive tests and COHb levels at 5 percent or
2-23
-------
greater. The question of groups at special risk for central nervous system
(CNS) effects, therefore, has not been explored. Of special note are those
individuals who are taking drugs which have primary or secondary depressant
effects which would be expected to exacerbate CO-related neurobehavioral
decrements. Other groups at possibly increased risk for CO-induced neuro-
behavioral effects are the elderly and ill, but these groups also have not been
evaluated for such risk.
Only relatively weak evidence points toward possible CO effects on fibri-
nolytic activity and, then, generally only at rather high CO exposure levels.
Similarly, whereas certain data also suggest that perinatal effects (e.g.,
reduced birth weight, slowed postnatal development, sudden infant death
syndrome) are associated with CO exposure, only insufficient evidence exists by
which to either confirm such association qualitatively or to establish any
pertinent exposure-effect relationships.
Based on currently available scientific evidence, onset of adverse health
effects occur, even in healthy individuals, at COHb levels in the range of 5 to
20 percent and greater. There are, however, uncertainties regarding the extent
to which adverse health effects are occurring in a large number of sensitive
individuals with COHb levels in the range of 3 to 5 percent. These uncertain-
ties were a major reason for EPA's decision not to change the existing 8-hour
primary standard level of 9 ppm (Federal Register, 1985). Additional studies
are now under way to identify the relationship between COHb and aggravation of
preexisting cardiovascular disease. The new information obtained from these
studies will allow a better determination of the level of COHb necessary to
cause adverse effects in the sensitive population and ultimately set an appro-
priate level for the CO NAAQS.
Angina patients or others with obstructed coronary arteries, but not yet
manifesting overt symptomatology of coronary artery disease, appear to be best
established as a sensitive group within the general population that is at
increased risk for experiencing health effects ('i.e., exacerbation of cardio-
vascular symptoms) of concern at ambient or near-ambient CO exposure levels.
Several other probable risk groups were identified, including: (1) fetuses
and young infants; (2) the elderly (especially those with compromised cardio-
pulmonary functions); (3) younger individuals with severe cardiac or acutely
severe respiratory diseases; (4) individuals with preexisting diseases such as
chronic bronchitis, emphysema, or congestive heart failure; (5) individuals
2-24
-------
with hematological diseases (e.g., anemia) that affect oxygen-carrying capacity
or transport in the blood; (6) individuals with genetically unusual forms of
hemoglobin associated with reduced oxygen-carrying capacity; and (7) individ-
uals using medicinal or illicit drugs having CMS depressant properties.
However, there is currently little empirical evidence available by which to
specify health effects associated with ambient or near-ambient CO exposures.
Nor does unambiguous evidence exist which clearly establishes that healthy
nonsensitive individuals. Those in the above probable risk categories are
affected at lower CO exposure levels under high altitude conditions than CO
exposure concentrations effective at lower altitudes.
2.2.3.4 Conclusions.
Major Knowns
1. CO causes cardiovascular effects. Chronic angina patients are
the most sensitive risk group currently known. Effects have
been observed at 2.9 to 4.5 percent COHb.
2. Higher levels of CO resulting in >5 percent COHb can cause
decrements in neurobehavioral functioning.
3. Several other types of health effects are observed after CO
exposure, but they are not as well defined as the cardiovascular
effects.
Major Unknowns
1. Patients with ischemic heart disease (Coronary Artery Disease -
CAD) who are exposed to low levels of CO have been reported to
experience persistent angina (chest pain) at COHb levels of
approximately 2 percent. Studies need to be completed to veri-
fy this report and to determine if all CAD patients or sensitive
segments of this population are at increased risk to CO.
2. While the incidence of angina pectoris has been identified as a
potential major health risk in patients with ischemic heart
disease, little is known about the reproducibility or recurrence
of this phenomena. Such information is vital to a proper
interpretation of the health data on CO effects.
3. A second group that may be at increased risk to low levels of
CO exposure and which needs to be studied are those with conges-
tive heart failure.
4. Patients with ischemic heart disease need to be studied to
determine whether they will develop cardiac arrhythmias fol-
lowing exposure to CO.
2-25
-------
2.2.4 Nitrogen Dioxide
Nitrogen dioxide, which is a red to brown gas with a pungent, acrid odor,
is produced during combustion at high temperatures from the combination of
nitrogen and oxygen from air. It is an oxidizing agent that is highly irritat-
ing to mucous membranes, and causes a wide variety of health effects.
2.2.4.1 Exposure. Human exposures to N02 in indoor microenvironments have
been obtained primarily from indoor concentration studies and PEM studies in
which the exposure is integrated over both indoor and outdoor microenviron-
ments. Thus it is difficult to construct a data set of N0? observations alone
obtained while the subject is in the target microenvironment which fits the
criterion of exposure to a single pure compound. All studies show that indoor
NC^ levels are lower than the immediate outdoor values when there are no indoor
sources (e.g., all-electric homes) due to the reactivity of NO,, with indoor
surface. In the presence of an indoor source such as a cigarette or a gas
stove in use, the NO^ levels are invariably higher than the outdoor NO,,.
However, for the cases of intermittent usage of the source, the all-electric
values approach the gas home values when the combustion sources are used
infrequently. Figure 2-1 presents mean indoor microenvironment N02 values from
a published study that measured indoor and outdoor concentrations simultane-
ously (Spengler et al., 1979). Short-term average N09 concentrations can reach
3
~1000 ug/m in gas homes during cooking in the kitchen.
2.2.4.2 Monitoring. Two main passive devices are available for measuring
integrated exposure to NCk: the Palmes tube and the Yanagisawa badge. The
Palmes tube (Palmes et al., 1976) is an acrylic or metal tube, normally about
1 cm in diameter and 7 cm long, containing three stainless steel grids coated
with triethanolamine (TEA) in a cap at the top of the tube. The bottom of the
tube is open to sample the air, allowing NOp to diffuse upward until it reacts
with the TEA, forming a stable complex for subsequent analysis by spectropho-
tometry. Although the sample has a sensitivity of several hundred ppb-hours,
with a lower limit of detection of 300-ppb-hr, requiring a minimum 2-day
collection time at normal indoor environmental levels, most studies use 1- to
2-week collection times. The Palmes tube has been used to study the relation-
ship between respiratory illness in children and the use of gas stoves for
cooking (Goldstein et al. , 1979).
2-26
-------
INDOOR NO2 LOWER, M9/m3 INDOOR NO2 HIGHER, M9/m3
-10 -50 5 10
ALL
HOMES |_
NO KITCHEN,
VENTING [
KITCHEN VENTING I
INSIDE I
KITCHEN VENTING]
OUTSIDE I
T
T
D
ELECTRIC
GAS
Figure 2-1. Mean indoor/outdoor difference in nitrogen dioxide concentrations
from cooking fuel and kitchen ventilation, average across all indoor/outdoor
sites (May 1977-April 1978).
Source: Spengler et al. (1979).
In an attempt to lower the minimum detection limit of the Palmes tube
sufficiently to achieve 8- to 24-hr utility, an ion chromatographic (1C)
analytical finish was explored. The application of 1C, which is inherently
more sensitive than UV spectrophotometry, combined with a specially-designed
concentrator column, improved the overall sensitivity by an order of magnitude
to 30 ppbv-hr + 20 percent (Miller, 1987, in press). While this analytical
improvement would theoretically permit one day or shorter exposure times, the
concentration step caused chloride ion interference with resolution of the
nitrite ion. It became necessary to build a special extraction apparatus to
extract the nitrite ion from the triethanolamine-coated screens in order to
prevent handling contamination. Finally, the acrylic tube of the Palmes N02
2-27
-------
device was found to act as a sink or a source of NO,, depending on how well it
was cleaned. Stainless-steel tubes were investigated to minimize contamination
and improve the minimum detection limit, but these proved too cumbersome to
recommend for routine personal monitoring.
A second method that appeared to have the sensitivity required was based
on a Thermosorb cartridge originally designed for nitrosamine collection and
analysis by GC-chemiluminescence analysis (Fine and Rounbehler, 1975). The
feasibility of converting the pumped Thermosorb cartridge to the passive mode
was investigated. However, results could not be replicated due to problems
with the chemistry of reaction of NOp with the morpholine-based sorbent.
A recent third approach modified the stainless-steel PSD developed for
volatile organic chemicals to make it responsive to NCy This was accomplished
simply by replacing the Tenax sorbent bed with TEA-coated glass fiber filters.
Ion chromatography (1C) can be employed for analysis.
o
The effective sampling rate of this PSD is 154 cm /min for NC^, making it
potentially 150 times more sensitive than the standard Palmes tube. These
devices have been evaluated in an N09 exposure chamber at concentrations
3 3
ranging from 20 mg/m to 460 mg/m (10 to 250 ppbv) and compared with a
continuously-reading Bendix NO chemiluminescent monitor. For 24-hr exposures,
/\
the correlation coefficient was 0.9955 over the range studied. Nitric oxide
o
(91 mg/m )caused no deleterious effect on the efficiency of the PSD at 57
percent RH. The response was also shown to be linear over exposure times from
2 to 30 hours at constant air concentrations (109 mg/m or 58 ppbv). The
minimum detection limit for the PSD was shown to be 30 ppb-hr when 1C analysis
was performed without the aid of a concentrator column. The PSD is a very
small, dual-faced cylinder, measuring 3.8 cm (1.5 in) o.d. x 1.2 cm (7/16 in)
and weighing only 36 g (1.3 oz). The device is undergoing further testing to
determine potential interferences from other chemical species. Parallel
comparisons with the Luminox LMA-3 monitor also need to be conducted.
The Yanagisawa badge (Nitta and Maeda, 1982) uses an absorbent sheet of
cellulose fiber coated with TEA. A five-layer mat of hydrophobic fiber reduces
the effect of face velocity on the diffusion characteristics of the badge.
The sensitivity of the badge is 66 ppb-hours, about ten times more sensitive
than that of the Palmes tube. Thus, sampling periods as short as eight hours
at environmental levels are possible. The badge has been used in a study of
the effect of unvented heaters on indoor NOp levels (Yanagisawa et al., 1981).
2-28
-------
As in the case of CO, there is a considerable need for monitoring systems
capable of determining continuous exposures to NQ^ in air. Systems designed
for outdoor ambient air monitoring are too large and complex to be used for
THE or IAQ monitoring purposes. While continuous monitoring is required to
determine brief exposures to N02 encountered during cooking with gas appli-
ances, more sensitive methods to measure low level, time-weighted average
concentrations are also needed for complete assessment of potential health
effects.
Over the past three years, considerable progress has been made in the
development of portable real-time N0? monitors based on the chemiluminescent
(light-producing) reaction between NOp and luminol (Wendel et al., 1983). Air
entering the system is pulled through a unit and across the face of a filter
wetted with a solution containing luminol. The chemiluminescence is detected
on a photodiode, which produces a voltage signal proportional to the NOp
concentration. Recent research and development sponsored by EPA has led to the
commercialization of' a small, light-weight monitor known as the Luminox LMA-3
(Schiff et al., 1986). The monitor was tested by EPA and found to have a
significant temperature dependence. Subsequent redesign has been successful in
moderating this dependence and existing units are currently being retrofitted
at the manufacturer's expense. Ultimate acceptance of the unit as a useful
means for directly monitoring indoor N0~ is expected if remaining questions
concerning the characteristics of the luminol reactant solution can be treated.
The exact composition of the solution determines the response characteristics.
During FY 87 the solution will be standardized to comply with the sensitivity
and linearity requirements of indoor air monitoring.
The Luminox LMA-3 monitor is 38 x 20 x 22 cm (15 x 8 x 8.5 in.) in size
and weighs 7 kg (15.5 Ib), which makes it easily transportable for an adult.
However, it needs to be operated on 115 VAC power, as the internal batteries
will only permit three hours of operation. It appears unlikely that the
luminol-based monitoring technology will permit significant improvements in
portability or battery operation. Consequently, further attempts at miniaturi-
zation of the Luminox monitor have been abandoned in favor of pursuing electro-
chemical sensor development.
A prototype electrochemical sensor system was recently developed by
Transducer Research, Inc. (TRI). With EPA support, TRI has significantly
enhanced sensitivity and demonstrated a total signal drift of less than 25 ppbv
2-29
-------
over 24 hours for the device. Interference from sulfur compounds remains a
problem and will be addressed as part of the development cycle. The FY 87
objectives include fabrication of units suitably packaged for testing and
evaluation as PEMs.
Improved PSDs for NOp have been developed in a parallel effort with con-
tinuous NOp monitor development. PSDs present no portability problems and, if
sufficiently sensitive, could serve as complementary tools for overall assess-
ment of NCL exposures.
Two other active N02 monitors are currently being tested for indoor
sampling application. The first is a small detector made by InterScan which
uses an electrochemical cell and is designed for use in occupational settings.
It has been modified by Harvard University and is being tested as a PEM to mea-
sure concentrations found in ambient settings. A miniature data system is
also under test to permit assessment of the exposure distributions. The
Ecolizer, made by Ecological Sciences, Inc., is currently being tested at
Columbia University to study gas stove emission exposures. It is small enough
to be worn on the person.
2.2.4.2.1 Absorption of NO^. Several studies have shown that various
materials including fabrics, paints, metals, and tatami mats absorb N0?. The
extent of absorption by a typical home can be estimated by comparing outdoor
N02 values with those in homes with no sources. Future studies needed in this
area include personal monitoring studies to answer the difficult question of
the influence of indoor concentrations on total exposure.
2.2.4.3 Health Effects of NOp. The health effects of N02 have been studied
extensively and reviewed by the U.S. Environmental Protection Agency (1982b)
and the World Health Organization (1987). Because of adverse health effects
associated with N0?, EPA has established a NAAQS of 0.05 ppm as an annual
average.
Knowledge of the health effects of N02 are derived from animal toxico-
logical, human clinical, and epidemiological studies. The animal and human
clinical data will be summarized briefly here. The discussion of the epidemio-
logy studies will be expanded because they are essentially studies of the
effects of the use of gas stoves and therefore the information is more broadly
useful to indoor air, since it is source-specific information.
Nitrogen dioxide, as described earlier in this chapter, is produced by
many combustion sources within homes as part of a very complex mixture from a
2-30
-------
given combustion source. However, with rare exception, controlled studies of
NCL (i.e., very controlled laboratory studies in which an animal or man is
exposed to known levels of N02 and effects are measured) are with N02 alone.
Thus, we have no significant knowledge of additivity, synergism, or antagonism
associated with NOp in mixtures with other combustion products, much less with
the full range of indoor air chemicals. The only exceptions with controlled
studies are with other common chemicals, namely 0^, NCL, and SCL (U.S. Environ-
mental Protection Agency, 1982b; Mustafa et al., 1984). The few studies of the
interaction of NCL and 0, indicate that NCL has either no influence, is addi-
tive, or is synergistic, depending upon the exposure regimen and end point
chosen. Human clinical studies of NCL in association with 03 and/or S02 show
no contribution of NCL to the effects, under the protocol used. Several
mixture studies with NCL were performed with animals in the 1960s and early
1970s, but the study designs do not permit an interpretation of the role of
N0?. The controlled interaction studies clearly indicate that when we have
knowledge of the health risks of NCL, this knowledge cannot be used to quanti-
tatively estimate the risks of NCL in mixture with other pollutants, including
indoor pollutants. It can only be roughly assumed that the effects would be a
combination of additivity and synergism. With this as background, a summary of
the effects of N02 in controlled human and animal studies follows.
Hundreds of animal toxicology studies have been performed with N02 (U.S.
Environmental Protection Agency, 1982b; World Health Organization, 1987).
Clearly, high levels of hKL (>2.5 ppm) cause increased susceptibility to
pulmonary bacterial infections and other changes in host defenses (acute
exposure), major structural alterations in the lung, including emphysemic-like
effects in several animal species (subchronic and chronic exposure), changes in
pulmonary function (subchronic exposure), and changes in lung biochemistry
(acute, subchronic, and chronic exposure). Lower levels of NOp (<1 ppm) also
cause decrements in lung host defenses (subchronic and chronic exposure), lung
structural changes that can be interpreted as possibly indicative of the
development of chronic lung disease (subchronic exposure), and changes in lung
metabolism (acute exposure). Very young animals, compared to young adults, do
not appear to be more susceptible to the subchronic effects of NO^. Extra-
pulmonary effects have also been observed at <1 ppm N02 (acute, subchronic, and
chronic exposure). Generally, the animal toxicology studies form the main
basis for the concern that chronic exposure of man to N02 could well result in
2-31
-------
chronic lung disease, likely to be irreversible, and that the concentration of
NOp is more of a determinant of health effects than is the length of exposure,
although length of exposure also plays a significant role.
Human clinical studies have demonstrated that acute (<2 hr) exposure of
resting, normal subjects only causes changes in pulmonary function at levels
rarely, if at all encountered indoors (>2 ppm) (U.S. Environmental Protection
Agency, 1982b). More recent studies are summarized in Table 2-10. A single
more recent study (Bylin et al., 1985) has reported pulmonary function changes
of normal subjects exercising in 0.2 ppm N0?, but this is in contrast to recent
reports by others (Linn et al., 1985). Current research has identified a sus-
ceptible subpopulation, that is, asthmatics (Table 2-10). Two institutions
have observed decrements in pulmonary function in asthmatics exercising in an
atmosphere of 0.3 ppm NOp (Horstman et al. 1987; Bauer et al., 1986); another
study did not find asthmatics to be more sensitive (Linn et al., 1985). These
and other studies also found that NO^-exposed asthmatics were also more sensi-
tive to bronchoconstrictor agents administered in the laboratory. The emerging
information currently indicates that asthmatics are a sensitive subpopulation,
but there is a variability in the sensitivity of asthmatics, with an unknown
causative factor for this variability.
As mentioned earlier, the literature on the epidemiology of N02 is based
on the study of people exposed to gas stoves that emit NO^. Gas stoves emit
other pollutants as well, but many experts interpret the effects to be due to
NO,,. For the purposes of assessing risks of exposures to indoor air, however,
the causative factor of effects in these studies is more directly and clearly
interpretable in terms of the source, independent of any single chemical
emitted from that source. It should be noted that of all the combustion
appliances, gas stoves are the only ones that have been well-studied.
In most of the epidemiological investigations, school-age children have
been the subjects, and symptom status and retrospective illness histories have
been obtained by parent-completed questionnaires. In some studies, a measure
of lung function has been obtained. In these investigations, the source of
NO,, exposure has primarily been emissions from gas-fueled cooking stoves.
Exposure has most often been categorized by simple questions concerning type of
cooking fuel; N02 levels have been directly measured in only a few of the
investigations. Stove type alone does not predict average exposures nor does
it provide any information concerning peak exposures during stove operation.
2-32
-------
TABLE 2-10. CONTROLLED HUMAN EXPOSURE STUDIES OF N02 EFFECTS
Pollutant/
Concentration
N02 0.1 ppm
Duration of
Exposure and
Activity
1 hr at rest
Number of
Subjects and Type
20 asthmatics
20 normals
Pulmonary Effects
SGaw decrease induced by
carbachol significantly
enhanced in normals
(p <0.005) and also in
asthmatics (p <0.05).
Symptoms References
None Ahmed et al. (1982)
CO
CO
N02 0.1 ppm
N02 0.1 ppm
N02 0.15 ppm
03 0.15 ppm
(N02 + 04) 0.15 ppm
N02 0.2 ppm
N02 4.0 ppm
1 hr at rest
1 hr at rest
2 hr intermittent
light exercise
2 hr intermittent
light exercise
75 min with two inter-
mittent exercises (Vf =
25 1/min and 50 1/mifi)
9 asthmatics
hypersensitive to
ragweed
15 normals
15 asthmatics
(atopies)
6 normals
31 asthmatics
23 asthmatics,
25 normals
N02 had no significant
effect on baseline SGaw,
No statistically significant
change in airway resistance
(SRaw) for either group.
Significant decrease (>5%)
in Gaw/Vtg with 03 for 5
of 6 subjects and all six
for combined 03-N02; very
small (<5%) decreases in
Gaw/Vtg. with N02 alone
in 3 of 6 subjects.
No effect on forced expira-
tory function or total
respiratory resistance
observed with N02 alone.
Statistically significant
small (X ~ 5.5%) exacerba-
tion by N02 of metacho line-
induced bronchoconstriction
in 17 of 21 subjects tested.
Small increases in SRaw
following exercise in
both clean air and N02.
No significant difference
between the two exposure
conditions.
None
None
Cough with 03 and 03 + N02
but not N02 alone.
Significantly fewer symp-
toms during N02 exposure
compared to air (p <0.05).
Ahmed et al. (1983)
Hazucha et al.
(1983)
Kagawa and Tsuru
(1979)
Kleinman et al.
(1983).
None
Linn et al. (1985)
(continued on following page)
-------
TABLE 2-10. (continued)
Pollutant/
Concentration
Duration of
Exposure and
Activity
Number of
Subjects and Type
Pulmonary Effects
Symptoms
References
N02 0.3 ppm
30 min with one
10 exercises (Vc
30 1/min) fc
15 asthmatics
N02 0.3 ppm
ro
i
GO
90 rain with three
intermittent exercises
(V£ = 45 1/min)
13 asthmatics
N02 0.12 ppm
N02 0.24 ppm
N02 0.5 ppm
20 min at rest
8 normals
8 asthmatics
Greater than 75% lung
deposition during both
rest and exercise. No
responses to at rest
exposures, decreases in
FEVt and PEFR at 60% TLC
were significantly greater
following exercise in N02
than that in clean air.
Small, but significant
increase in airway reac-
tivity (bronchoconstrictor
response) to cold air
provocation.
Decrease in FEV! following
initial 10 min exercise in
N02 was significantly longer
than that observed in clean
air. After the second and
third exercise, SRaw in-
and FEVt and FVL decreases
significantly greater in
N02 than in clear air.
In normals, increased SRaw
at 0.24 ppm and decreased
SRaw at 0.5 ppm; no effect
on TGV.
In asthmatics, no signif-
icant effect on SRaw; lower
TGV after 0.5 ppm.
In both groups, no signif-
icant change in respiratory
rate. Bronchial reactivity
to histamine studied after
0.5 ppm: increased reac-
tivity (SRaw) in asthmatics,
but not normals.
None reported
Bauer et al. (1984)
Slight cough and dry mouth/
throat after initial
exercise in N02
Rogers et al.
(in press)
Not studied
Bylin et al. (1985)
SRaw = Specific Airway Resistance
SGaw = Specific Airway Conductance
Source: Grant (1984).
-------
Consistent evidence of excess respiratory symptoms and illnesses in
children living in homes with gas stoves has not been found (Table 2-11).
Melia and co-workers (1977) published one of the first reports of adverse
health effects associated with exposure to gas stove emissions. Significant-
ly higher prevalences of bronchitis, day or night cough, and colds going to the
chest were found for English schoolchildren living in homes with gas as com-
pared with electric stoves. In follow-up assessments of these children, the
pattern of an adverse effect of gas stove exposure was less evident (Melia et
al., 1979). Other studies in England by this and other groups have not shown
consistent evidence of harmful effects of indoor NO^ exposure on symptoms and
illnesses in children (Florey et al., 1979; Melia et al., 1982a,b, 1983; Ogston
et al., 1985).
In a 1980 report from Harvard's Air Pollution Health Study in six U.S.
communities, the initial findings of the British investigators were corrobo-
rated (Speizer et al., 1980). In a cross-sectional study of 8,120 children
ages 6 to 10 years, report of a serious respiratory illness before age two
years was weakly but significantly (odds ratio = 1.23) associated with current
use of a gas stove. However, with expansion of the cohort to 10,106 children,
the odds ratio for respiratory illness before age two declined to 1.13 (p =
0.07) (Ware et al., 1984).
Other investigations, both cross-sectional and longitudinal, have also
examined the associations between gas stove exposure and respiratory illnesses
and symptoms in children. The findings do not consistently provide evidence
of effects of gas stove exposure (Keller, 1979a,b; Dodge, 1982; Ekwo et al.,
1983; Schenker et al., 1983).
The data concerned with lung function level in children are similarly
inconclusive. Of four investigations with large sample sizes (Ware et al.,
1984; Speizer et al., 1980; Hasselblad et al., 1981; Vedal et al., 1984) two
have demonstrated unequivocal and statistically significant effects (Speizer et
al., 1980; Hasselblad et al., 1981). The magnitude of effect was extremely
small, on average. While lung function has been considered in other studies,
the sample sizes were inadequate for detecting effects of the magnitude found
in the larger studies (Table 2-12).
Only a few investigations provide data on effects of indoor NO^ on adults.
Prospective studies of acute respiratory illness occurrence have not demon-
strated excesses in residents of homes with gas stoves (Keller et al., 1979a,b;
Love et al., 1982). Studies of lung function level and of chronic respiratory
2-35
-------
TABLE 2-11. EFFECTS OF GAS COOKING ON RESPIRATORY ILLNESSES AND
SYMPTOMS IN CHILDREN
Study Population
Outcome Measure
Results
BRITISH STUDIES:
5758 children, 6 to
11 yrs, England and
Scotland (Melia,
1977).
2408 children, 42%
of original 5758
in above study (Melia
et al., 1979).
4827 children, 5 to
11 yrs, England and
Scotland (Melia
et al., 1979).
808 children, 6 to 7
yrs, United Kingdom,
(Florey et al., 1979).
191 children,
5 to 6 yrs,
England (Melia
et al., 1982a,b).
390 infants, 0 to 1
yrs, England (Melia
et al., 1983).
1565 infants, 0 to 1
yrs, England
(Ogston et al., 1985).
Major respiratory
symptoms and diseases
individually, and as a
single composite vari-
able describing the
presence of any 1 of
6 symptoms or diseases.
Single composite
variable as described
above.
Single composite
variable as described
above.
Single composite
variable as described
above.
Single composite
variable as described
above.
Respiratory illnesses
and symptoms requiring
physician visits,
assessed prospectively.
Respiratory illnesses
and hospitalizations
assessed prospectively
to I yr.
Significant associa-
tions gas between cooking
and selected symptoms
and diseases, and of a
composite variable.
Relative risk for
composite variable
generally exceeded
1.0; risk varied
and decreased with
age.
Significant effect of
gas stoves on com-
posite variable in
urban areas only.
Borderline signifi-
cant association
between composite
variable and gas
stoves. Increased
prevalence as bedroom
N02 levels increased
in a sample with
measurements (n = 80).
No significant asso-
ciation between bedroom
N02 levels and
prevalence of com-
posite variable.
No association between
gas stove use and respi-
ratory illnesses and
symptoms.
No significant asso-
ciation between ill-
ness or hospital-
izations and use
of gas for cooking.
(continued on following page)
2-36
-------
TABLE 2-11. (continued)
Study Population
Outcome Measure
Results
OHIO STUDIES:
441 upper-middle class
families including
898 children under age
16 (Keller et al.,
1979a).
120 families from
first study, in-
cluding 176 children
under age 12 (Keller
et al., 1979b).
Incidence of acute
respiratory illness,
determined by bi-
weekly telephone
calls.
Incidence of acute
respiratory illness,
determined by bi-
weekly telephone calls
and validated by home
visits.
HARVARD AIR POLLUTION HEALTH STUDY:
8120 children, 6 to 10
yrs, six U.S. cities
(Speizer et al., 1980).
10,106 children, 6 to
10 yrs, six U.S.
cities. Expansion of
above study (Ware
et al., 1984).
OTHER STUDIES:
676 children, third
and fourth grades,
Arizona (Dodge,
1982).
4071 children, 5 to
14 yrs, Pennsylvania
(Schenker et al.,
1983).
History of physician-
diagnosed bronchitis,
of serious respiratory
illness before age 2, of
respiratory illness
in last year.
Same as above.
Prevalence of asthma,
wheeze, sputum, cough
as determined by parent
completed questionnaire.
Major respiratory ill-
nesses and symptoms
as determined by parent-
completed questionnaires.
Respiratory illness
incidence similar in
homes using gas and
electric stoves.
Respiratory illness
incidence similar
in homes using gas
and electric stoves.
Significant associa-
tion between current
use of gas stove and
history of respira-
tory illness before
age 2 (Odds ratio =
1.23).
Odds ratio for his-
tory of respiratory
illness before age 2
decreased to 1.12
(p = 0.07).
Significant asso-
ciation between use
of gas stove and
prevalence of cough
(prevalence rate
ratio = 1.97).
No significant asso-
ciation between use
of gas stove and any
symptom or illness
variable.
(continued on following page)
2-37
-------
TABLE 2-11. (continued)
Study Population
Outcome Measure
Results
1138 children, 6 to
12 yrs, Iowa (Ekwo
et al., 1983).
121 children, 0 to 13
yrs, Connecticut
(Berwick et al.,
1984).
231 children, 6 yrs.,
Netherlands (Hoek
et al., 1984).
Major respiratory symp-
toms and illnesses as
determined by parent-
completed questionnaires.
Number of days of
illness.
Comparison between N02
levels in homes of
cases (children
with asthma) and
controls.
Significant asso-
ciation between
current gas stove use
and hospitalization
for respiratory
illness before age 2
(Odds ratio = 2.4).
Number of days of
illness associated
with average hours of
heater use.
N02 distributions
similar in homes
of cases and
controls.
symptoms have not shown consistent adverse effects of gas stoves (Comstock et
al., 1981; Jones et al., 1983; Fischer et al., 1985) (Table 2-13).
Definitive statements concerning the health risks of indoor NO/, exposure
cannot be made at present. Many studies have examined respiratory illnesses,
respiratory symptoms, and lung function in children and adults, but their
results are not consistent and are not adequate for establishing a causal
relationship. Variations in the characteristics of the study populations and
differing end points may partially explain the differences among the studies.
NOp exposures were directly measured in only a few of the studies. Surrogate
measures of NO^ exposures, such as stove type, do not accurately classify the
exposures of individual subjects. As a result, the studies that have used such
surrogates may have been biased toward not detecting an effect.
2.2.4.4 Summary of Knowns and Unknowns for NOg. Since N02 is a NAAQS pol-
lutant, it is the subject of study under other portions of ORD research pro-
gram. Research needs to address major uncertainties in health risk assess-
ment are multiple and are separately documented by CASAC and ORD. Therefore,
these uncertainties and research needs will not be repeated here. Rather, the
discussion will focus on certainties and uncertainties regarding NOp only in
the context of indoor air. These issues are not duplicative of those relating
to outdoor air.
2-38
-------
TABLE 2-12. EFFECTS OF GAS COOKING ON LUNG FUNCTION IN CHILDREN
Study Population
Lung Function Measure
Results
808 children, 6 to 7
yrs, United Kingdom
(Florey et al., 1979).
898 children, 0 to 15
yrs, from 441 families,
Ohio (Keller et al.,
1979a,b).
8120 children, 6 to 10
yrs, six U.S. cities
(Speizer et al., 1980).
16,689 children, 6 to
13 yrs, 7 areas in U.S.
(Hasselblad et al.,
1981).
676 children, 9 to 11 yrs
Arizona (Dodge, 1982).
183 children, 6 to 12
yrs, Iowa
(Ekwo et al., 1983).
9720 children, 6 to
10 yrs, six U.S.
cities (Ware et al.,
1984).
3175 children, 5 to
14 yrs, Pennsylvania
(Vedal et al., 1984).
PEFR,
FEF25.75
FVC, FEVo-75
FVC,
FEVo.
0-75
FEVi
FEVls FEF75,
FEF25_75
FEVls FVC
in FEVj,
FVC, FEV0.75,
FEF25_75,
Vmax7S, Vmax90
No association with
N02 levels or presence
of gas stove.
Data on children not
presented separately.
No association with
presence of a gas
stove.
Overall reduction of
16 ml and 18 ml respec-
tively, for FEVj and FVC
in children from homes
with gas stoves.
Significant reduction
of 19 ml associated
with gas stove
use in older girls
only.
No effect of gas
stoves on pulmonary
level or rate of
growth.
No change after
isoproterenol
challenge in children
from homes with gas
stoves.
Significant reduction
of 0.6% and FVC of 0.7%.
Not significant after
adjustment for parental
education.
No association
with use of gas
stove.
2-39
-------
TABLE 2-13. EFFECTS OF GAS COOKING ON PULMONARY ILLNESS, SYMPTOMS,
AND FUNCTION OF ADULTS
Study Population
Outcome Measure
Results
441 upper-middle class
families, including
1054 adults over
15 yrs, Ohio
(Keller et al., 1979a).
120 families from first
study, including
269 adults over 18
yrs, Ohio (Keller
et al., 1979b).
1724 adults, ages
>20 yrs, Maryland
(Comstock et al.,
1981).
708 adults, ages >20
yrs. Nonsmoking
sample of above
population (Helsing
et al., 1982).
102 nonsmoking women in
lowest quartile of
FEVj compared
to 103 nonsmoking
women in highest
quartile, Michigan
(Jones et al., 1983).
97 nonsmoking adult
females, Netherlands
(Fischer et al., 1985).
Incidence of acute
respiratory illness,
determined by bi-
weekly telephone
calls.
Incidence of acute
respiratory illness,
determined by bi-
weekly telephone calls
and validated by home
visit.
Major chronic re-
spiratory symptoms,
, FVC.
Major chronic re-
spiratory symptoms,
FEVlf FVC.
Comparison of pro-
portions of cases and
and controls cur-
rently using gas
stoves.
IVC, FEV, FVC, PEF,
MEF75, MEF25,
MMEF.
Respiratory illness
incidence similar
in homes using gas
and electric stoves.
Respiratory illness
incidence similar
in homes with gas
and electric stoves.
Association between
gas stove use and in-
creased prevalence of
respiratory symptoms,
FEV! <80% predicted,
FEVi/FVC <70%, found
in nonsmoking males
only.
Significant associa-
tion between gas
stove use and
increased prevalence
of chronic cough and
phlegm, low FEVx/FVC.
Marginal association
between use of gas
stove and lower
lung function, (odds
ratio = 1.8, p = 0.08).
Cross-sectional analy-
sis showed an associa-
tion between current
N02 exposure and de-
creases in most pulmo-
nary function measures.
No association with lon-
gitudinal decline in
pulmonary function.
2-40
-------
Major knowns:
1. N02 levels observed in some homes can decrease pulmonary function
in asthmatics.
2. As shown in animal toxicology studies, exposure to N02 can cause
effects on lung host defenses, biochemistry, function, and
structure, and some of the effects are likely to be irrevers-
ible. Chronic exposure is a significant cause for concern.
3. Epidemiologic evidence is sufficient to suggest that exposure to
emissions from gas stoves can cause effects on the pulmonary
systems of children and perhaps adults.
4. N02 is likely to have additive or synergistic effects with other
indoor pollutants.
Major unknowns:
1. Except for N02 associated with gas stoves, the effects of N02 in
combination with other indoor pollutants are totally unknown.
Given the nature of the toxicity of N02, this is a substantial
information gap. Even if there were far more knowledge of the
effects of N02 alone, there would still be a need to assess N02's
contribution to the effects in a mixture to enable assessment of
an N02 source and development of an effective mitigation
strategy.
2. Explicit information on the number of different susceptible sub-
populations is unavailable.
3. As a whole, the epidemiologic studies on the effects of cooking
with gas stoves are not conclusive. Most of the quantitative
exposure assessment was not adequate in these studies. Other
elements of the study design also contribute to the lack of clear
conclusions. The causative agent in these studies is also not
definitive. It is assumed, for sound reasons (U.S. Environmental
Protection Agency, 1982b), to be N02, but other chemicals are
emitted, making additive and perhaps even synergistic effects
likely.
2.2.5 Sulfur Dioxide
2.2.5.1 Monitoring. Many measurements have been made of indoor versus outdoor
particulate matter, as well as organic and inorganic compounds associated with
particulate matter, to obtain a meaningful relationship between the two (van
Houdt et al., 1984; Alzona et al., 1979; Yocom et al., 1970). In general,
indoor-to-outdoor ratios (I/O) are close to unity for inhalable particulate
matter (IPM) or respirable particulate matter (RPM) when no smokers are present
2-41
-------
in the home. By contrast, I/O ratios are generally less than unity when total
suspended particulate matter is involved due to the filtering action of
crevices (Yocom et al., 1982; Yamanaka and Maruoka, 1984; Cohen and Cohen,
1980).
Indoor S02 will invariably be approximately 30 percent less than the out-
door values due to both the preponderance of outdoor sources, and the chemical
reactivity of SO^ with interior surfaces and ammonia generated by humans and
animals in the indoor microenvironment. In the special case only, where one
burns kerosene indoors in an unvented device using a poor grade of fuel that
contains sulfur, will S0? be generated indoors, and will indoor values exceed
outdoor values. Table 2-14 reports results of concentrations measured in such
homes by Leaderer et al. (1984). These results indicate that people in such
homes may be exposed to levels of SO,, that may be of concern for susceptible
individuals such as infants or the elderly patient with respiratory distress.
2.2.5.2 Health Effects. The health effects associated with exposure to SOp
in combination with particles are discussed in Section 2.3.2.4. Reports of
the effects of exposure to SO^ alone are discussed here. Exposure of animals
to levels of S02 <1 ppm have shown only decrements in pulmonary function (U.S.
Environmental Protection Agency, 1982a).
Asthmatic subjects are at least one order of magnitude more sensitive to
S02 inhalation than are otherwise normal individuals (U.S. Environmental
Protection Agency, 1982a, 1986a). Mildly asthmatic subjects have been studied,
and it has been observed that approximately 50 percent of the asthmatic volun-
teers experience at least a doubling of airway resistance at concentrations at
or below 0.75 ppm SO- (Roger et al., 1985; Horstman et al., 1986). In addition,
asthmatic subjects have been studied following very short exposure to S02, and
significant responses have been measured following exposures of only one minute
duration. Other health effects associated with exposure to SO^ are discussed
in Section 2.3.2.4, with the discussion of particulate matter.
2.3 PARTICLES AND OTHER COMBUSTION PRODUCTS
2.3.1 Introduction
The term particulate matter represents a broad class of substances both
chemically and physically. It consists of liquids, aerosols, or solid parti-
cles capable of suspension in air. Airborne particles exist in diverse sizes
and compositions that can vary widely under changing influences of source
2-42
-------
TABLE 2-14. TWO-WEEK AVERAGE S02 LEVELS BY LOCATION FOR HOMES
IN SIX PRINCIPAL SOURCE CATEGORIES*
Source Category
Location
NO K** Heater or Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
One K Heater, No Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
No K Heater, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
One K Heater, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
Two K Heaters, No Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
Two K Heaters, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
N
-
11
-
12
13
-
25
-
25
25
-
5
-
5
5
-
3
-
3
3
-
5
-
5
5
-
2
-
2
2
<
Mean
-
0.6
-
1.1
1.7
—
68.4
—
72.4
62.9
-
0.5
—
1.1
0.0
-
89.9
-
45.6
134.1
-
120.4
-
90.1
150.7
-
110.0
—
118.0
101.9
50* (ug/m3)
sb
-
2.0
—
3.7
4.2
—
86.8
—
92.3
98.0
—
0.7
~
1.5
0.0
—
91.2
—
57.3
179.4
—
66.4
-
62.4
146.5
—
81.5
-
102.4
60.7
% >80Hg/nr*
—
0
~
0
0
"•
24.0
~
28.0
20.0
—
0
—
0
0
™
33.3
~
33.3
33.3
~
40.0
~
60.0
60.0
"~
50.0
~
50.0
50.0
*Repeat monitors period data (n = 19) are included.
Samples were lost for two residences; one residence the monitors were capped
early by the residents and for the 2nd residence repeated efforts by the
interviewers to retrieve the monitors failed.
**K = kerosene
Source: Leaderer (1984a).
2-43
-------
contributions and interactive conditions (U.S. Environmental Protection Agency,
1982a). In regard to size, airborne particles tend to cluster in two groups:
coarse particles, generally larger than 2 to 3 pm in diameter; and fine parti-
cles, generally smaller than 2 to 3 urn in diameter. Chemical composition is
less definable due to interactions among particles and other pollutants,
although photochemical oxidation plays a negligible role in transforming che-
micals indoors. Some particles are highly reactive (e.g., acidic or basic) with
other pollutants or with biological systems and indoor materials. Condensations
from particle-particle, or gas-particle interactions do still occur. While the
relationship between chemical type and particle size has been relatively well
characterized for ambient air (the fine mode consisting primarily of sulfates,
organics, ammonium, nitrates, carbon, lead, and other trace metal constituents;
the coarse mode consisting of silicon components, iron, aluminum, sea salt,
and plant particles, with some overlap between fine and coarse modes) charac-
terization of indoor particles is less well defined (U.S. Environmental Protec-
tion Agency, 1982a).
2.3.2 Particles and Organics from Combustion
Combustion sources (combustion appliances and tobacco smoking) are proba-
bly the chief indoor generators of fine-mode particles which contain a host of
organic and inorganic material about which little is known. Spray and cooking
aerosols may also contribute to the total fraction of fine mode particles.
Biological contaminants, including viruses, bacteria, fungal spores and frag-
ments, pollens, fragments of house dust-mite feces, and dried, reentrained
animal secretions (e.g., urine, saliva) and animal dander, may also be found
primarily in the fine-mode fraction. Coarse-mode fractions which may consist
largely of material carried in from outdoors such as dusts, or larger biologi-
cal fragments such as mold parts or insect fragments, and which may settle on
floors and carpeting and be reentrained through human activity, have not been
well-characterized.
It has been determined that the I/O ratios are close to unity for IPM or
RPM when no smokers are present. In contrast, I/O ratios are generally less
than unity when total suspended particulate matter is involved due to the
filtering action of the dwelling's shell (Yocom et al., 1982; Yamanaka and
Maruoka, 1984; Cohen and Cohen, 1980).
2-44
-------
2.3.2.1 Occurrence and Sources.
2.3.2.1.1 Unvented kerosene space heaters. Emission rates for particles
generated by combustion from kerosene space heaters are reported in Table 2-15.
A study conducted by Yamanaka and Maruoka (1984) near the city of Kyoto,
Japan, examined the mutagenicity of the extract recovered from airborne parti-
cles inside and outside a home with unvented kerosene heaters. Samples were
collected over a one-month period inside and outside the home without a smoker,
and were examined for mutagenicity using the Ames bioassay techniques. Muta-
genic activity on the order of two- to threefold greater was found in indoor
compared to the outdoor samples, suggesting that this increase in mutagenicity
was a consequence of the kerosene heater use.
TABLE 2-15. PARTICULATE EMISSION RATES FROM KEROSENE SPACE HEATERS3
Radiant (7760 Btu/hr) Emission Rate (mg/hr)
New Unit, Normal Flame 0.16
New Unit, Low Flame 0.19
Old Unit, Normal Flame 0.13
Convective (7430 Btu/hr)
New Unit, Normal Flame <0.03
New Unit, Low Flame <0.02
Old Unit, Normal Flame 0.034
aMass of particles-from 0.005 to 0.4 urn diameter determined by electrical
mobility detector, particle density of 2.0 g/cm.
Source: Traynor et al. (1982a, 1983).
Traynor et al. (1986) measured particulate emissions of PAH and nitrated
PAH from kerosene combustion. They used a radiant and a convective heater in a
27 m environmental chamber operated at approximately 1.1 ach to collect
particulate emissions on a XAD/filter collection device operated at 6.8 m/s.
Because the particulate emissions from a well tuned convective heater are
negligible, it was necessary in this study to maladjust the convective heaters
to force particulate production. The convective heaters were maladjusted by
lifting the exterior shells approximately 1 cm to induce sooting. The well-
tuned radiant heaters' emission rates were consistent with the published data.
2-45
-------
The study reported a positive response in the bioassay analysis of the
extract of both sets of heater filter, indicating the presence of nitrated
PAH's. In addition, the radiant heater particulates were demonstrated to con-
tain highly mutagenic dinitro-PAHs (Table 2-16). Particulate matter from the
convective heater showed no such response (Traynor et al., 1986).
TABLE 2-16. NITRATED PAH SOURCE STRENGTHS FROM WELL-TUNED RADIANT
AND MALADJUSTED CONVECTIVE KEROSENE SPACE HEATERS (ng/hr)
Compound
1-nitronaphthalene
9- ni troanthracene
3-nitroflouranthene
1-nitropyrene
Radiant 1,2,4,5
280
--
1.9
47
Radiant 3
140
56
--
8.2
Maladjusted
Convective 1,2
380
--
--
aAll measurements represent total for XAD/filter extract except 1-nitropyrene
which was filter extract only.
Analytical detection limits for nitro-PAH was 1.0 ng/hr.
cDinitropyrenes (1,3-DNP, 1,6-DNP, 1,8-DNP) were measured by other researchers
using bioassay techniques to give a source strength of 0.2 ng/hr.
Source: Traynor et al. (1983).
2.3.2.1.2 Gas appliances. Gas appliances have been found to be a source of
particulate matter (Gas Research Institute, 1985; Girman et al., 1982; Traynor
et al., 1982b,c). Girman et al. (1982) reported that the average emission rate
for particles less than 25 urn was 0.24 to 0.62 ng/kJ for gas stoves. GRI re-
ported a value of 0.25 ug/kJ. The dominant element of RSP was reported by
Traynor to be carbon (Traynor et al., 1982b,c).
Table 2-17 contains particulate emissions reported by Traynor et al.
(1983).
Emissions of RSP from natural gas-fired appliances have been reported to
be on the order of 4.6 mg/hr per burner for a gas stove, 0.1 mg/hr for a gas
oven, and 0.2 to 3.2 mg/hr for unvented space heaters (Fisk et al., 1985). RSP
resulted from the use of top burners rather than from the use of the ovens
(U.S. Department of Energy, 1985).
2-46
-------
TABLE 2-17. GAS RANGE AND OVEN PARTICULATE EMISSION RATES (mg/hr)
Gas-fired burner operating 16 min, Fuel input 8730 Btu/hr
Average 4.6 , 3.8
Range 1.9 to 9.5a
2.2 to 5.7b
Gas-fired oven at 180°C for 1 hr, Fuel input rate 7970 Btu/hr
Average 0.126a, <0.42b
Range 0.118 to 0.126a
aParticles (<0.5 pm) based on electrical mobility analyser; particle
density assumed to be 2.0 g/cm.
Particles (<2.5 pm) based on gravimetric analysis of filter catch.
Table 2-18 contains reported particulate emissions for gas-fired space
heaters.
TABLE 2-18. PARTICULATE EMISSION RATES FOR A GAS-FIRED
BLUE-FLAME SPACE HEATER
Test Conditions Emissions (mg/hr)
Well-tuned, in laboratory, full input Range 0.21 to 3.23
Average 1.31
Well-tuned, in laboratory, partial input Range <0.026 to 0.0264
Average <0.93
Sexton and Repetto examined the particulate matter from cooking stoves
3
and cigarette smoke. The mutagenic density as revertants per m in Salmonella
typhimurium strain TA98 was determined for the particulate extracts. The
particle emissions from gas stoves were not found to be highly mutagenic
(Sexton and Repetto, 1982).
2.3.2.1.3 Wood-burning stoves and fireplaces. In certain parts of the coun-
try, wood-burning stoves and fireplaces have a strong influence on ambient air
quality and hence on the quality of the air entering the home. It is estimated
that in the Pacific Northwest, up to 50 percent of the homes rely on wood fuel
2-47
-------
to provide at least a portion of their their heating needs. The impact of
residential firewood use on air quality has been detailed by Lipfert (1982).
In six large cities, the impact of wood burning on ambient total suspended
particles (TSP), NO , CO, and BaP was significant. Woodsmoke from residential
/\
wood stoves can also be a prime source of mutagens in indoor air. It has been
reported that the reaction of the wood smoke with NO^ and 0, greatly increases
its mutagenic activity.
Particulate emissions from woodstoves and fireplaces vary widely, depend-
ing on the design and operation of the unit. A joint TVA/BPA study released in
1985 indicated that airtight and non-airtight wood heaters were a statistically
significant source of indoor TSP and RSP. In addition, these combustion
devices were found to contribute significant quantities of PAHs (Neulicht and
Core, 1982; Hytonen et al., 1983; Moschandreas et al., 1981a). In general,
airtight wood stoves and catalytic wood stoves contributed less pollution than
the non-airtight units.
Samples of airborne particulates collected in a room which was alternately
heated via electricity and woodburning were examined in the presence and
absence of tobacco smoking. An airtight heater was found to cause only minor
changes in the PAH concentration and the NO mutagenic activity. The most
significant increase occurred when wood was burned in an open fireplace. There
were notable mutagenic effects compared to those activities resulting from
tobacco smoke (Alfheim and Ramdahl, 1984).
2.3.2.1.4 Attached garages. The potential impact on indoor air quality from
this type of source is seen in a study by Vu Due and co-workers (1981). They
characterized motor exhausts in an underground car park using a Hi-Vol cascade
impactor to determine the size distribution of the exhaust particulate. They
reported that 60 percent of the lead and cadmium detected were located in the
particles of submicron aerodynamic diameters and that 50 percent of the PAHs
were absorbed on particles of less than 1.1 urn (Vu Due and Favez, 1981). Since
there are no significant indoor sources of lead, lead found in the indoor
environment is generally attributed to vehicle exhaust (Winchester and Nelson,
1979; Yocom et al., 1982).
2.3.2.1.5 Conclusions. A summary table of emission rates for particulates
from indoor combustion sources has been published by the U.S. Department of
Energy (1985) and is presented in Table 2-19.
2-48
-------
TABLE 2-19. EMISSION RATES FOR PARTICULATE
AND PARTICULATE-BOUND MATERIALS (mg/hr)
Source
Kerosene
Space
Heaters
Gas Space
Heaters
Appliance
Type
Radiant
Convective
Particles
0.13-0.16
<0. 03-0. 034
0.21-3.23
Benzo(a)pyrene
;:
--
Wood Heaters 2.6 1.4 x 10E-5
3.5 x IDE 3
Gas Appliance Range 1.9-30
(1 burner)
Oven 0.118-0.126
Source: U.S. Department of Energy (1985).
2.3.2.2 Exposure. The U.S. EPA's 1982 Criteria Document for Particulate
Matter and Sulfur Oxides (U.S. Environmental Protection Agency, 1982a) and its
first and second addenda (U.S. Environmental Protection Agency, 1982b, 1986a)
review the scientific bases for the National Ambient Air Quality Standards
(NAAQS) for particulate matter (PM) and SO . This assessment does not attempt
/\
to duplicate the references in the air quality criteria document but, instead,
focuses on work, published after 1969, on total human exposure to air
pollutants. The 1984 symposium sponsored by the Swedish Council for Building
Research, Stockholm (Berglund et al., 1984), provides a broad background on the
worldwide interest in indoor air pollution and human exposure.
The earliest data concerned with human exposure to indoor particles were
more related to pesticides adsorbed on particles than to the particles them-
selves (Starr et al., 1974). Fugas (1975) used indoor monitoring networks to
measure lead. This work showed some of the relationships of I/O concentra-
tions in European towns. Shortly thereafter, Binder et al. (1976) determined
that particulate dosage appeared to be caused more by exposure to indoor,
rather than outdoor, pollutants.
At the same time, evidence began to emerge concerning the indoor pollutant
load caused by cigarette smoking. Although the earliest work was more con-
cerned with CO, PM from tobacco smoking quickly became an important area of
2-49
-------
scientific interest. Dockery and Spengler (1977), Repace and Lowrey (1980),
Repace (1981), Bock (1982), Girman et al., (1982), Johnson et al. (1984), and
many others demonstrated the widespread effect of smoking on human health, both
alone and in conjunction with other air pollutants such as participates, CO,
and VOCs.
By conventional measures, the quality of ambient air has steadily improved
over the years. However, although outdoor TSP concentrations may have de-
creased, human exposures to inhalable particulates, RSP, that is, below 2.5 u
in size, probably has increased due to concentrations of fine aerosols and
ultrafine particulates generated within homes and offices. In addition,
because most people spend more than 90 percent of their time indoors, they will
be proportionally subjected to any elevated levels of pollutants from indoor
sources. Modern studies tend to assume that central monitoring stations do not
reflect or predict actual personal exposures. Continued research is needed on
the relationships between ambient concentrations and actual exposure.
PM is produced by incomplete combustion and suspension of ashes even when
combustion is complete (except for a properly tuned gas flame). Consequently,
cigarette smoking, woodstoves, and kerosene heaters will generate combustion
particles into the indoor microenvironment which add to the noncombustion par-
ticle loading. It is unfortunate that the measurement techniques routinely used
in reported studies are unable to discriminate between the combustion particles
and the other household particles generated from human activities and intrusion
from the outside air. As described by Mage et al. (1985) the particle mass
collected is an heterogeneous mixture of material, with chemical properties
ranging from inert to highly toxic and carcinogenic. Thus, a mere measure of
o
mass concentration in ug/m has little direct relationship to a health effect,
since the percentage of materials thought to be inert can range from 0 to 100
percent. Further evidence shows that much of the RSP in an indoor micro-
environment is generated by human activity (e.g., particle resuspension when
walking on carpets, shedding of hair and skin detritus, cooking, vacuuming,
etc.) so that an average taken over a 24-hour period will smooth out the high
periods of human exposure with the low periods of quiescent air when no one is
home. Spengler et al. (1985) show that personal 24-hour RSP exposures are
consistently higher than the time-weighted average concentration predicted by
24-hour indoor and outdoor values from stationary monitors.
2-50
-------
Additional complications exist because all these data are fragmentary in
time (24-hour averages), and they do not cover seasons which are known to intro-
duce trends in the data due to variations in activity patterns and air exchange
(air conditioning versus heating). Table 2-20 gives the range of indoor and
outdoor RSP concentrations and the comparative personal exposures of the people
living in the homes under study. It is interesting to note that at most per-
centiles the personal exposure exceeds the indoor microenvironment concentra-
tion; this cannot be credited to outdoor concentrations which are invariably
lower than those indoors. Other indoor RSP data from Spengler et al. (1985)
are also given in Table 2-20.
TABLE 2-20. QUANTILE DESCRIPTORS OF PERSONAL, INDOOR, AND OUTDOOR RSP
CONCENTRATIONS, BY LOCATION
City
Kingston
Harriman
Total3
RSP Sample
Group
personal
indoor
outdoor
personal
indoor
outdoor
personal
indoor
outdoor
N
133
138
40
93
106
21
249
266
71
95%
99
110
28
122
129
34
113
119
33
RSP Quantile, M9/m3
75%
47
47
22
54
45
23
48
46
23
50%
34
31
16
35
27
15
34
29
17
25%
26
20
12
24
18
13
26
20
13
5%
19
10
6
15
10
9
17
10
7
Mean
42
42
17
47
42
18
44
42
18
SE
2.5
3.5
2.7
4.8
4.1
4.0
2.8
2.6
2.1
alncludes samples from 13 subjects living outside Kingston and Harriman
Tennessee town limits and from four field personnel.
Source: Spengler et al. (1985).
2.3.2.3 Monitoring of PM. Particle samplers for outdoor ambient air monitor-
ing are large, noisy and cumbersome to use. Some continuous inhaled particle
(IP) monitors are available (e.g., the piezobalance), but also are too large,
inaccurate or too complex for THE monitoring. The smaller commercial IP moni-
tors (e.g., GCA Miniram) have poor sensitivities and/or unknown validity at
the low particulate concentrations found in nonoccupational environments.
None of the continuous (direct-reading) monitors will collect a physical sample
for subsequent chemical analysis. An accurate size-selective sampler (under
10 mm) that also can provide direct-reading information is needed.
2-51
-------
A small, unobtrusive version of the dichotomous sampler is needed for
indoor microenvironmental measurements. A true particulate PEM that collects
enough sample to accurately compare with microenvironmental measurements is
also needed. While pilot studies on real populations are now feasible with
collection-type PEMs for airborne IP matter, additional laboratory development
of this type of PEM must be conducted before precise studies are possible.
The developmental efforts for exposure monitors for inhaled particulate
matter have either been directed towards a true PEM or toward a portable
sampler that can be carried easily from one microenvironment to another.
Because of gaps in the technology, the latter approach has been pursued more
often to provide an unobtrusive PEM that collects enough particulate matter for
subsequent analyses. Fletcher (1984) reviewed many of the exposure samplers
available. Most studies have compromised by accepting the potential errors
associated with predicting particulate exposure from microenvironmental
measurements, as compared to those induced by the precision problems of PEMS
and the excessive burden of an obtrusive PEM.
A small, portable, single-channel sampler was designed for a joint effort
between Harvard and the Electric Power Research Institute (EPRI) (Dockery and
Spengler, 1981). It operates at 1.7 1/min and is provided with a 2.5-mm
cyclone inlet. This sampler has been widely used and was utilized by Harvard
in conjunction with an activity diary to relate the microenvironmental measure-
ments to exposure. The low flowrate and unavailability of inlets with other
cutpoint size are the major drawbacks to this sampler.
In an EPA-supported effort National Bureau of Standards (NBS) developed
an alternative sampler (McKenzie et al., 1982) with real channel capabilities
(2.5 and 10 urn D ) in a portable version similar in size to the Harvard/EPRI
36
sampler. This NBS version collects a 0 to 2.5 pm and a 2.5 to 10 urn fraction,
operates at 6 1/min, and is powered by a battery pack. The particle separation
between the channels is geometric, and is achieved by using an 8-(jm Nuclepore
filter followed by an absolute filter. This is less desirable than an aerody-
namic separator and requires that the coarse particle filter substrate be of
polycarbonate material.
A developmental effort is currently under way by EPA to produce a higher
flow rate PEM with interchangeable cutpoint inlets. Using the low current-
drain pump and 2.5- and 10-um inlets developed for a 10 1/min dual-channel
microenvironment sampler in the Indoor Air Program (IAP), the effort is being
2-52
-------
directed toward a sampler that is minimally obtrusive. Sampling at 10 1/min
should substantially improve the analytical capabilities and possibly permit
sampling times as short as 12 hours.
Nongravimetric samplers, which use techniques such as light scattering and
piezoelectric crystal frequency shift, are also available commercially. These
samplers have proven useful in screening studies, but all have significant
technical limitations, such as poor sensitivities or interference problems.
Their utility as exposure monitors, either as PEMs or microenvironment sam-
plers, has not been adequately demonstrated.
Table 2-21 summarizes the mass collected (at 100 percent efficiency) by
several commonly used particulate samplers for various sampling times and flow
rates.
2.3.2.4 Health Effects. The U.S. Environmental Protection Agency (1982a,
1986b) has extensively reviewed the health effects from exposure to particles.
The most extensive data are derived from epidemiological studies of outdoor
air. In these studies, the exposure assessments were limited primarily to mea-
surement of the mass of PM; only rarely was the mass characterized chemically.
Additional supporting data from animal toxicological and human clinical studies
are chemical-specific. Careful evaluation of the entire data base indicates
that PM causes adverse health effects. However, relating this entire data base
to the indoor situation is not scientifically possible, inasmuch as specific
chemicals are different indoors and outdoors, and for the most part specific
chemicals in PM have not been characterized indoors or outdoors.
However, there are chemical-specific elements of the entire PM data base
that can be useful in assessing the potential effects of exposure to indoor
PM. Examples of useful data base elements include: ETS (discussed in a
separate section), soot, particle-bound PAHs, and other combustion particles.
The respiratory system is the major route of human exposure to particles,
and its structure and function, along with the physical and aerodynamic pro-
perties of the particles, determine where and how particles are deposited,
retained, or cleared. Breathing patterns, as they relate to route and ventila-
tion level, greatly influence where the deposition of inhaled particles occurs.
The aerodynamic diameter (D ) generally has been characterized into two modes
clc
by EPA, the fine mode (<2.5 urn) and the coarse mode (>2.5 pm to ~15 urn, or
greater). These distinctions relate to particle deposition, with fine-mode and
2-53
-------
TABLE 2-21. QUANTITY OF PARTICLES COLLECTED IN MICROGRAMS
AT A CONCENTRATION OF 30 pg/m3
SAMPLING PERIOD
Sampler
Harvard/
EPRI
Harvard
Aerosol
Impactor
NBS
Prototype
EPA
Prototype
Dichotomous
Sampler
1-CFM
Andersen
Impactor
4-CFM
Andersen
8-CFM
Battell e
(Prototype)
PM10HI VOL
Type3
PEM
ME
ME
ME
A/ME
A/ME
A/ME
ME
A
Flow rate
1/m 1
1.8 3
4 7
6 11
10 18
16.7 30
28.3 51
113 204
226 408
1,133 2,039
3
10
43
32
54
90
153
612
1,224
6,118
6
19
86
65
108
180
306
1,224
2,448
12,236
12
39
86
130
216
360
612
2,448
4,896
24,473
(hr)
24
78
173
259
432
720
1,223
4,896
9,792
48,946
168
544
1,210
1,814
3,024
5,050
8,564
34,262
68,524
342,619
PEM - Personal (portable) exposure monitor
ME - Microenvironment sampler
A - Ambient
Sample divided between 2 or more substrates
Source: Rodes (1986).
small coarse-mode particles depositing principally in the thoracic (tracheo-
bronchial and pulmonary) region and larger coarse-mode particles depositing
primarily in the extrathoracic (nasopharyngeal) region during nose-breathing.
Oronasal breathing associated with minute ventilations exceeding 0.35 1/min can
significantly alter deposition patterns (U.S. Environmental Protection Agency,
1982a).
2-54
-------
The Air Quality Criteria for Particulate Matter and Sulfur Oxides document
(U.S. Environmental Protection Agency, 1982a) lists a number of possibilities
of adverse effects from inhaled particles. Included are possible irritant
effects resulting in decreased air flow as a result of airway constriction,
altered mucociliary transport, and changes i'n alveolar macrophage activity.
These effects apply across a wide range of inhaled particles, acting alone, or
in concert with common gaseous air pollutants, such as SCL, NO , or ozone.
Other toxic effects are more chemical-specific and, depending on the nature of
the chemical, may include organs outside of the respiratory tract.
Bronchoconstriction, arising from chemical and/or mechanical stimulation
of irritant neural receptors in the bronchi, has been reported as a response to
short-term exposure to high levels of various inert dusts, as well as to acid
and alkaline aerosols. Neurological receptors tend to concentrate near airway
bifurcations, where particle deposition is greatest, so that stimulation may
result in pulmonary reflexes, such as bronchoconstriction and coughing. These
reflex actions may be related to the effects observed in various epidemiological
studies, such as aggravation of chronic respiratory disease states including
asthma, bronchitis, and emphysema. Individuals with asthma or emphysema and
other respiratory diseases may have increased particulate deposition due to
altered breathing patterns or airway structural changes, which may then contri-
bute in a cascading effect to even more bronchoconstriction and particulate
deposition (U.S. Environmental Protection Agency, 1986a).
New information discussed in the second addendum to the 1982 Air Quality
Criteria for Particulate Matter and Sulfur Oxides document (U.S. Environmental
Protection Agency, 1986a) supports the conclusions of the earlier document.
Newly available studies that classify thoracic deposition and clearance of large
particles are presented; deposition during oronasal breathing and deposition in
possibly susceptible subpopulations such as children are assessed, and new in-
formation that could lead to health effects data refinement or reinterpretation
are evaluated.
Tables 2-22 and 2-23 summarize the quantitative health data derived from
epidemiologic studies reviewed in the second addendum to the 1982 criteria
document on PM and SO (U.S. Environmental Protection Agency, 1986a).
J\
It should be emphasized that the health data obtained for PM are generally
the result of epidemiologic studies that correlate health effects observations
in an exposed population with exposure to ambient air pollutant levels. The
2-55
-------
ro
i
01
CTi
Morbidity
TABLE 2-22. SUMMARY OF KEY QUANTITATIVE CONCLUSIONS BASED ON NEWLY AVAILABLE EPIDEMIOLOGICAL STUDIES OR
ANALYSES RELATING HEALTH EFFECTS TO ACUTE EXPOSURE TO AMBIENT AIR LEVELS OF S02 AND/OR PM
Type of Study
Mortality
Results Obtained
Indications of increased mortality during
24-hr
BS*
<500
average pollutant level
TSP
—
(ug/m3)
S02
-700-750
Reference
Mazumdar et al. (1982)
London winters of 1958-59 to 1971-72, with
most marked S02 effects evident at ~700-750
ug/m3 and indications of small increases at
BS levels <500 pg/m3 and possibly as low as
150-300 ug/m3.
New analyses of same 1958-59 to 1971-72
London winter mortality data indicative of
increased mortality at BS levels <500 pg/m3
and no evident threshold at 150 ug/m3.
Unpublished reanalysis of same 14 year London
mortality data using spectral transform multiple
regression analyses confirming significant asso-
ciations for total, cardiovascular and respira-
tory mortality, accounting for autocorrelation
and temperature. Suggestion of more pronounced
effects with 7-21 day cycles of exposure.
Unpublished reanalysis of same 14 year London
mortality data using regression analyses that
detrended data for time series autocorrela-
tion, humidity, and temperature indicating
significant associations between mortality and
BS to below 100 ug/m3, but not for S02 at <500
ug/m3.
Evidence for reversible (-2-3 wk) small (2-3%),
but statistically significant decrements in FVC
of school children following episodes in Steuben-
ville, Ohio when 24-h TSP and S02 levels respec-
tively ranged up to 220-420 and 280-460 ug/m3,
but not after "sham" episode with TSP = 160 and
S02 = 190 ug/m3. Larger decrements seen in sub-
set of children.
Evidence for reversible (~2-3 wk) small (3-5%),
but statistically significant decrements in pul-
monary function measures (FVC, FEV^o MEF) for
school children in the Netherlands during and
after pollution episode when 24-h TSP, RSP, and
S02 levels ranged up to 200-250 ug/m3, but no
effect shortly after day when same pollutants
averaged 100-150 ug/m3.
<150-500
<500
Ostro (1984)
Shumway et al. (1983)
Continuous
association
from lowest
(<100 ug/m3)
BS levels
>500
Schwartz and Marcus (1986)
220-420
280-460
Dockery et al. (1982)
200-250
200-250
Dassen et al. (1986)
Source: U.S. Environmental Protection Agency (1986a).
-------
TABLE 2-23. SUMMARY OF KEY QUANTITATIVE CONCLUSIONS BASED ON NEWLY AVAILABLE EPIDEMIOLOGICAL STUDIES
RELATING HUMAN HEALTH EFFECTS TO LONG-TERM EXPOSURES OF S02 AND/OR PM
Type of Study
Results Obtained
Annual-Average
Pollutants
Levels (|jg/m3)
TSP
SO,
Reference
Initial Cross-Sectional
Analyses of Ongoing
Longitudinal Study
cr,
—i
Cross-Sectional
Study
Longitudinal Study
in Southwestern U.S.
towns
Increased rates of cough, bronchitis
and lower respiratory disease (in the
absence of lung function changes) among
school children in 6 U.S. cities signi-
ficantly associated with annual-average
TSP levels across range of approximately
30 to 150 ug/rn3 when analyzed for between
city effects but not in relation to PM
gradients within individual cities.
Effects most clear for highest PM areas
(-60-150 ug/m3) versus lowest (-40-60 (jg/m3).
No significant association with S02 except
for cough.
Significantly increased rates of
persistent cough and phlegm (PCP) among
young adults^associated with annual-
average S02 =115 jjg/m3 in highest expo-
sure Utah community versus 3 lower expo-
sure towns with S02 in 11-36 ug/m3 range.
No TSP gradient across four communities.
Effects possibly due to intermittant high
S02 peaks.
Significantly increased prevalence of
cough among children from highest pollu-
tion area (annual average S02 = 103 ug/m3;
intermittant 3 h peaks often exceeded 2,500
ug/m3 or ~1 ppm) in comparison to lower
pollution towns (annual S02 = 14 ug/m3).
No TSP gradient across high and low pollu-
tion towns. Effects possibly due to inter-
mittant high S02 peaks.
-60-150
Ware et al. (1986)
115 Chapman et al. (1985)
103
Dodge et al. (1985)
*BS = British Smoke Method
Source: U.S. Environmental Protection Agency (1986a).
-------
studies do not specifically involve measurement of indoor particles and, hence,
do not assess effects from indoor air pollution. Furthermore, extrapolation
of health effects from epidemiologic studies of ambient particle pollution
should be approached with great caution because sampling methods for particles
outdoors do not correlate with one another and often cannot be directly compared
with the methods developed for indoor particle sampling.
Particles studied indoors have primarily been those in the fine-mode,
primarily resulting from cigarette smoking, or as emissions from combustion ap-
pliances. Coarse-mode particles resulting from reentrainment of fibers, dust,
house dust mite fecal pellets, animal and human dander, mold spores and frag-
ments, probably constitute the second most common form of indoor particle pol-
lution. Methods for routine sampling of many of these types of particles have
not been developed, and their health impact is, for the most part, unknown.
2.3.2.4.1 Health effects associated with exposure to soot. Virtually every
combustion process in which carbonaceous fuels are burned results in the
production of soot. Soot often consists of clusters of very small (~30 nm)
spheroids fused together to form particles with a 0.1 to 0.2 u diameter. These
soot particles contain an organic fraction (solvent extractable) and an ele-
mental carbon fraction which appears to be similar to graphite. The combus-
tion of diesel fuel, coal, fuel oil, and wood has been shown to produce soot
particles with extractable organic matter that is mutagenic in short-term bio-
assays and, in several of these sources, has been shown to be tumorigenic in
animals. Chemical characterization of these organics shows that they contain
carcinogenic PAHs, methylated PAHs, nitrated PAHs, and oxidized PAHs.
Tokiwa et al. (1985) reported the presence of highly mutagenic and carci-
nogenic dinitropyrenes in the emissions of kerosene heaters, gas burners, and
liquefied petroleum gas burners. The organic extracts of these emission
particles showed mutagenic activity in the Ames Salmonella mutation assay in
strains TA98 (1.5 to 5.3 revertants/pg) and TA97 (2.0 to 6.0 revertants/ug)
without metabolic activation of S9. The organic extracts from kerosene heater
emissions also showed positive in TA1538 and TA100 in the absence of S9. More
mutagenic activity was found in the emission particles collected in the first
20 minutes of burning.
Higher mutagenic activity has been reported in samples extracted from air
in homes' that burn wood than from those that do not (Morin, 1985). Previous
studies by Lewtas (1985) have shown that the organic extracts of the emission
2-58
-------
particles collected after a dilution tunnel from the air-tight wood stove were
mutagenic in the Ames Salmonella mutation assay (1.3 revertants/ug for pine,
and 0.93 revertants/ug for oak) in strain TA98 with metabolic activation of S9
mitochondrial fraction. The organic extract of the wood combustion emission
using pine also showed dose-response skin tumor initiation activity in Sencar
female mice in a 6-month study (Lewtas et al., unpublished data).
Occupational studies have shown the increased risk of lung, larynx, and
skin cancer after exposure to coal soot, and increased incidence of scrotum and
bladder cancer after exposure to coal tar in gashouse workers, stokers, as-
phalt, coal tar and pitch workers, coke-oven workers, and chimney sweeps (Cole
and Goldman, 1975). Nonoccupation-related lung cancer has been linked to the
exposure of indoor unvented coal combustion emissions in a study conducted in a
rural county, Xuan Wei, located in Yunnan province in China (Mumford et al.,
1987). The Xuan Wei residents have been exposed domestically to smoke from
unvented coal and wood combustion. The county has unusually high lung cancer
mortality rates that cannot be attributed to tobacco use or occupational
exposure. The Xuan Wei women, mostly nonsmokers, have the highest lung cancer
rates in China, and men's rates are among the highest. Lung cancer mortality
rates are associated with the usage of smoky coal (comparable to U.S. medium-
volatile bituminous coal) but not associated with the usage of wood. This is
in agreement with the reports by DeKoning et al. (1984), who indicated no
association of lung cancer with domestic burning of biomass (including wood,
crop residues, and dung) in other rural areas of developing countries.
Studies of the noncarcinogenic effects of soot are relatively rare and
consist primarily of animal inhalation toxicology studies of fly ash that was
collected, stored, and reaerosolized (U.S. Environmental Protection Agency,
1982a). Generally, these studies showed that fly ash is of low toxicity com-
pared to other aerosols such as sulfuric acid and of metals. However, high
o
levels (>1 mg/m ) did cause several types of pulmonary effects in animals after
3
prolonged exposure. High-level exposure (>5 mg/m ) to diesel particles also
results in pulmonary effects after chronic exposure (McClellan et al., 1986).
2.3.3 Polycyclic Aromatic Hydrocarbons
Polycyclic aromatic hydrocarbons (PAHs) represent a diverse group of
combustion products whose common characteristic is a nucleus of five- or six-
membered carbon rings in which interlinked rings have at least two carbon atoms
2-59
-------
in common (Zander, 1983). The number of rings varies from two to many.
Configuration of ring structures determines physical and chemical properties
and biological activities. Some examples of simple PAHs are depicted below.
phenanthrene
anthracene
benzo(a)pyrene
The PAHs are of specific human health concern because many of them are
procarcinogenic, cocarcinogenic, or carcinogenic, and they affect the immune
and cardiovascular systems. They are breathed into the lung as volatiles or
they can adsorb to the surface of deposited particles and be inhaled.
2.3.4 Other Combustion Organics
PAHs can form many substituted compounds, which also may be pro-, co-, or
frankly carcinogenic. In addition, combustion products include heterocyclic
compounds, as well as compounds of nearly every class known, including alde-
hydes, ketones, nonpolycyclic hydrocarbons, aromatics, organic acids, and
nitrosamines, among others.
2.3.5 Interaction of Particulate Matter and Organics
Beyond the specific issues discussed above, one of the concerns for both
carcinogenic and noncarcinogenic effects of PM is related to the organics bound
to the particles. The particles and bound organics have properties that influ-
ence deposition, retention, and bioavailability. Hence, the specific chemical-
physical characteristics of the particles and organics have major influence on
the health effects. One of the most classical examples is that
BaP alone has low carcinogenic potency; ferric oxide alone is not a carcinogen.
When BaP is coprecipitated with ferric oxide and instilled into the lungs of
animals, cancers result. Many particles adsorb organics as they are created in
combustion processes. However, once within the air, they can adsorb other
2-60
-------
materials (i.e., radon, pesticides, other organics from home products). These
absorbed materials will have different pulmonary deposition sites and deposi-
tion efficiencies when adsorbed on a particle, as opposed to existing in
vapor phase. Retention and pharmacokinetics would also be affected. Thus,
even an inert particle could contribute (in a currently unknown way) to health
risks from indoor air. Assessment of the risks of individual pollutants does
not take into account such complex interactions. Risk assessments on complex
mixtures must be done if a clear picture of the health effects of indoor pol-
lutants is to emerge.
2.3.6 Monitoring
Concentrations of polynuclear aromatic hydrocarbons (PAHs) and related
compounds, like those of the volatile organics (Pellizzari et al., 1982;
Wallace, 1986b), have often been found to be higher indoors than outdoors
(Chuang et al., 1987; Wilson et al., 1985; National Research Council, 1986c).
Because many of these compounds are known carcinogens, and because many more
show potential carcinogenicity (as evidenced by mutagenicity or other bio-
assays) exposure to them is of major concern. Sources of PAHs in the indoor
environment include combustion devices, such as furnaces, woodstoves, fire-
places, and kerosene heaters. Additional contributions to the PAH levels can
come from personal activities, which include cooking and especially smoking.
There may also be minor contributions from other sources, such as building
materials, furnishings, polishes, and waxes.
Because the PAHs and other SVOCs are distributed in air between the par-
ticulate and vapor phases, depending on the vapor pressures of the compounds,
the nature of available adsorptive surfaces, temperature, and other environ-
mental conditions, sampling systems for PAHs must collect both particles and
vapor. Thus, a filter may be used to collect the particles with the filter
followed in series by a vapor trap, containing a sorbent such as polyurethane
foam (PUF) or XAD-2 resin to collect that portion of the PNAs that is in the
gas phase (Chuang et al., 1987a,b; Wilson et al., 1985). Analytical procedures
usually involve separate extraction of the filter and sorbent, with subsequent
analysis by high pressure liquid chromatography (HPLC) or gas chromatography/
mass spectroscopy (GC/MS). These traditional sampling techniques may, however,
greatly disturb the phase distributions existing in the air at the time of
sampling. Consequently, particulate-associated chemicals may be underestimated,
2-61
-------
while the vapor phase concentrations may be inflated. Equivalent results for
total PAH concentrations (particles plus vapor) are obtained by pooling the fil-
ter and sorbent extracts prior to analysis or by analyzing the extracts sepa-
rately and adding the results together. Pooling the extracts eliminates the
possibility that an inaccurate assessment of phase distribution will be made
and should improve analytical detection capabilities.
The utility of the sampling system just described was evaluated in a small
indoor monitoring study performed in Columbus, Ohio in 1985 (Chuang et al. ,
1987b). A sampling head identical to those used in the PS-1 ambient air sampler
(General Metal Works, Cleves, Ohio) was located indoors and connected by vacuum
tubing through a window port to an outdoor pumping system. The sampling head
contained a quartz fiber filter followed by a PDF adsorbent cartridge. Samples
were obtained at a flow rate of 7 cfm (200 1/Min), which allowed collection of
quantities sufficient for both chemical analysis and bioassay in eight hours,
yet was low enough that only 5 to 10 percent of the total air volume in a given
house was sampled. Analyses were performed on the sample extracts without
cleanup, using chemical ionization GC/MS. The sample size was more than ade-
o
quate for most of the 17 target PAHs, which were present in the 1 to 3 ng/m
o
range for large molecules, such as benzo(e)pyrene, and in the 20 to 150 ng/m
range for smaller molecules, such as phenanthrene and fluoranthene. To analyze
for the two target nitro-PAHs, 1-nitropyrene and 2-nitrofluoranthene, which were
o
present in the 10 to 100 pg/m range, it was necessary to pool three 8-hr ex-
tracts and employ negative chemical ionization mass spectrometry. Sample sizes
were barely adequate for microbioassay (Lewtas et al., 1987).
Subsequent studies have indicated the desirability of going to a lower
flow rate, 1 to 2 cfm (28 to 57 1/min) to minimize perturbation of the indoor
air environment by the sampling itself. These flow rates should still provide
adequate sample for chemical analysis, but if bioassay is planned, sampling
over longer time periods will be necessary to collect adequate material.
The adsorbent used in the indoor monitoring study described above was PUF.
Good collection efficiencies and sample recoveries (>80 percent) were obtained
for PAHs having three rings or more. More studies of the relative suitability
of PUF and XAD-2 resin for the collection and quantification of PAHs (Chuang
et al., 1987b) have shown that XAD-2 is preferable to PUF for sampling of
PAHs with three rings or fewer, both because it adsorbs a greater fraction of
the sample, and because a smaller fraction is lost by volatilization after
2-62
-------
collection. Sample storage in sealed containers at 20 to 60°C in the dark for
periods up to thirty days does not appear to produce significant losses of most
PAHs from XAD-2 or PUF. Similar storage of filters does not lead to significant
sample losses, except for benzo(a)pyrene and cyclopenta (c,d)pyrene, which are
gradually lost, presumably because of chemical reaction. The level of
benzo(a)pyrene, for example, decreases in 30 days to about 40 percent of its
initial value. Extraction and cold storage of the filter extracts soon after
collection can minimize this problem. The sampling efficiencies, recoveries,
and storage stabilities for collection media used to collect the substituted
PNAs and other semi volatile compounds comprising the more polar constituents
of air samples have not been thoroughly investigated.
Although the nonpolar fraction of air samples generally contains most of
the PAHs, it only accounts for about one-third of the mutagenicity observed in
bioassays. Other compounds, including some substituted and derivatized PNAs,
are found in the polar fractions and must account for the remaining bioactivity.
Many of these compounds have not been characterized chemically, and the analy-
tical methods necessary for their separation, identification and quantifica-
tion, have not been fully developed. Therefore, a great deal of methods devel-
opment research remains to be done before confident assessments of exposure to
these carcinogens and toxic chemicals can be made.
One of the most striking findings of the indoor air methods evaluation
studies mentioned above (Wilson et al., 1985; Lewtas et al., 1987) was that of
the variables—fireplace use, gas or electric heating system and appliances,
air exchange rate, and cigarette smoking --the experimental variable that had
the greatest effect on both PAH levels and on mutagenicity was smoking.
Mainstream cigarette smoke and its effects have been well-documented in the
literature. However, the chemical characteristics of sidestream smoke and the
associated exposures have not been as thoroughly studied. Many research needs
are apparent (National Research Council, 1986c). The needs of exposure mea-
surement methods include reliable methods for collection and quantification of
nicotine, which can serve as a marker for tobacco smoke gas phase constituents,
identification of marker compounds for the particulate phase constituents, and
development of sampling and analytical methodology for these markers. Analy-
tical methodology for the more polar PAHs and other semivolatile compounds
associated with ETS must be developed or improved.
2-63
-------
To ascertain true exposure to the PAHs and other semi volatile compounds in
indoor and ambient air, it is necessary to know whether they are in the gas
phase or are associated with particles, so that the particle size distribution
determines to a large extent whether they are retained in the body. Few careful
and accurate studies of the phase distribution of these compounds have been done
(National Research Council, 1986c). Development and evaluation of sampling
methodology that will quantitatively separate the vapor and particles, and that
will maintain the integrity of the separately collected portions of the samples,
is necessary. Preliminary work on denuder-based samplers to determine phase
distributions has been done (Coutant et al., 1987), but further development is
needed. Since sampler design affects the subsequent methods of analysis further
analytical methods development for studies of the phase distribution of PNAs
and related compounds is implied.
2.3.7 Woodsmoke
Woodsmoke is made up of a complex mixture of compounds, including alde-
hydes, such as acrolein, and PAHs, many of which are mutagenic. Some studies
from developing countries indicate an association between high-level smoke
exposure (i.e., relative to the U.S.) in dwellings and chronic pulmonary
disease.
2.3.7.1 Health Effects. Some studies from developing countries indicate an
association between intense smoke exposure in dwellings and chronic pulmonary
disease. It should be noted that the following studies describe effects from
smoke exposure in unventilated dwellings, so that concentrations are orders of
magnitude greater than those expected indoors in the U.S. In a house-to-house
survey of adults in a village in Nepal, Pandey (1984a,b,c) found that the
prevalence of chronic bronchitis increased with the extent of domestic smoke
exposure, as measured by the number of hours spent daily near the stove. In a
subsequent study, Pandey et al. (1985) evaluated respiratory function of 150
women ages 30 to 44 years from two rural villages in Nepal. Domestic smoke
exposure appeared to adversely affect lung function of smokers but not of
nonsmokers. Epidemiological and clinical studies involving adults in New
Guinea have also suggested adverse effects of domestic smoke exposure (Master,
1974; Anderson, 1979).
2-64
-------
Respiratory effects of woodsmoke have also been examined in children from
these countries. Anderson (1978) studied 1650 children drawn from two con-
trasting New Guinea communities, one at sea level where wood was not burned
indoors and one in the highlands where wood was commonly burned indoors. A
cross-sectional survey showed no differences between the two groups on spiro-
metric testing, by physical examination, or by clinical history. During a
30-week surveillance period involving some of the wood smoke-exposed children
and an unexposed control group, Anderson (1978) did not find a consistent
relationship between woodsmoke exposure and respiratory abnormalities. In
contrast, Kossove (1982) reported that Zulu infants under 13 months of age with
severe lower respiratory tract disease were twice as likely to have a history
of daily heavy smoke exposure compared to unexposed control infants. It has
been estimated that about one-half of the world's households cook and/or heat
daily with biomass fuels (Smith, 1986). In most cases, the fuels are burned
under unvented conditions. The major health effects from high exposures to
biomass combustion products are chronic obstructive lung disease, heart
diseases including cor pulmonale, and acute respiratory infection diseases
which cause high infant mortality and low birth weights.
While these studies implicate domestic smoke exposure as a possible risk
factor for the development of respiratory disease in developing nations, their
results should not be generalized to developed nations, where exposures are
generally much lower and biomass fuels are not widely used.
Data on health effects of residential wood combustion in the U.S. are
sparse. Honicky et al. (1983) described the case of an infant who was repeated-
ly hospitalized for severe lower respiratory tract disease characterized by
wheezing and pneumonia. The child's illnesses stopped when the parents removed
the woodstove from the home. This case prompted Honicky and colleagues (1985)
to conduct a prevalence study of respiratory symptoms in 62 Michigan children,
31 from homes with and 31 from homes without woodburning stoves. The propor-
tion of children with moderate or severe symptoms during the previous winter
was much greater in the group from homes with woodstoves: 84 percent had at
least one severe symptom as compared to 3 percent of the control group. How-
ever, in a similar study in Massachusetts, Tuthill (1984) found no association
between use of a woodburning stove and chronic respiratory disease, respiratory
symptoms, and excess respiratory symptoms. Woodsmoke components were not mea-
sured in the subjects' homes in either investigation.
2-65
-------
2.3.8 Major Knowns and Unknowns: Health Effects
This section only addresses the health effects of PM as they relate to
indoor exposures. Thus, it does not include issues critical to outdoor PM. PM
as part of ETS is discussed in Section 2.4.
Major Knowns:
1. There is a cancer potential associated with exposure to PM.
2. The relationship between exposure and dose of PM is heavily
influenced by several factors including activity patterns, age
of the person, and disease status of the person. Organics
adsorbed onto particles also have a different dosimetry than the
organics alone, thereby influencing both risk assessment and
control strategies.
3. It can be hypothesized that particles from incomplete combustion
have a noncancer health risk potential.
Major Unknowns:
1. There is evidence that there is a cancer potential associated
with exposure to particulate matter, but the dimensions of that
potential are not known.
2. The quantitative relationship between exposure to PM and deliv-
ered dose to man is relatively unknown for the indoor air
situation. There is insufficient knowledge of personal activity
patterns indoors, especially exercise levels, to permit applica-
tion of existing and future pulmonary dosimetry models to
evaluate the relationship between exposure and dose. The
dosimetry, pharmacokinetics, and bioavailability of indoor
organics adsorbed into particles is not known. These gaps in
the data base prohibit even semiquantitative application of
models relating monitored exposure to dose, dose being the
factor that is critical to health effects. For example, ten
people within an identical indoor environment, receiving identi-
cal exposures, could receive very different doses and hence
experience different degrees of health risk, depending on
whether these people had different activity patterns, different
baseline pulmonary functions (i.e., males versus females, nor-
mals versus those with existing pulmonary disease), or different
stages of lung development (i.e., child versus adult).
3. The noncancer health effects of indoor combustion particles,
except for ETS, are almost totally unknown. Even if indoor PM
was better characterized chemically, these effects would still
be unknown, given the paucity of the general inhalation toxicol-
ogy data base and the need to address PM in context of the
2-66
-------
mixtures they are in. Given what is known, this huge gap in the
data base is cause for concern.
2.3.9 Mitigation and Control Options for Combustion Sources
Indoor air controls that have application to combustion sources can be
grouped in five general categories (Table 2-24).
TABLE 2-24. METHODS OF IAQ CONTROL FOR COMBUSTION SOURCES
Control Method Examples
General Ventilation Infiltration of outdoor air, whole building venti-
lation with outdoor air by mechanical equipment
Local Ventilation Exhaust fan near a source (kitchen fan, range hood)
Air Cleaning Filtration or electrostatic devices for particle
removal in central air handling systems, absorption
devices for gaseous pollutant removals
Source Modification Proper tuning and operation of combustion appliances
Source Removal Substitution of alternative nonpolluting source of
heat
Building Design Detachment of parking facilities, relocation of
and Operation air intakes, maintenance of ventilation systems
General ventilation provides a multiple point source control by dilution
with outdoor air. This type of air exchange can also introduce outdoor air
pollutants into the occupied space. Therefore, general ventilation is only
applicable when outdoor air has lower pollutant concentrations or if the air is
cleaned prior to introduction. General ventilation can also make air quality
in some areas of a building worse through the distribution of a pollutant. The
energy requirements associated with whole-building ventilation make it imprac-
tical to increase the air change per hour (ach) to greater than 1 through mecha-
nical means.
Avoiding unusually low ach is equally important for maintaining acceptable
indoor air quality because indoor concentrations of contaminants appear to
increase rapidly as ach decreases.
Local ventilation can be regarded as a source control measure for combus-
tion sources because it effectively reduces pollutant source strength. Local
2-67
-------
exhausts have been demonstrated to minimize transport of pollutants from a
source to an occupied space. Local ventilation can maintain low indoor pollu-
tant concentrations for a concentrated source with less exhausted air and
consequently less cost than general dilution ventilation. Local ventilation
may have particular application to episodic operation such as occurs during the
reloading of a wood burning stove.
Few data are available on the effectiveness of local exhaust from specific
rooms. Evidence indicates that local exhaust can reduce transport of tracer
gases to an adjoining room even when the doors between the rooms remain open,
however, significant air flow rates are required. The practicality of exhaust-
ing a residence from an individual room has not been demonstrated.
Local exhausts such as range hoods are an important reduction strategy for
NOp. Range hoods have been demonstrated to reduce gas stove NO levels in
^ /\
kitchens by 30 percent at a flow rate of 93 1/s. The use of a range hood is
demonstrated to be highly effective and to have near a linear relationship
between range hood efficiency and flow rate. Such use should also effectively
reduce concentrations of cooking aerosols.
Revza (1984) utilized sodium hexafluoride (SF6) as a surrogate pollutant
source to evaluate the efficiency of range hoods and window fans operated as a
local exhaust in a two-room test space (ach <0.1/hr). Using a range hood and
both heated and unheated tracer gases he compared the concentrations of SF6
measured at multiple points within the space to those predicted by a simple
two-room mass balance model. Revza found ventilation efficiency to be roughly
linear over flow rates of 10.3 to 60 1/s for the heated tracer. The highest
efficiency was 77 percent with heated tracer gas. Revza inferred that a flow
rate approaching 75 to 100 1/s would produce 100 percent removal of the unheat-
ed tracer. With unheated tracer gas, the effectiveness of the range hood was
found to be highly dependent on the ambient temperature. The variable buoyancy
of the tracer made the calculation of an efficiency impossible (Revza, 1984).
Window fans were tested with the source in each of the two rooms and the
fan remaining fixed. The fan was exhausting at a rate varying from 10.3 to
45.2 1/s. When the source was in the same room as the fan, the SF6 concentra-
tions were found not to agree with the predictive model; when the source was in
another room, fairly good agreement with the model was obtained. In neither
case did the source differ by more than 50 percent from the predictive model
(Revza, 1984).
2-68
-------
Although use of air cleaners is an established and growing practice for IA
control it does not appear to be directly applicable to inorganic combustion
gases. There are two basic modes for applying air cleaners to indoor air. The
first is to interpose the air cleaner between the source and the indoor envi-
ronment, e.g., a recirculating range hood. The second is to treat the indoor
air after the contaminant has been dispersed, e.g., central heating, ventila-
tion, air conditioning (HVAC), or free-standing air cleaners. These techno-
logies have potential application for removal of RSP and unburned hydrocarbons
(UNBH), however, their efficiency and capacity for removal of specific pollut-
ants have not been evaluated in indoor situations.
Residential furnace filters and panel filters such as those used by some
portable air cleaners are not effective for particle removal. Electrostatic
precipitators are demonstrated to be 50 to 90 percent efficient and extended
surface area filters may remove 70 to 99 percent of the particles (Fisk et
al., 1985).
The source modification control option for combustion gases has not been
studied fully, even though some rather detailed work exists for the control of
NO through modification of gas appliances. Modification can include replace-
/\
ment of standing pilot lights with other ignition devices such as electronic
ignitions.
Zawacki et al. (1984) investigated simple methods of reducing NO emissions
J\
through modification in typical unvented gas space heaters. They reported that
a significant reduction in NO could be obtained using appropriate burner
/\
inserts. This approach was not suitable for reduction of N02 (Zawacki et al.,
1984).
Burner modifications, such as the use of wire mesh in the flame to conduct
heat away from the hottest portion of the flame to reduce N02 formation, must
be done in a way not to reduce the flame temperature. One mesh configuration
has been shown to reduce N0? emissions by 70 percent with the concurrent 7500
percent increase of CO (Fisk et al., 1985). A double cylinder configuration
was reported to reduce NO by approximately 62 percent while allowing CO
/\
burnout to proceed to completion; steel ring inserts in a range top burner
demonstrated a 26 percent reduction in NOp without a dramatic increase in CO
(Fisk et al., 1985). The basic strategy of burner inserts has been adopted
recently in a modification introduced by the gas industry to meet the new
California Standards for controlling N02 emissions from furnaces and water
heaters (Fisk et al., 1985).
2-69
-------
Source removal or source exclusion has application for the control of
indoor air contamination arising from combustion sources. For example, the
temperature for the formation of NCL which is produced in kerosene combustion
appliances and in gas flames is not produced in appliances using electrical
resistance elements. CO is also eliminated through the replacement of combus-
tion sources with an appropriate electrical counterpart. S02 is produced by
combustion of a fuel with a high sulfur content. It can be controlled through
source elimination or fuel substitution.
Although not a source control, appropriate building design and operation
as a control option is the goal of many building professionals. Design may
include the location of specific subunits and systems as well as the selection
of building materials. Because CO is a major exhaust pollutant from internal
combustion engines, the isolation of garages and loading docks from occupied
building is an acceptable design strategy. Where this cannot be accomplished
it appears essential to monitor the leakage paths into the occupied areas for
CO. The location of the air intakes and the location and condition of the
ducting system are also crucial. Most air pollutants appear to increase in
concentration from the air intake along the path of the duct work of the
ventilation system (Berglund et al., 1982).
The deliberate use of a pollutant's reactivity also has potential as a
design strategy. However, before reactivity can be considered as a control
strategy it is essential to understand the mechanism of a pollutant's genera-
tion and decay. Indoor air pollutant concentration is a function of pollutant
source strength, air transport and diffusion, and the chemical reactions occur-
ring within and on the various surfaces of a building. The chemical reactivity
and the internal air flow are both affected by temperature, and reactivity and
may be affected by humidity as well (GEOMET Technologies, Inc., 1976).
The use of reactivity to control N0? has been proposed by Fisk and co-
workers (1985), who reported a removal rate by reactivity of 0.16 to 1.29 per
hour for NOp. Sulfur dioxide is also a highly reactive compound and may be
differentially adsorbed by interior walls, ceilings, and floor coverings.
Consequently, design changes which increase air circulation or inclusion of
specific surface material may be helpful in the removal of SOp ( Fisk et al.,
1985).
Proper operation and maintenance are also important control options. In
those occupied spaces affected by leaking combustion products, the replacement
2-70
-------
of cracked heat exchangers in furnaces, or improvements to the venting system
can provide control.
There are basically two strategies for RSP control. The first is source
exclusion or reduction; the second is the removal of RSP from the air.
Reduction in the emissions from wood stoves and fireplaces can be accomplished
through structural changes such as wood stove inserts in a fireplace, or by
replacing fireplace screens with tempered glass doors, or by operational
changes, such as mode of loading, amount of air supplied, or type and dryness
of wood burned.
2.3.9.1 Major Knowns and Unknowns. The general absence of data and the
absence of reliable particulate emission factors from the indoor sources makes
it impossible to establish a relative ranking between reported sources and
their contribution to indoor air pollution. In general, the absence of stan-
dardized test procedures, including sampling and analysis methods, and lack of
uniformity in test conditions precludes the use of much of the existing data
for the predictive modeling which facilitates risk assessment calculations.
Major sources of uncertainty related to indoor air combustion particulates
are the source and emission rates of particulates, the characteristic size
distribution, the presence and identity of absorbed chemical compounds, and the
applicability of a given control option to a specific pollutant source.
Indoor air source characterization unknowns related to particles emitted
from combustion sources include the following:
characterization of various sources with particular emphasis on
size and mass distribution of particles produced by unvented
sources
expanded characterization of particle-bound organics from com-
bustion devices
survey of indoor combustion sources to gather statistics on age,
condition, and operational parameters of combustors
investigation of the factors affecting leakage rates of parti-
cles from vented combustion sources
studies of sink rates of particles on indoor materials
2-71
-------
2.4 ENVIRONMENTAL TOBACCO SMOKE
2.4.1 Introduction
Tobacco smoke is a major contributor of particulate and organic matter in
the indoor environment (Samfield, 1985; National Research Council, 1981b;
Repace, 1982). The health effects of smoking on smokers has been extensively
studied, but the effect of passive smoking on nonsmokers has received far less
attention (Taylor et al., 1978; Surgeon General of the U.S., 1986; National
Research Council, 1986a).
Smoking contributes enormously to RPM in buildings. The following are
some results from the Harvard Six Cities Study (Spengler et al., 1981;
Samfield, 1985) which indicate that smoking can triple I/O ratios of RPM.
Mean
Concentration
Location ug/m3 I/O Ratio
Outdoor 21.1
Indoor - no smoking 23.4 1.16
Indoor - 1 smoker 36.5 1.73
Indoor - 2 smokers 70.4 3.34
Nonsmokers inhale environmental tobacco smoke, which is the combination of
sidestream smoke (SS) released into the air from the cigarette's burning end,
and mainstream smoke (MS), exhaled by the active smoker. Because tobacco
smoking releases myriad chemical species into the air, ETS is an extremely
complex mixture that changes as it ages. The exposures to passive and active
smoking differ qualitatively and quantitatively (U.S. Department of Health and
Human Services, 1984). SS has higher concentrations of some toxic and carcino-
genic substances than MS; however, dilution by room air markedly reduces the
concentrations inhaled by the passive smoker in comparison with those inhaled
by the active smoker. Individual components of tobacco smoke can be measured
in indoor air, but no single component has been shown to accurately measure
the disease-causing potential of ETS.
Klus and Kuhn (1982) actually measured the mainstream and sidestream emis-
sions of various compounds in cigarette smoke. Some of their results are given
in Table 2-25. Note that ammonia appears in the sidestream at a ratio of
10-24:1 to that in the mainstream. This may explain the somewhat more basic
character of SS to MS.
2-72
-------
TABLE 2-25. SOME SIDESTREAM TO MAINSTREAM (SS/MS) RATIOS DETERMINED FOR
VARIOUS ELEMENTS AND COMPOUNDS IN CIGARETTE SMOKE
Substance
SS/MS
Nicotine
Carbon monoxide
Carbon dioxide
Propanal
Formaldehyde
Acrolein
Hydrogen cyanide
Acrolein
Hydrogen cyanide
Acetonitrile
Ammonia
Aniline
Pyridine
N-nitrosodimethyl amine
N'-nitrosonornicotine
N'-ni trosoanatabi ne
4-(N'-methyl-N1-nitrosoamino)-l(3-pyridyl)-l-butanone
Toluol
Benzo(a)pyrene
Quinoline
Cadmi urn
Nickel
Zinc
2.6 to 3.3
2.5 to 4.7
8.9 to 11.3
approximately 2.5
approximately 51
approximately 12
0.056 to 0.37
3.41 to 5.30
73 to 170
29.7
10 to 24
19.1 to 100
10 to 24
19.1 to 100
0.48 to 2.7
0.31 to 0.95
1.27 to 3.7
5.6 to 83
2.1
10.8
3.58 to 7.2
0.19 to 31.0
0.2 to 6.66
2-73
-------
Repace (1984) and co-workers have conducted extensive field studies and
measured the CO and particulate phase emissions in rooms where cigarette
smokers were present. One of their conclusions was that there is good
evidence for treating the emissions of all cigarettes alike. He hypothesized
a tar yield of 32.5 ing/cigarette. Repace and Lowry (1985) estimated that the
"typical" U.S. nonsmoker is exposed to 1.4 mg of tobacco tar per day; the range
for U.S. nonsmokers is given as 0 to 14 mg/day (Samfield, 1985).
White and Froeb (1980) concluded that chronic exposure to tobacco smoke
in the workplace is equivalent to smoking 1 to 10 cigarettes per day. Several
epidemiological studies purport to show that nonsmoking spouses of smokers are
at a high risk than nonsmokers married to nonsmokers (Hirayama, 1981;
Garfinkel, 1981).
2.4.2 Source Characterization
As the sidstream smoke and exhaled mainstream smoke, together with other
gases and particles emitted from burning tobacco, are diluted and circulated in
indoor air, the chemical and physical character of the constituents are thought
to change substantially from either MS or SS, based upon those few constituents
which have been studied in ETS. The median diameter of particles in ETS is
smaller than in undiluted SS (Wynder and Hoffmann, 1967). Nicotine, which is
primarily associated with the particle phase of undiluted SS, is thought to
redistribute between the particle and vapor phase as the SS dilutes, so that it
is found almost exclusively in the vapor phase in ETS (Eudy et al., 1985, 1986;
Hammond et al., 1987). More data are needed on the distribution of ETS compo-
nents between the gas and particle phases since this factor is critical to the
human dosimetry and retention of ETS components in the lung.
Most of the chemical characterization data available which is pertinent
to ETS comes from studies that demonstrated that 300 to 400 of the more than
3800 constituents in MS have also been measured in SS (Wynder and Hoffman,
1967). Certain constituents are found in much higher concentrations in SS as
compared to MS, including the carcinogenic volatile nitrosamines and the vola-
tile pyridines. Since SS also has high concentrations of nitrogen dioxide,
which is a nitrosating agent, chemical transformations are expected to occur
between the point at which chemicals are emitted as SS and the point at which
nonsmokers are exposed to ETS. Another change which occurs as SS is diluted
to form ETS is that in the presence of radon, the short-lived radon daughters
2-74
-------
will attach to ETS particles and become more available for inhalation than in
the absence of ETS. Bergman et al. (1986) have shown that indoor radon-
daughter concentrations can more than double in the presence of tobacco smoke.
It is critical, therefore, that research be conducted to determine the presence,
concentration and phase distributions of toxic and carcinogenic constituents
of ETS in indoor air spaces. Interactions with non-smoke toxic compounds need
to be clarified. Special consideration needs to be given to the possibility
of secondary reactions of amines in indoor air containing ETS that could result
in the production of carcinogenic N-nitrosation and condensation products.
A critical link between source characterization and exposure assessment is
source emission factors. Studies have been conducted to determine particle,
tar, nicotine, and CO emission factors for SS. Rickert et al. (1984) showed
that these SS emission factors were independent of MS emissions. These and
other studies suggest that the SS emission factors for these components are
similar across all brands of cigarettes. Unfortunately, most of these studies
have been conducted using SS collection devices rather than chamber studies
where ETS collection is performed. Studies are needed to determine the emis-
sion factors and surface removal rates for critical constituents of ETS.
The emission rates for carcinogenic and toxic constituents especially need to
be determined.
2.4.3 Exposure Assessment
Three approaches can be used to assess human exposure to ETS:
»
1. Exposure modeling using emission rates, dispersion models, and
time-activity patterns
2. Monitoring exposure concentrations by using either personal
monitoring or microenvironmental measurements combined with
time-activity patterns
3. Human exposure monitoring using biological markers of exposure
All three approaches are being used to address the important question of
human exposure assessment to ETS. Unfortunately, the most imprecise, yet
simplest method, that of employing a simple questionnaire (e.g., Do you live
with, or work with, or have regular contact with persons who are smokers?), has
been the most widely used method of estimating human exposure to ETS. This
2-75
-------
method has been used to classify individuals into broad categories of exposure
while recognizing the problems associated with misclassification due to errors
in reporting current smoking habits, neglecting certain exposures, and the
reporting of exsmokers as nonsmokers.
2.4.3.1 Exposure Modeling. A model for exposure of nonsmoking adults in the
U.S. to ETS particles has been recently developed by Repace and Lowrey (1985)
and is expanded upon by Repace (in press). This model is dependent upon parti-
cle emission factors, time-dependent growth and decay of ETS concentrations in
indoor air spaces, and ventilation and mixing of ETS in room air, to develop a
general equilibrium model and equation for ETS. This model predicts that the
exposure of U.S. nonsmokers ranges from 0 to 14 mg of cigarette tar per day,
depending upon the nonsmokers's lifestyle. The average population exposure
for adults of working age, averaging over the work and home microenvironments,
is about 1.43 mg/day (Repace and Lowrey, 1983) with an 86 percent exposure
probability.
Exposure modeling methods for ETS have recently been reviewed by the
National Research Council (1986a). The modeling efforts to date have concen-
trated on modeling the respirable particles (RSP) (particles <2.5 microns)
emitted from ETS in indoor environments. These models generally use a simpli-
fied form of the mass-balance equation. The models are generally single-
compartment models that assume steady-state or equilibrium conditions to
estimate RSP concentrations. The input parameters are the RSP-emission rate
for ETS, number of cigarettes consumed, ventilation or infiltration rates,
removal rates by surfaces, air mixfng in the environment, and volume of the
space. The variability of one or more of the input parameters can make a
difference of as much as an order of magnitude in the estimated RSP concentra-
tion (National Research Council, 1986a). To the extent that the equilibrium
assumptions do not hold, for example if the source is intermittent rather than
continuous, then additional variability in the exposure assessment is
introduced.
The limited field tests of these general equilibrium models have predicted
RSP concentrations reasonably well over a wide range of input values for the
equations (National Research Council, 1986a). Using this approach, the pre-
dicted RSP exposures due to ETS are highly variable, but always consistently
predict substantial increases in human exposure when smoking occurs in the
indoor space being modeled. The National Research Council (1986a) recommends
2-76
-------
that when field measurement studies of ETS constituents are conducted, data on
model input parameters (e.g., smoking rates, room volume, air exchange rate,
etc.) should be collected so that field evaluations of the equilibrium model
can be made. More information is also needed on current and past distributions
of the input parameters for the mass-balance models of RSP concentrations for a
range of microenvironments in which individuals spend most of their time.
2.4.3.2 Monitoring Exposure. Measurement and monitoring of ETS are hampered
by the fact ETS is a very complex mixture of organic and inorganic gases and
particles. All of the thousands of ETS constituents cannot be monitored on a
routine basis, therefore a marker or tracer for quantifying ETS concentrations
indoors and personal exposure is needed. The National Research Council (1986a)
reviews the tracers that have been used for ETS and suggest that a suitable
tracer should:
be unique or nearly unique to tobacco smoke;
be present in sufficient quantities to be detected at low
smoking rates;
have similar emission rates for a variety of tobacco products;
and
have a fairly consistent ratio to the constituents of interest
(e.g., the carcinogenic constituents) over a range of environ-
mental conditions and for a variety of tobacco products.
While several constituents have been used as tracers of ETS, no single
measure has met all the criteria listed by the NAS, nor has any one of these
measures been recognized as representing ETS exposure. The constituents
reviewed by the National Research Council (1986a) which have been examined as
tracers include: CO, RSP, nicotine, aromatic hydrocarbons, tobacco-specific
nitrosamines, nitrogen oxides, acrolein, acetone, and polonium-210. The NAS
report repeatedly recommends that research efforts need to be directed toward
identifying a tracer contaminant for ETS that meets the criteria listed above.
At present, RSP is the most widely used as a general measure of ETS exposure
indoors. Smoking, however, is not the only source of RSP in indoor air.
Nicotine has many of the characteristics necessary to serve as a tracer for
ETS. The major problems with nicotine as a tracer are: (1) the ratio of
nicotine to other ETS constituents for a variety of tobacco products is not
2-77
-------
known, (2) the reactivity rate of nicotine relative to other ETS constituents
is not known, (3) particulate- or vapor-phase nicotine may be re-emitted once
deposited on surfaces, and (4) until very recently, efficient sampling methods
for total nicotine were not available.
The most extensive indoor air monitoring data base on a constituent
associated with ETS is that for RSP. In addition to microenvironmental mea-
surements of indoor air spaces, personal monitoring methods are available for
RSP. Several studies have recently utilized personal monitors of either RSP or
nicotine to determine total human exposure to ETS (Muramatsu et al., 1984;
Schenker et al., 1984; Sexton et al., 1984; Spengler et al., 1985, Hammond et
al., in press). Spengler et al. (1981) has conducted an extensive study of
exposure to RSP indoors and outdoors, and more recently (Spengler et al., 1985)
using personal monitors. These studies demonstrate the dominant role of ETS in
increasing RSP levels indoors and in personal exposures to RSP.
2.4.3.3 Biological Markers. Exposure to ETS depends upon many factors. The
National Research Council (1986a) concluded that an optimal assessment of
exposure should be done by analysis of the physiological fluids of exposed
individuals rather than just relying upon the indoor air assessment. The
methods which have been used to date are reviewed by the National Research
Council (1986a) and the Surgeon General (1986). These methods include measure-
ments in physiological fluids of: thiocyanate, carboxyhemoglobin, nicotine and
its metabolite cotinine, hydroxyproline, N-nitrosoproline, aromatic amines, and
urinary mutagens.
Nicotine and cotinine are currently the best available markers of human
exposure to ETS. Both nicotine and cotinine in biological fluids are highly
specific to tobacco smoke exposure. Cotinine measurements offer several
advantages over nicotine, and particularly because of its long half-life in
adult nonsmokers, it has become the current method of choice for assessing
human exposure to ETS. In order to use this biological marker more effective-
ly, improved and standardized methods for measuring cotinine are needed, and
studies quantifying the relationship between exposure, dose, and clearance
rate for nonsmokers of diverse age, sex, and race are needed. Current studies
being conducted by EPA suggest that the clearance rate for cotinine in infants
is much faster than in adults.
Nicotine in ETS appears to be primarily in the vapor phase and may not be
directly related to the carcinogenic potential of ETS. The National Research
2-78
-------
Council (1986a) recommends that indicators that are related to the carcinogenic
risk are needed. Sensitive dosimetry methods for tobacco-specific compounds
are urgently needed. They suggest that new methods for immunoassays and
postlabelling be used in the development of dosimetry studies in nonsmokers
exposed to ETS. Protein and DNA adducts are recommended as possible exposure
and dosimetry measures that could be effectively used in both exposure and
epidemiologic studies.
2.4.4 Health Effects
2.4.4.1 Introduction. The health effects of environmental tobacco smoke (ETS)
or passive smoking have recently been thoroughly reviewed by the World Health
Organization (International Agency for Research on Cancer, 1986), National
Research Council of the National Academy of Sciences (National Research
Council, 1986a) and the U.S. Department of Health and Human Services (Surgeon
General of the U.S., 1986). Effects can be divided into acute irritating and
immune effects, respiratory effects, cancer, and cardiovascular effects, among
others.
2.4.4.2 Acute Irritating and Immune Effects. The most common acute effects
associated with exposure to ETS are eye, nose and throat irritation, and objec-
tions to the smell of tobacco smoke. Particle filtration causes little decline
in irritation or odor, which suggests that these effects may be caused by gas-
phase rather than particle-phase constituents of ETS (National Research
Council, 1986a).
Eye-blink rate as a measure of irritation correlates positively with ETS
concentration and is likewise correlated with sensory irritation such as
burning eyes or nasal mucosa (Weber-Tschopp et al., 1976; Weber, 1984).
Cigarette smoke contains immunogens, which are substances that activate
the immune system. Approximately one-half of allergy-prone individuals react
to tobacco extracts or smoke extracts in skin tests, but the specific immuno-
genic compounds have not been isolated from these extracts. Self-reported
complaints of sensitivity to smoke do not correlate well with skin-sensitivity
tests (Lehrer et al., 1984).
2.4.4.3 Respiratory Effects. The evidence for health effects from involuntary
exposure to tobacco smoke is strong for children. Such exposure is currently
widespread, and because of its high prevalence even small effects on morbidity
and mortality have important public health implications.
2-79
-------
Epidemiological investigations have linked passive smoking in children to
increased occurrence of lower respiratory tract illness (presumably of infec-
tious etiology) during infancy (Table 2-26), to increased prevalence of respi-
ratory symptoms, and to reduction of lung function. These studies indicate a
significantly increased frequency of bronchitis and pneumonia during the first
year of life in children with smoking parents. Although outcome measures
varied somewhat among the studies, the relative risks associated with passive
smoking were similar, and dose-response relationships between infant illness
and extent of parental smoking were demonstrable. An effect of passive smoking
was not readily identified after the first year of life. Retrospective
collection of information from the parents of older children confirms excessive
lower respiratory illness experienced by infants whose parents smoke (Schenker
et al., 1983; Ware et a!., 1984).
For respiratory symptoms in children, the effect of passive smoking is not
as large as for lower respiratory illness (Table 2-27). However, data from
numerous cross-sectional studies demonstrate a greater frequency of the most
common respiratory symptoms—cough, phlegm, and wheeze--in the children of
smokers (U.S. Department of Health and Human Services, 1984). In these studies
the subjects have generally been schoolchildren. Recent results from several
large studies, the Harvard Air Pollution Health Study (Ware et al., 1984) and
an English study reported by Charlton (1984) provide convincing evidence that
passive smoking increases the occurrence of cough and phlegm in the children of
smokers, although earlier data from smaller studies had been ambiguous (Table
2-26). The evidence also indicates an excess of chronic wheeze associated with
longitudinal investigations have provided the previously missing evidence of
the consequences of passive smoking during lung growth and development. Tager
et al. (1979) reported that the FEFpry,- declined with the number of smoking
parents in the household, based on cross-sectional data from children in East
Boston, Massachusetts.
In 1983, these investigators provided follow-up results for these children
over a seven-year period (Tager et al., 1983). Using a multivariate statis-
tical analysis, they showed that both maternal smoking and active smoking by
the child reduced the growth rate of the FEV-.. Longitudinal data from the
passive smoking (Table 2-26), although passive smoking has not been signifi-
cantly associated with wheezing in all studies (Schenker et al., 1983; Leeder
et al., 1976; Tashkin, 1984). Evidence for association of passive smoking with
2-80
-------
TABLE 2-26. PREVALENCE OF RESPIRATORY SYMPTOMS IN SELECTED INVESTIGATIONS
OF CHILDREN, BY NUMBER OF SMOKING PARENTS*
Subjects
Respiratory
System
Prevalence (per 100) by
Number of Smoking Parents
0 12
2426 children, aged
6 to 14, England
(Col ley, 1974).
3105 nonsmoking chil-
dren, aged 12 to 13,
England (Bland et al. ,
1978).
650 children, aged
5 to 9, Massachusetts
(Weiss et al., 1980).
Chronic cough 15.6
Cough during the day 16.4
or at night
Chronic cough and 1.7
phlegm
17.2
19.0
2.7
22.2
23.5
3.4
4071 children, aged
5 to 14, Pennsylvania
(Schenker et al. ,
1983).
8528 children, aged
5 to 9, six U.S.
cities (Ware et al. ,
1984).
1733 nonsmoking chil-
dren aged 8 to 10,
England (Charlton,
1984)
Persistent wheeze
Chronic cough
Chronic cough
Persistent wheeze
Chronic cough
Persistent wheeze
Frequent cough
(boys)
Frequent cough
(girls)
1.8
6.3
4.1
7.2
7.7
9.9
35
32
6.8
7.0
4.8
7.7
8.4
11.0
42
40
11.8
8.3
4.0
5.4
10.6
13.1
48
52
^Abstracted from Table 4,
(1984) and from Charlton
U.S. Department of Health and Human Services
(1984).
the development of childhood asthma is conflicting (Gortmaker et al., 1982;
Burchfiel, 1984; Leeder et al., 1976; Horwood et al., 1985; Schenker et al.,
1983; Tashkin et al., 1984).
Substantial data are now available on the effects of parental smoking on
children's lung function. The Surgeon General's 1984 report on smoking and
health (U.S. Department of Health and Human Services, 1984), based on cross-
sectional observations, concluded that passive smoking had a small but
measurable effect on the lung function of children. However, the long-term
consequences of these changes were considered to be unknown. More recent data
2-81
-------
TABLE 2-27. EPIDEMIOLOGICAL STUDIES OF EARLY CHILDHOOD RESPIRATORY ILLNESSES AND PASSIVE SMOKING
Study Population
Study Design
Effect of
Passive Smoking
Comments
10,672 births in
Israel, 1967-1968
(Harlap, 1974)
2205 births in
England, 1963-1965
(Leeder et al., 1976)
12,068 births in
Finland, 1966
(Rantakallio, 1978)
Antenatal maternal
smoking history, moni-
toring of admissions
during first year of
life
Prospective cohort with
annual questionnaire
00
PO
Prospective cohort with
follow-up of hospital-
izations, physician
visits, and mortality
Significant increase
in hospitalization for
pneumonia and bron-
chitis, RR* =1.4
Significant increase
in bronchitis or pneu-
monia in first year of
life, RR = 1.7 if one
parent smoked, RR = 2.6
if both smoked
Significant increase of
hospitalization for
respiratory diseases
during first 5 years,
RR = 1.7
Dose-response relationship
present. Maternal smoking
only
Sex of smoking parent not
examined
Effect largest during first
year. Maternal smoking only
and measured during pregnancy
1265 births in
New Zealand, 1977
(Fergusson et al.,
1985)
130 children with
respiratory virus
infection in infancy,
England (Pullan and
Hey, 1982)
1058 births in
China, 1981 (Chen
et al., 1986)
Prospective cohort with
diaries, physician and
hospital record review
Case-control with 111
controls, performed 10
years after index
illness
Prospective cohort with
illnesses ascertained
at age 18 months by a
questionnaire
Significant increase in
bronchitis or pneumonia
during the first year
of life, RR = 2.0, if
mother smoked
Significant effect of
maternal smoking at
time of illness, RR =
3.2
Significant effect of
smoke exposure at
home, RR = 1.9 if
>10 cigarettes per
day consumed by family
No effect of paternal smoking.
Effect of maternal smoking
equivocal in second year, .
absent in third. Dose-
response present in first
Effect of paternal smoking
also present
None of the mothers smoked
*RR = relative risk. Calculated from published data if not provided by the authors.
-------
from three longitudinal investigations have provided the previously missing
evidence of the consequences of passive smoking during lung growth and develop-
ment. Tager et al. (1979) reported that the FEF25-75 declined with the number
of smoking parents in the household, based on cross-sectional data from
children in East Boston, Massachusetts.
In 1983, these investigators provided follow-up results for these children
over a seven-year period (Tager et al., 1983). Using a multivariate statis-
tical analysis, they showed that both maternal smoking and active smoking by
the child reduced the growth rate of the FEV-.. Longitudinal data from the
Harvard Air Pollution Health Study (Ware et al., 1984) also showed reduced
growth of the FEV-, in children whose mothers smoked cigarettes (Berkey et al.,
1986). In children aged six to ten years, a statistical model estimated that
FEV, growth rate is reduced by 0.17 percent per pack of cigarettes smoked
daily by the mother. Burchfiel (1984) examined the effects of parental smoking
on 15-year lung function change of subjects in the Tecumseh study, first exa-
mined at ages 10 through 19 years. Among subjects who remained nonsmokers
across the follow-up period, parental smoking reduced the growth of the FEV-,,
the FVC, and the Vinax™ in males but had no effect in females.
The health effects of passive smoking on adults have not been as compre-
hensively examined and therefore remain controversial. Only a few cross-
sectional investigations provide information on the association between
respiratory symptoms in nonsmokers and passive smoking, and their results do
not provide consistent evidence of an effect of passive smoking (Lebowitz,
1976; Schilling et al., 1977; Comstock et al., 1981; Schenker et al., 1982).
With regard to pulmonary function in adults, exposure to passive smoking
has been associated with reduction of the FEF25_75 in two cross-sectional
investigations. White and Froeb (1980) compared spirometric test results in
middle-aged nonsmokers with at least 20 years of passive smoking exposure in
the workplace to the results in a control group who did not have similar expo-
sure. The mean FEF25_7r of the exP°sed group was significantly reduced, by 15
percent of the predicted value in women, and 13 percent in men. A recent
population-based French investigation examined the effect of marriage to a
smoker in 849 male and 826 female nonsmokers (Kauffmann et al., 1983). Above
age 40 years, the FEF?r_7c was reduced in nonsmoking men and women with a
smoking spouse. The results of an investigation of 163 nonsmoking women in the
Netherlands also suggested adverse effects of tobacco smoke exposure in the
home (Remijn et al., 1985; Brunekreef et al., 1985).
2-83
-------
Other studies have not indicated chronic effects of passive tobacco smoke
exposure on adult nonsmokers. In two cross-sectional studies, marriage to a
smoker was not significantly associated with reduction of ventilatory function
(Schilling et al., 1977; Comstock et al., 1981). A case-control study of 20-
to 39-year-old nonsmoking women in the Tecumseh Community Health Study cohort
also did not show an effect of marriage to a smoker on lung function level
(Burchfiel, 1984). Kentner et al. (1984) examined the effects of passive and
active smoking in 1351 German white collar workers. Self-reported exposure to
ETS at home and at work was not associated with reduction of lung function, as
assessed by spirometry.
Neither epidemiological nor experimental studies have established the
importance of ETS in exacerbating asthma in adults. Of three studies involving
experimental exposure of adult asthmatics to tobacco smoke (Shephard et al.,
1979; Dahms, 1981; Wiedemann et al., 1986), only one showed a definite adverse
effect (Dahms, 1981). In a population sample in Tucson, AZ, Lebowitz et al.
(1984) examined the relationship between passive smoking and daily symptom
occurrence and daily level of peak flow. Peak flow rates in the asthmatic par-
ticipants were not affected by ETS exposure, although effects on symptom status
were reported.
In 1981, reports were published from Japan (Hirayama, 1981) and Greece
(Trichopoulos et al., 1981) that showed increased lung cancer risk in nonsmok-
ing women married to cigarette smokers. Subsequently, involuntary exposure to
tobacco smoke has been examined as a risk factor for lung cancer in studies
conducted throughout the world (Table 2-28). While not all of the studies have
shown significantly increased risk associated with exposure, the weight of the
epidemiological evidence supports a conclusion that involuntary exposure to
tobacco smoke is a risk factor for lung cancer. This conclusion can be further
supported on a biological basis. SS contains many of the same carcinogens pre-
sent in MS; tobacco smoke materials are absorbed during involuntary exposure;
and respiratory carcinogenesis appears to take place without any threshold of
exposure. These biological considerations underlie the recent conclusion by
the International Agency for Research on Cancer of the World Health Organiza-
tion (1986) that "passive smoking gives rise to some risk of cancer."
All of the epidemiological data discussed here provide evidence of adverse
effects in children and adults exposed to ETS. In all of the studies exposure
2-84
-------
TABLE 2-28. COHORT AND CASE-CONTROL STUDIES OF PASSIVE EXPOSURE
TO TOBACCO SMOKE AND LUNG CANCER
Study
Findings
Comment
Prospective cohort
study in Japan of
91,540 nonsmoking
females, 1966-1981
(Hirayama, 1984)
Case-control study
in Greece of 40
nonsmoking female
cases, 149 controls,
1978-1980 (Trich-
opoulos et al., 1981)
Prospective cohort
study in the U.S.
of 176,139 non-
smoking females,
1960-1972
(Garfinkel, 1981)
Case-control study
in Hong-Kong of 84
female cases and
139 controls, 1976-
1977 (Chan et al.,
1979; Chan and Fung,
1982)
Case-control study
in the US with 22
female and 8 male
nonsmoking cases,
133 female and
180 male controls
(Correa et al.,
1983)
Case-control study
in the U.S.A. 25 male
and 53 nonsmoking
female cases with
matched controls,
1971-1980 (Kabat
and Wynder, 1984)
Age-occupation adjusted
SMRs, by husband
smoking:
Nonsmokers - 1.00
Exsmokers - 1.36
1-19/day - 1.45
>20/day - 1.91
Odds ratios by husband
smoking:
Nonsmokers - 1.0
Exsmokers - 1.8
Current smokers
<20/day - 2.4
>21/day - 3.4
Age-adjusted SMRs, by
husband smoking:
Nonsmokers - 1.00
Current smokers
<20/day - 1.27
>20/day - 1.10
Crude odds ratio of 0.75
associated with smoking
spouse.
Odds ratios by spouse
smoking:
Nonsmokers - 1.00
<40 pack years - 1.48
>41 pack years - 3.11
Odds ratio not signifi-
cantly increased for
current exposure at home:
Males - 1.26
Females - 0.92
Trend statistically
significant; All
histologies
Trend statistically
significant; His-
tologies other than
adenocarcinoma and
bronchioloalveolar
carcinoma
All histologies;
Effect of husband
smoking not signifi-
cant
All histologies.
Two reports are in-
consistent on the
exposure variable
Significant increase
for >41 pack years.
Bronchioloalveolar
carcinoma excluded
All histologies.
Findings negative
for spouse smoking
variable as well
(continued on following page)
2-85
-------
TABLE 2-28. (continued)
Study
Findings
Comment
Prospective cohort
study in Scotland
of 8128 males and
females, 1972-1982
(Gillis et al.,
1984)
Case-control study
in Hong Kong with
88 nonsmoking female
cases, 1981-1982
(Koo et al., 1984)
Case-control study
in the U.S. with
31 nonsmoking
and 189 smoking
female cases
(Wu et al., 1985)
Case-control study
in the U.S. with
134 nonsmoking
female cases
(Garfinkel et al.,
1985)
Case-control study
in Japan with 19
male and 94 female
nonsmoking cases,
and 110 male and
270 female non-
smoking controls
(Akiba et al.,
1986)
Case-control study
in Louisiana, Texas,
and New Jersey with
99 nonsmoking cases
and 736 controls
(Dalager et al.,
1986)
Age-adjusted mortality
ratios for domestic
exposure:
Males - 3.25
Females - 1.00
Odds ratio of 1.24
(p >0.40) for combined
home and workplace
exposure. No associ-
ation with cumulative
hours of exposure
No significant effects
of exposure from
parents, spouse, or
workplace in smokers
and nonsmokers
Nonsignificant odds ratio
of 1.22 if husband smoked.
Significantly increased
odds ratio of 2.11 if hus-
band smoked 20 or more
cigarettes daily at home.
Significant trend with
number of cigarettes smoked
at home by the husband
For females, odds ratio of
1.5 if husband smoked; for
males, odds ratio of 1.8
if wife smoked
Adjusted odds ratio for
marriage to a smoking
spouse was 1.5
Preliminary, small
numbers of cases
All histologies
Adenocarcinoma
and squamous cell
carcinoma only
All histologies;
Careful exclusion
of smokers from
the case group
Clinical or radio-
logical diagnosis
for 43%. All
histologies
Nearly 100% histo-
logical confirmation.
All histologies
SMR = standard mortality ratio
2-86
-------
has been assessed qualitatively rather than by measurements of personal expo-
sure or smoke concentrations in rooms. Information of quantitative dose-
response relationships for these effects is lacking.
2.4.4.4 Lung Cancer and Other Cancers. Smoking remains the largest single
preventable cause of death and disability for the U.S. population. Recently
the WHO/IARC issued a comprehensive monograph (International Agency for
Research on Cancer, 1986) on the carcinogenic risk of tobacco smoke to humans,
which concludes that cigarette smoking is a major cause of cancer and is most
strongly associated with cancers of the lung and respiratory tract. Smoking
also causes cancers at other sites, including the pancreas and urinary bladder.
The National Research Council (1986a) and the Surgeon General of the U.S.
(1986) released reports in December of 1986 on ETS which included extensive
reviews of studies examining cancer risk from passive smoking or involuntary
exposure to tobacco smoke. The summary that follows is taken from these two
reports.
Exposure to ETS has been examined as a risk factor for lung cancer in non-
smokers in numerous recent epidemiologic studies. These studies have compared
the risks for subjects exposed to ETS with the risks for people not reported
to be exposed. Because exposure to ETS is an almost universal experience in
the more developed countries, these studies involve comparison of more exposed
and less exposed people, rather than comparison of exposed and unexposed
people. The studies are therefore inherently conservative in assessing the
consequences of exposure to ETS. Interpretation of these studies must consider
the extent to which populations with different ETS exposures have been identi-
fied, the gradient in ETS exposure from the lower exposure to the higher
exposure groups, and the magnitude of the increased lung cancer risk that
results from increased ETS exposure.
To date, questionnaires have been use to classify ETS exposure. Quantifi-
cation of exposure by questionnaire, particularly lifetime exposure, is diffi-
cult and has not been validated. However, spousal and parental smoking status
identify individuals with different levels of exposure to ETS. Therefore,
investigation has focused on the children and nonsmoking spouses of smokers,
groups for whom greater ETS exposure would be expected, and for whom increased
nicotine absorption has been documented compared to the children and nonsmoking
spouses of nonsmokers.
2-87
-------
Most (11 of 13) epidemiologic studies reviewed that have examined the
association between involuntary smoking and lung cancer have shown a positive
association with exposure, and in 6 the association reached statistical
significance (Surgeon General of the U.S., 1986; National Research Council,
1986a).
There were five additional studies that were excluded because of inadequa-
cies either of documentation or of the study itself. The NRC report summarized
the relative risk estimate from each of these studies and determined a summary
estimate based on the combined studies.
This risk was 1.34 (95 percent confidence limits 1.18 to 1.53) overall
(National Research Council, 1986a). For all women the relative risk was 1.32
(1.16 to 1.51); for men it was 1.62 (0.99 to 2.64). The wide confidence limits
for men reflect the fact that most of the data were based on nonsmoking women
rather than nonsmoking men. For studies conducted in the United States, the
relative risk was 1.14 (0.92 to 1.40). Considering only the largest studies
(those with expected number of lung cancer deaths of 20 or more), the relative
risk estimate was 1.32 (1.15 to 1.52) (National Research Council, 1986a). The
confidence limits on each of these estimates all include the overall summary
estimate of 1.34. Given the difficulty in identifying groups with differing
ETS exposures, the low-dose range of exposure examined, and the small numbers
of subjects in some series, it is not surprising that some studies have found
no association, and that in others the association did not reach a conventional
level of statistical significance. The question is not whether cigarette smoke
can cause lung cancer; that question has been answered unequivocally by examin-
ing the evidence for active smoking. The question is, can tobacco smoke at a
lower dose and through a different mode of exposure cause lung cancer in
nonsmokers? The answer must be sought in the coherence and trends of the
epidemiologic evidence available on this low-dose exposure to a known human
carcinogen. In general, those studies with larger population sizes, more care-
fully validated diagnosis of lung cancer, and more careful assessment of ETS
exposure status have shown statistically significant associations. A number of
these studies have demonstrated a dose-response relationship between the level
of ETS exposure and lung cancer risk. By using data on nicotine absorption by
the nonsmoker, the nonsmoker's risk of developing lung cancer observed in human
epidemiologic studies can be compared with the level of risk expected from an
2-88
-------
extrapolation of the dose-response data for the active smoker. This extrapola-
tion yields estimates of an expected lung cancer risk that approximate the
observed lung cancer risk in epidemiologic studies of involuntary smoking.
The weight of evidence derived from epidemiologic studies shows an associ-
ation between ETS exposure of nonsmokers and lung cancer that, taken as a
whole, is unlikely to be due to chance or systematic bias. The observed
estimate of increased risk is 34 percent, largely for spouses of smokers
compared with spouses of nonsmokers. One must consider the alternative
explanations that this excess either reflects bias inherent in most of the
studies or that it represents a causal effect. Misclassification of smokers,
exsmokers and nonsmokers may have contributed to the result to some extent.
Computations of the effect of two sources of misclassification were presented.
Computations taking into account the possible effects of misclassified
exsmokers and the tendency for spouses to have similar smoking habits placed
the best estimate of increased risk of lung cancer at about 25 percent in
persons exposed to ETS, at a level typical of that experienced by nonsmokers
married to smokers, compared with those married to nonsmokers. Another compu-
tation using information from cotinine levels observed in nonsmokers, and
taking into account the effect of making comparisons with a reference popula-
tion that is truly unexposed, leads to an estimated increased risk of about
one-third when exposed spouses were compared with a truly unexposed population.
The data presented in the Surgeon General and NRC reports establish that a
substantial number of the lung cancer deaths that occur among nonsmokers can be
attributed to involuntary smoking. However, better data on the extent and
variability of ETS exposure are needed to estimate the number of deaths with
confidence.
2.4.4.5 Cardiovascular Disease and Other Effects. Tobacco smoking and cardio-
vascular disease are causally associated (U.S. Department of Health and Human
Services, 1984), and the effects of smoking on exercise tolerance and blood
pressure are well documented and reviewed elsewhere (U.S. Department of Health
and Human Services, 1984). The constituents of tobacco smoke thought to be
associated with cardiovascular disease are CO and nicotine.
The effects of ETS on cardiovascular disease have recently been reviewed
by the National Research Council (1986a) and the Surgeon General of the U.S.
(1986). A summary from the Surgeon General's report follows.
2-89
-------
The relationship between cardiovascular disease and ETS has been examined
in one case-control study and three prospective studies. In the case-control
study by Lee and colleagues (1986), ischemic heart disease cases and controls
did not show a statistically significant difference in their exposure to ETS
based on the smoking habits of spouses or on an index accounting for exposure
at home, at work, and during travel and leisure. In a Japanese cohort study,
Hirayama (1984, 1985) reported an elevated risk for ischemic heart disease (N =
494) in nonsmoking women married to smokers. The standardized mortality ratios
when the husbands were nonsmokers, exsmokers or smokers of 19 or more ciga-
rettes per day, and smokers of 20 or more cigarettes per day were 1.0, 1.10,
and 1.30, respectively (one-sided p for trend, 0.019).
In a Scottish study by Gill is et al. (1984), nonsmokers not exposed to
tobacco smoke were compared with nonsmokers exposed to tobacco smoke, with
respect to the prevalence of cardiovascular symptoms at entry and mortality due
to coronary heart disease. There was no consistent pattern of differences in
coronary heart disease or symptoms between nonsmoking men exposed to tobacco
smoke and their nonexposed counterparts. Nonsmoking women exposed to tobacco
smoke exhibited a higher prevalence of angina and major electrocardiogram (ECG)
abnormality at entry, and also a higher mortality rate for all coronary di-
seases. However, rates of myocardial infarction mortality were higher for
exposed nonsmoking men and women compared with the nonexposed nonsmokers. The
rates were 31 and 4 per 10,000, respectively, for the nonexposed nonsmoking
men and women, and 45 and 12 per 10,000, respectively, for the exposed non-
smoking men and women. None of the differences were tested for statistical
significance.
In the Japanese and the Scottish studies, other known risk factors for
cardiovascular diseases (i.e., systolic blood pressure, plasma cholesterol)
were not taken into account in the analysis.
In a study of heart disease, Garland et al. (1985) enrolled 82 percent of
adults aged 50 to 79 between 1972 and 1974 in a predominantly white, upper-
middle-class community in San Diego, California. Blood pressure and plasma
cholesterol were measured at entry, and all participants responded to a
standard interview which asked about smoking habits, history of heart disease,
and other health-related variables. Excluding women who had a previous history
of heart disease or stroke or who had ever smoked, 695 currently married
nonsmoking women were classified by their husbands' self-reported smoking
2-90
-------
status at enrollment. After 10 years of follow-up, there were 19 deaths due to
ischemia heart disease; the age-standardized mortality rates for nonsmoking
wives whose husbands were nonsmokers, exsmokers, and current smokers were 1.2,
3.6, and 2.7, respectively (one-sided p for trend, <0.10). After adjustment
for age, systolic blood pressure, total plasma cholesterol, obesity index, and
years of marriage, the relative risk for death due to ischemic heart disease
for women married to current or former smokers at entry compared with women
married to never smokers was 2.7 (one-sided p <0.10).
The study's findings are not convincing from the point of view of sample
stability. The total number of deaths due to ischemic heart disease was small,
and the denominator in the relative risk calculation is unstable, based on the
deaths of two women whose husbands had never smoked. Moreover, it is well
established that the risk of coronary heart disease is substantially lower
among those who have stopped smoking (U.S. Department of Health and Human
Services, 1984), although the amount of time required for this change after
cessation of smoking is not clear (Kannel, 1981). In this study, 15 of 19
deaths occurred in nonsmoking women married to husbands who had stopped smoking
at entry, and the age-standardized rate for ischemic heart disease was highest
in this group. The high proportion of deaths in nonsmoking women married to
men who stopped smoking implies that the excess resulted from sustained effects
of involuntary smoking. More detailed characterizations of exposure to ETS and
specific types of cardiovascular disease associated with this exposure are
needed before an effect of involuntary smoking on the etiology of cardio-
vascular disease can be established.
One study (Aronow, 1978) suggested that involuntary smoking aggravates
angina pectoris. This study was criticized because the end point, angina, was
based on subjective evaluation, and because other factors such as stress were
not taken into account (Coodley, 1978; Robinson, 1978; Waite, 1978; Wakehan,
1978). More importantly, the validity of Aronow's work has been questioned
(Budiansky, 1983).
Both the National Research Council (1986a) and the Surgeon General of the
U.S. (1986) concluded that although several studies show excessive risk of
cardiovascular disease in ETS-exposed nonsmokers, methodological problems in
study design and analysis preclude any firm conclusions on the association
between ETS exposure and cardiovascular disease.
2-91
-------
Other health effects of ETS which have been examined are primarily related
to studies of growth and health of children exposed to ETS. National Research
Council (1986a) reviewed these studies, particularly those associated with the
influence of ETS exposure on the birthweight of newborn babies born to non-
smoking women exposed to ETS while pregnant, and potential ETS effects on
childhood growth, and its role in the excess development of ear infections.
All of these studies were unable to differentiate effects of i_n utero exposure,
from childhood exposures to ETS.
2.4.5 IAQ Control Options
The obvious methods for the control of exposure of nonsmokers to the
sidestream and exhaled mainstream smoke of smokers are:
1. The prohibition of smoking in public places.
2. The elimination of smoking in the workplace where nonsmokers
are present or the complete segregation of smokers.
3. The educating of the nonsmoking public (particularly the
nonsmoking spouses of smokers) to the hazards of passive smoke.
A public awareness campaign on the part of federal, state, and city
governments has had a significant effect in reducing exposure of nonsmokers to
SS of smokers, but complete elimination has not been achieved.
Fortunately, other control options are also available:
1. Properly designed air ventilation systems can be effective in
reducing the concentration of ETS in the indoor environment,
especially if air from smoking areas is exahusted from the
building. This however, carries an energy penalty. This may
be partly offset through the use of air-to-air heat exchangers.
2. The use of high-efficiency filters on electrostatic devices,
either installed in the ductwork of central air handling systems
or as portable units, strategically placed, can be effective in
removing the PM of ETS. Little is known about the effect of
such devices on the gas phase of smoke. Questions have also
been raised concerning ozone production by these electrostatic
devices.
3. Negative ion generators, when properly built and installed, are
said to provide an effective means of removal of smoke con-
stituents. More research is needed to determine removal
effectiveness and design parameters.
2-92
-------
2.4.6 Conclusions
2.4.6.1 What is Known. Evidence has accumulated indicating that nonsmoking
pregnant women exposed to ETS on a daily basis for several hours are at in-
creased risk for producing babies of low birthweight. The mechanisms that
reduce birthweight are, as yet, unknown. Recent studies show a dose-response
relationship between the number of cigarettes smoked by the father and birth-
weight of the children of nonsmoking pregnant women.
A few studies have reported that children of smokers have reduced growth
and development. These require further corroboration to differentiate ui
utero exposure from subsequent childhood exposures.
Household exposure to ETS is linked with increased rates of chronic ear
infections and middle-ear effusions in young children. For children with nasal
allergies and recurrent otitis media, ETS exposures may synergistically in-
crease their risk for persistent middle-ear effusions.
2.4.6.2 What Scientific Information is Missing. Experimental studies should
be developed to articulate possible mechanisms through which paternal smoking
adversely effects fetal growth in nonsmoking pregnant women. Special emphasis
should be placed on identifying relevant effects of pregnancy on excretion and
absorption of ETS, including transplacental metabolism.
Additional study is needed to corroborate one finding of a dose-response
relationship between reduced height of children and increasing numbers of
cigarettes smoked in the home, regardless of whether the mother smoked during
pregnancy and regardless of which parent smoked.
Research should be conducted to explore the mechanisms by which exposure
to ETS might adversely affect the functioning of the ear and to study possible
long-term consequences of ETS exposure for the auditory apparatus.
2.4.6.3 Research Needs. The National Research Council report (1986a) on ETS
also provides an extensive review of the data available and research needed on
the assessment of human exposure and dosimetry of ETS.
The National Research Council and the Surgeon General concluded that
laboratory studies can contribute to a better understanding of the factors and
mechanisms involved in the induction of cancer and the cancer potency of ETS.
There have been numerous bioassays conducted on MS. In examining the effects
of MS, many research workers used condensates of the smoke painted on the
shaved skin of mice. This contrasts with human exposure, which is mainly to
the respiratory tract. Nonetheless, these skin-painting studies have been
2-93
-------
useful in examining the carcinogenicity of different tobacco constituents and
have advanced knowledge of the actions of MS on a gross exposure level. Except
for one study which provides suggestive evidence that SS is more tumorigenie
than MS, similar work with skin painting has not been done with ETS and would
be of value for assessing the differential toxicity and carcinogenicity of ETS
and MS.
In contrast to MS exposure, ETS exposure involves proportionately more
exposure to gas phase than to particulate phase constituents. There have been
no studies of the effects of exposure to aged ETS. The relative ijn vivo
toxicities of MS, SS, and ETS need to be assessed.
Some studies have attempted to evaluate the gas phase of MS, SS, and ETS
in short-term, |n vitro assays. A solution of the gas phase of MS has been
shown to induce dose-dependent increases in sister-chromatid exchanges in
cultured human lymphocytes. Mutagenic activity has been found in the PM of SS
and in condensates of ETS. However, the work done to date is too sparse to
permit any estimates of the mutagenicity of ETS per se, even though most of
ETS consists of SS. Further ui vitro assays of ETS are needed.
There have been few studies of risk for cancers other than lung in non-
smokers exposed to ETS. Some of the sites considered have been the brain, the
hematopoietic system, and all sites combined. The results of these studies
have been inconsistent. Whether there is an association between ETS exposure
and cancers of any site other than the lung is an important topic for future
epidemiologic inquiries.
2.5 NONCOMBUSTION PARTICLES
2.5.1 Asbestos
Asbestos is a generic term that applies to a group of impure hydrated
silicate minerals which occur in various fibrous forms (Hawley, 1981). They
are incombustible and separable into filaments. Types of asbestos include
amphiboles, such as amosite, crocidolite, tremolite, anthophyllite, and actino-
lite, and chrysotile. Chrysotile, a fibrous form of serpentine, is a magnesium
silicate whose fibers are strong and flexible, and its longer fibers can be
spun into thread for weaving (Hawley, 1981; Sittig, 1985). It is the most
widely used form of asbestos in the United States. Amphibole asbestos includes
various silicates of magnesium, iron, calcium, and sodium. Its fibers are
2-94
-------
generally brittle and cannot be spun, but it is more resistant to heat than
chrysotile asbestos (Hawley, 1981).
Asbestos has been used in fireproof fabrics, brake linings, gaskets,
roofing compositions, electrical and heat insulations, paint fillers, and
chemical filters; as a reinforcing agent in rubber and plastics; and as a
component of paper dryer felts and diaphragm cells (Hawley, 1981). Most
asbestos has been used in the construction industry. Most (92 percent) of the
one-half million tons used in the U.S. is firmly bonded in such products as
floor tiles, asbestos cements, roofing felts, and shingles. The rest is fri-
able or in powder form, present in insulation materials, asbestos cement pow-
ders, and acoustical products (Sittig, 1985). Inhalation of free fibers can
result in a fibrosis of the lung known as asbestosis, and in the development
of mesotheliomas, and cancers of the lung and gastrointestinal tract.
2.5.1.1 Sources. While no longer installed in homes or commercial buildings,
asbestos was widely used as a constituent of ceiling tiles and for pipe, duct,
and attic insulation. The typical size of asbestos fibers is 0.1 to 10 urn in
length, and when disturbed, asbestos fibers may become suspended in the air for
many hours, thus increasing the extent of asbestos exposure for individuals
within the area (U.S. Environmental Protection Agency, 1985c). Fallout from
asbestos installations may occur without overt physical disruption of the
fiber-bearing material and may simply be a function of adhesive degradation,
vibration, humidity variations, air movement from heating and ventilating
equipment, and air turbulence and vibration caused by human activity (U.S.
Environmental Protection Agency, 1978). Usually, asbestos fibers released in
buildings have been associated with visible damage or erosion of asbestos
materials such as insulation, asbestos cement piping and insulation, floor
tiling, shingles, or other asbestos-containing material. Many buildings with
such materials intact have no increased concentrations of asbestos in air.
During 1974, 116 samples of indoor and outdoor air were collected in 19
buildings (usually 4 to 6 indoor samples and 1 ambient air control sample per
building) in 5 U.S. cities to assess whether contamination of the building
air resulted from the presence of asbestos-containing surfacing materials in
rooms or return air plenums (Nicholson et al., 1975). The asbestos materials
in the buildings were of two main types: 1) a cementitious or pi aster-like
material that had been sprayed as a slurry onto steelwork or building surfaces,
and 2) a loosely bonded fibrous mat that had been applied by blowing a dry
2-95
-------
mixture of fibers and binders through a water spray onto a target surface.
The friability of the two types of materials differed considerably; the
cementitious spray surfaces were relatively impervious to damage while the
fibrous sprays were highly friable. The results of air sampling in these
buildings provide evidence that the air of buildings with fibrous asbestos-
containing materials may often be contaminated.
Weathering of asbestos cement wall and roofing materials was shown to be a
source of asbestos air pollution by analyzing air samples taken in buildings
constructed of such material (Nicholson, 1978). Seven samples taken in a
school after a heavy rainfall showed asbestos concentrations from 20 to 4500
33 3
ng/m (arithmetic mean = 780 ng/m ); all but two samples exceeded 100 ng/m .
The source was attributed to asbestos washed from asbestos cement walkways and
asbestos cement roof panels. No significantly elevated concentrations were
observed in a concurrent study of houses constructed of asbestos cement
materials. Roof water runoff from the homes landed on the ground and was not
reentrained, while that of the schools fell to a smooth walkway, which allowed
easy reentrainment when dry. Contamination from asbestos cement siding has
also been documented by Spurny and co-workers (1980).
Asbestos may enter buildings from outside sources. One of the most
significant remaining contributions to environmental asbestos concentrations
may be emissions from braking of automobiles and other vehicles. Measurements
of brake and clutch emissions reveal that, annually, 2.5 tons of unaltered
asbestos are released to the atmosphere, and an additional 68 tons fall to
roadways where some of the asbestos is dispersed by passing traffic (Jacko et
al., 1973). Infiltration of outside air, or entry through building air intakes
may disperse asbestos fibers from these sources indoors.
Asbestos flooring is used in a large number of buildings and is the third
largest use of asbestos fibers. Sebastien et al. (1982) measured concentra-
o
tions of indoor airborne asbestos up to 170 ng/m in a building with weathered
asbestos floor tiles.
2.5.1.2 Monitoring. The analysis of ambient air samples for asbestos has
utilized techniques different from those used in occupational circumstances.
2
This situation occurred because typical urban air may contain up to 100 (jg/m
of particulate matter in which the researcher is attempting to quantify asbes-
3 3
tos concentrations from about 0.1 ng/m to perhaps 1000 ng/m . Thus, asbestos
may constitute only 0.0001 to 1 percent of the particulate matter in a given
2-96
-------
air sample. Asbestos found in ambient air has a size distribution such that
the vast majority of fibers are too short or too thin to be seen with an
optical microscope. In many cases, these fibers and fibrils will be agglomer-
ated with a variety of other materials present in the air samples.
The only effective method of analysis uses electron microscopy to enumer-
ate and size all asbestos fibers (Nicholson and Pundsack, 1973; Samudra et al.,
1978). Samples for such analysis are usually collected either on a Nuclepore®
(polycarbonate filter with a pore size of 0.4 urn or on a Millipore® (cellulose
ester) filter with a pore size of 0.8 urn. In some cases the Millipore® is
backed by nylon mesh. Samples collected on Nuclepore® filters are prepared for
direct analysis by carbon coating the filter to entrap the collected particles.
A segment of the coated filter is then mounted on an electron microscope grid,
which is placed on a filter paper saturated with chloroform so that the chloro-
form vapors dissolve the filter material. (Earlier electron microscopic analy-
sis utilized a rub-out technique in which the ash residue was dispersed in a
nitrocellulose film on a microscope slide and a portion of the film was then
mounted on an electron microscope grid for scanning.)
Samples collected on Millipore® filters are prepared for indirect analysis
by ashing a portion of the filter in a low-temperature oxygen furnace. This
removes the membrane filter material and all organic material collected in the
sample. The residue is recovered in a liquid phase, dispersed by ultrasonifi-
cation, and filtered on a Nuclepore® filter. The refiltered material is coated
with carbon and mounted on a grid as described above. The samples are then
subjected to analysis. Chrysotile asbestos is identified on the basis of its
morphology as seen with the electron microscope, and amphiboles are identified
on the basis of morphology under electron microscopy and by their selected area
electron diffraction patterns, supplemented by energy-dispersive X-ray analysis.
o
Fiber concentrations in fibers per unit of volume (such as fibers/cm , fibers/
3
m , etc.) are calculated based on sample volume and filter area counted. In
some cases, mass concentrations are reported using fiber volume and density
relationships. However, mass concentrations may not be reliable if the sample
contains fibrous forms, such as clusters, bundles, and matrices, in which fiber
volume is difficult to determine. These materials may constitute most of the
asbestos mass in some samples, particularly those reflecting emission sources.
Current fiber counting methods do not include those clumps. However, many of
them are respirable and to the extent that they are broken apart in the lungs
2-97
-------
into individual fibers, they may add to the carcinogenic risk. On the other
hand, methods that break up fibers generally disperse the clumps as well. In
such analyses, the clumps would contribute to the mass.
In much of the earlier analyses of chrysotile concentrations in the United
States, the ashed material was either physically dispersed or disrupted by
ultrasonification. Thus, no information was obtained on the size distribution
of the fibers in the original aerosol. Air concentrations were given only in
terms of total mass of asbestos present in a given air volume, usually nano-
3
grams per cubic meter (ng/m ). With the use of Nuclepore® filters and appro-
priate care in the collection of samples and their processing, information on
the fiber size distribution can be obtained, and concentrations of fibers of
selected dimensions can be calculated. Samples collected on Millipore® filters
can be ashed and the residue resuspended and filtered through Nuclepore® fil-
ters. However, some breakage of fibers for electron microscopic analysis has
been reported by Burdett and Rood (1983) and is being tested by several labora-
tories. However, the utility and reliability of this technique is unknown at
present.
Current measurements of low-level contamination with asbestos use electron
microscope techniques, which determine the total mass of asbestos present in a
given volume of air. Previous measurements of concentrations of fibers longer
than 5 urn were made using optical microscopy, or from optical microscopy of
total particulate samples. Occupational studies used the latter techniques.
If information regarding health effects from these studies is to be extrap-
olated to measurements made in nonindustrial indoor spaces, a relationship
between optical fiber counts and mass of asbestos determined by electron
microscopy must be established. Crude estimates of a conversion factor
relating fiber concentration in fibers per milliliter (f/ml) to airborne
o
asbestos in micrograms per cubic meter (ng/m ) is derived from several studies,
and is detailed in the Airborne Asbestos Health Update (Nicholson, 1986).
These crude conversion factors relating mass concentrations to optical fiber
concentrations range from 5 to 150, and by necessity introduces large
uncertainties. For extrapolation of low-mass concentrations the geometric mean
to fiber count of the above range of conversion factors, which is 30 ug/m /f/ml
is used in the EPA asbestos health assessment (U.S. Environmental Protection
Agency, 1986). The geometric standard deviation of this value is 4, and this
uncertainty severely limits any extrapolation in which it is used. In the case
2-98
-------
of amosite, the data of Davis et al. (1978) suggest that a conversion factor of
18 is appropriate. However these data yield lower chrysotile values than all
other chrysotile estimates; therefore, they may also be low for amosite
(Nicholson, 1986).
2.5.1.3 Exposure. Measurements using electron microscopy techniques estab-
lished the presence of asbestos in the urban ambient air, usually at concentra-
3 33
tions less than 10 ng/m . Concentrations of 100 ng/m to 1000 ng/m were
measured near specific asbestos emission sources, in schools where asbestos-
containing materials are used for sound control, and in office buildings where
o
similar materials are used for fire control. The value of 300 ng/m corre-
sponds to about 10,000 fibers/m , but this is a crude estimate. Excess concen-
trations in buildings have usually been associated with visible damage or
erosion of the asbestos materials. Many buildings with intact material have no
increased concentrations of asbestos. Most ambient measurements were taken
over ten years ago, therefore it is very important to obtain more current data.
Asbestos exposure data pertain essentially to conspicuous episodes; the studies
were not designed to provide representative measures of ambient concentrations
throughout the United States.
Table 2-29 summarizes those studies of the general ambient air or of
specific pollution circumstances that have a sufficient number of samples for
comparative analysis. The data are remarkably consistent. Average 24-hour
samples of general ambient air indicate asbestos concentrations of 1 to 2
o
ng/m (two U.S. samples that may have been affected by specific sources were
not included). Short-term daytime samples are generally higher, reflecting the
possible contributions of traffic, construction, and other human activities.
In buildings having asbestos surfacing materials, average concentrations 100
times greater than ambient air are seen in some schools, and concentrations 5
to 30 times greater than that in ambient air are seen in some other buildings.
Of concern was the discovery of extensive asbestos use in public school
buildings (Nicholson et al., 1978). Asbestos surfaces were found in more than
10 percent of pupil-use areas in New Jersey schools, with two-thirds of the
surfaces showing some evidence of damage. Because these values appear to be
typical of conditions in many other states, it was estimated that 2 to 6
million pupils and 100,000 to 300,000 teachers may be exposed to released
asbestos fibers in schools across the nation. To obtain a measure of
contamination for this use of asbestos, 10 schools were sampled in the urban
2-99
-------
TABLE 2-29. SUMMARY OF ENVIRONMENTAL ASBESTOS SAMPLING STUDIES
Mean
Collection Number Concentration
Sample Set Period of Samples ng/m3
Quarterly composites of 5 to 7 24-hr 1969-70 187 3.3 Ca
U.S. samples (Nicholson, 1971;
Nicholson and Pundsack, 1973)
Quarterly composite of 5 to 7 24-hr 1969-70 127 3.4 C
U.S. samples (U.S. Environmental
Protection Agency, 1974)
5-day samples of Paris, France 1974-75 161 0.96 C
(Sebastien et al., 1980)
6- to 8-hour samples of New York City 1969 22 16 C
(Nicholson et al., 1971)
5-day, 7-hour control samples for U.S. 1980-81 31 6.5 (6C, 0.5Ab)
school study (Constant et al., 1983)
16-hour samples of 5 U.S. cities (U.S. 1974 34 13 C
Environmental Protection Agency, 1974)
New Jersey schools with damaged asbes- 1977 27 217 C
tos surfacing materials in pupil use
areas (Nicholson et al., 1978)
U.S. school rooms/areas with asbestos 1980-81 54 183 (179C, 4A)
surfacing material (Constant, 1983)
U.S. school rooms/areas in building 1980-81 31 61 (53C, 8A)
with asbestos surfacing material
(Constant, 1983)
Buildings with asbestos materials in 1976-77 135 35 (25C, 10A)
Paris, France (Sebastien et al., 1980)
U.S. buildings with friable asbestos in 1974 54 48 C
plenums or as surfacing materials
(Nicholson et al., 1975; Nicholson et
al., 1976)
U.S. buildings with cementitious asbes- 1974 28 15 C
tos material in plenums or as surfacing
materials (Nicholson et al., 1975, 1976)
Ontario buildings with asbestos insula- 1982 63 2.1
tion (Ontario Royal Commission, 1984)
aC = Chrysotile
A = Amphibole
2-100
-------
centers of New York and New Jersey and in suburban areas of Massachusetts and
New Jersey. Schools were selected for sampling because of visible damage, in
some cases extensive.
Samples were taken over 4 to 8 hours in 10 schools (1 to 5 samples per
0
school). Chrysotile asbestos concentrations ranged from 9 ng/m to 1950
o 3
ng/m , with an average of 217 ng/m . Outside air samples at 3 of the schools
33 3
varied from 3 ng/m to 30 ng/m , with an average of 14 ng/m . In all samples
o
but two (which measured 320 ng/m ) no asbestos was visible on the floor of the
sampled area, although surface damage was generally present near the area.
•3
The highest value (1950 ng/m ) was in a sample that followed routine sweeping
of a hallway in a school with water damage to the asbestos surface, although
no visible asbestos was seen on the hallway floor. It is emphasized that the
schools were selected in testing on the basis of the presence of visible
damage. Although the results cannot be considered typical of all schools
having asbestos surfaces, the results do illustrate the extent to which
contamination can exist.
A recent study suggests that the above school samples may not be atypical
(Constant et al., 1983). Concentrations similar to those indicated above were
found in the analysis of samples collected during a 5-day period in 25 schools
that had asbestos surfacing materials. The schools were in a single district
and were selected by a random procedure, not because of the presence or absence
of damaged material. A population-weighted arithmetic mean concentration of
o
179 ng/m was measured in 54 samples collected in rooms or areas that had
3
asbestos surfacing material. In contrast, a concentration of 6 ng/m was
measured in 31 samples of outdoor air taken at the same time. Of special
concern are 31 samples collected in the schools that used asbestos, but taken
in areas where asbestos was not used. These data showed an average concentra-
tion of 53 ng/m , indicating dispersal of asbestos from the source. As pub-
lished fiber counts were fibers of all sizes, only the fiber mass data are
listed in the table. Additionally, fiber clumps were noted in many samples,
but were not included in the tabulated masses.
2.5.1.4 Health Effects. The most serious health effects from asbestos expo-
sure are lung cancer and mesothelioma (cancer of tissue of mesothelial origin),
and these have been established as the most important causes of death from
asbestos exposure. While the relationship of occupational exposure to asbestos
with these cancers has been established from occupational/epidemiologic studies,
2-101
-------
the occurrence of asbestosis has been documented more extensively than the risk
of the asbestos-related malignancies. In part, this documentation resulted
from knowledge of this disease extending back to the turn of the century,
whereas the malignant potential of asbestos was not suggested until 1935 (Lynch
and Smith, 1935; Gloyne, 1936) and not widely appreciated until the 1940s
(Merewether, 1949). Asbestosis had been documented in a wide variety of work
circumstances and associated with all commercial types of asbestos fibers.
Among some heavily exposed groups, 50 to 80 percent of individuals employed for
20 or more years in asbestos-associated industries were found to have abnormal
X-ray films characteristic of asbestos exposure (Selikoff et al., 1965;
Lewinsohn, 1972). A lower percentage of abnormal X-ray films was present in
lesser exposed groups. Company data supplied to the British Occupational
Hygiene Society (BOHS) (British Occupational Hygiene Society, 1968) on X-ray
and clinical abnormalities among 209 employees of a large textile production
facility in Great Britain were analyzed by Berry (1973) in terms of a fiber
exposure-response relationship. The results were utilized in establishing the
1969 British regulation on asbestos. These data suggested that the risk of
developing the earliest signs of asbestosis (rales) was less than 1 percent
for accumulated fiber exposure of 100 fiber-years/ml (f-y/ml), (e.g., 2 fibers/
milliliter (f/ml) for 50 years). However, shortly after the establishment of
the British standard, additional data from the same factory population sug-
gested a much greater prevalence of X-ray abnormalities than was believed to
exist at the time the British standard was set (Lewinsohn, 1972). These data
resulted from use of the new International Labour Office (ILO) U/C standard
classification of X-rays (International Labour Office, 1971) and the longer
time from onset of employment. Of the 290 employees whose clinical data were
reviewed by the BOHS, only 13 had been employed for 30 or more years and 172
had less than 20 years of employment. The progression of asbestosis depends on
both cumulative exposure and time from exposure; therefore, analysis in terms
of only one variable can be misleading.
Extrapolations of risks of asbestos cancers from occupational circum-
stances have been made, although numerical estimates in a specific exposure
circumstance have a large (approximately tenfold) uncertainty. Because of this
uncertainty, calculations of unit risk values for asbestos at the low concen-
trations measured in the environment must be viewed with caution. The best
estimate of risk to the United States general population for a lifetime
2-102
-------
o
continuous exposure to 100 fibers/m is 28 mesothelioma deaths and 5 excess
lung cancer deaths per million females. Corresponding numbers for males are 19
mesothelioma deaths and 17 excess lung cancer deaths per million individuals.
Excess gastrointestinal (GI) cancer mortality is approximately 10 to 30 percent
that of excess lung cancer mortality. These risks are subjective, to some
extent, and are also subject to the following limitations in data: (1) varia-
bility in the exposure-response relationship at high exposures; (2) uncertainty
in extrapolating to exposures 1/100 as much; and (3) uncertainties in conver-
sion of optical fiber counts to electron microscopic fiber counts or mass
determinations (Nicholson, 1986).
2.5.2 Dusts, Sprays, and Cooking Aerosols
2.5.2.1 Introduction. House dust potentially can be a significant source for
the intake of toxic materials through both inhalation and ingestion, especially
for young children who spend much of their time on the floor inside homes
(Roberts et al., 1987). Dust can be a medium for the transfer of toxics from
sources in and outside the house to people and a medium for the reservoir
of toxics, especially pesticides. An estimate of average daily soil dust
ingestion by small children ranges from 0.12 to 1.8 grams (Binder et al.,
1986). The composition of house dust is highly variable, and includes residues
from food and food preparation; hair and skin scale from humans and animals;
fibers from clothing, household furnishings, and building materials; aerosols
from cleaning compounds, waxes, and other consumer products; fragments of
vegetation and humus; and mineral particles. Van Houdt and Boleij (1984)
reported that the indoor airborne particulate matter collected in homes (living
rooms and kitchens) was more mutagenic than the samples collected outdoors.
Studies by Roberts et al. (1987) showed that the dust samples (50 mesh frac-
tions) collected from the rugs in homes by vaccum cleaners were positive in
Ames Salmonella mutation assay and E. coli DMA repair assay.
During the process of cooking, carcinogenic compounds such as PAHs,
(parent compounds), nitro-PAHs, and heterocyclic amines have been found. These
compounds can be emitted in the air, contained or adsorbed onto the food
surfaces. Lijinski and Shubik (1964) first reported the appearance on the
benzo(a)pyrene and other PAH compounds on the surface of charcoal-broiled beef
steaks. Highly mutagenic and carcinogenic heterocyclic amines have been
isolated from cooked food, pyrolysates of ami no acids and proteins, and the
2-103
-------
heated mixtures of creatinine, sugars, and ami no acids (Sugimura, 1985).
Formation of these compounds is highly temperature-dependent. Ohnishi et al.
(1984) reported the mutagenic activity of chicken and the detection of
1-nitropyrene in the meat. Relatively few studies have assessed the health
effects of air emissions during cooking. Lewtas et al. (1987) reported
mutagenic activity in kitchens in a pilot home study. As to the health effects
of sprays used in homes, little information is available.
2.5.2.2 Monitoring. The need for a sampling device that would collect and
maintain the integrity of organic chemicals on house dust was recognized when
the U.S. EPA's Total Exposure Assessment Methodology (TEAM) studies were
planned in 1978. The absence of such a sampler, however, resulted in the
elimination of such monitoring from the TEAM program. The need still exists
and is important to the THE effort. Consequently, initial efforts were begun
in early 1987 to develop a sampler that will combine the features of a home
vacuum cleaner and air sampler, such as that used for high-volume sampling of
pesticides and related SVOCs (Lewis and Jackson, 1982). Further work needs to
be conducted to test the prototype device and refine it as necessary.
2.5.3 IAQ Control Options
The resuspension of particulate matter using common vacuum machines has
been studied and tests have indicated that a sizeable portion of fine particu-
late matter escapes the bags used for collection. The development of a more
efficient vacuum cleaning method for household and office use would be helpful.
Periodic cleaning of furnace and air conditioning filters is helpful in
removing the larger particles. Negative ion generators have been shown to be
helpful in removing particulate matter but their performance is highly variable
among different manufacturers.
The U.S. EPA has issued guidelines for the removal and disposal of
asbestos (U.S. Environmental Protection Agency, 1978). Asbestos-containing
materials that are exposed to indoor air should either be covered to prevent
fibers getting into the air, or removed.
Exposure to fine liquid aerosols from various consumer "aerosol" spray
products can be controlled best by avoiding use of such products. Use under
well ventilated conditions is the minimum prudent option. Face masks designed
for dust are ineffective for such aerosols. Masks designed for organic vapor
control will be partially effective, but may let a substantial fraction of
2-104
-------
submicron particles through. Research on such devices might lead to improved,
and highly effective, designs.
The use of efficient range hoods is advocated during cooking operations.
Currently, however, there is little information on the effectiveness of range
hoods in removing vapor phase compounds, aerosols, and odors from cooking.
This is an area needing investigation.
2.6 NONCOMBUSTION GAS-PHASE ORGANIC COMPOUNDS
2.6.1 Gas-Phase Organic Compounds (Volatile Organic Compounds)
2.6.1.1 Introduction. As many as 300 organic compounds were identified in
homes during the TEAM study conducted by the U.S. Environmental Protection
Agency (Wallace et al., 1986). These compounds can originate as combustion
products or can be emitted from building materials and household chemicals.
It is conceivable that several to many may be present in concentrations that
could be of concern. Each individual compound has physical and chemical pro-
perties that in themselves may cause it to have interactions with biological
systems as well as effects on materials. Some of the compounds have been iden-
tified as irritants and/or neurotoxicants, and some can act as carcinogens,
cocarcinogens, or promoters of cancer in animals and/or humans (Ammann et al.,
1986).
2.6.1.2 Occurrence and Sources of Gas-Phase Organic Compounds. Various stu-
dies of indoor air quality have identified more than 250 different organics
at levels greater than 1 ppb (Sterling, 1985). Many hundreds of additional
compounds undoubtedly exist at lower levels. Individual researchers commonly
measure 30 to 50 separate compounds (De Bortoli et al., 1986; Lebret et al.,
1984; Molhave, 1982; Seifert et al., 1986). An evaluation of the data reported
in these and other studies indicates the following:
1. an extremely wide variety of organic compounds are found in the
indoor environment
2. the range of measured concentrations between different organic
compounds is extremely wide, often two or more orders of magni-
tude; also, the range of concentrations for a specific compound
can vary widely between measurements
2-105
-------
3. in most cases, the concentrations of specific organic compounds
exceeds the outdoor concentration, thus indicating that the
source of the compound is indoors
4. the sources of the organic compounds are quite numerous within
any indoor environment; the sources vary depending on the type
of building studied
While the available IAQ data present a complex picture regarding organic
vapors, there is substantial evidence relating indoor occurrence to emission
sources.
2.6.1.2.1 Outdoor sources of volatile organic compounds. Dozens of organic
compounds have been found in outdoor air and in the exhaled breath of indi-
viduals exposed to these compounds (Wallace, 1986b). Many organic compounds
are also contained in water supplies in sufficient quantity to provide most of
the observed exposure to such compounds. Water has been shown to provide
nearly all of the exposure to three brominated trihalomethanes and to chloro-
form (Wallace, 1986b).
2.6.1.2.2 Sources of indoor organic vapors. Many indoor materials, including
paints, stains, adhesives, and caulks, contain petroleum-based solvents. Such
solvents are comprised of a variety of organic compounds often found in indoor
environments. All general surveys of indoor air quality include the detection
of solvent-based compounds, and there is consistency between the surveys in
the detection of specific organics. For example, Molhave (1982) identified 22
compounds of the same organic species, among the 52 compounds emitted from 42
common materials used in Denmark, that were also identified by Seifert et al.
(1986) from among 43 compounds detected in 159 samples from public buildings
in West Germany. Chlorinated solvents are also used in a wide variety of con-
sumer products and are commonlyfound in indoor air. For example, an EPA study
by Steele (1985) determined the chlorocarbon content of more than 1100 house-
hold products, including shoe polishes, water repellents, cleaning fluids,
epoxy paint sprays, brush cleaners, primers, stains, and varnishes. Table 2-30
indicates the variety of compounds which are emitted from widely used solvents.
This table is meant to illustrate the complexity of the organic emissions from
solvents in indoor materials; it is not an exhaustive list of all solvent-based
organics. It should be noted that many of the compounds in Table 2-30 are not
emitted exclusively from solvents. For example, gasoline vapors from attached
garages and stored fuel may contribute significant quantities of benzene,
2-106
-------
TABLE 2-30. SOLVENT BASED ORGANIC COMPOUNDS
Compound Class Most Commonly Found Compounds
n-Alkanes C-9 (Nonane), C-10 (Decane),
C-ll (Undecane)
Isoalkanes C-9 (Isooctane), C-10, C-ll
Cycloalkanes C-6 (Cyclohexane), C-7, C-8, C-9
Aromatics C-6 (Benzene), C-7 (Toluene),
C-8 (Xylene), C-9 (Trimethylbenzene)
Ketones C-3 (Acetone), C-4 (MEK)
Alcohols C-3 (n-Propanol, iso-Propanol),
C-4 (n-Butanol)
Esters C-4 (Ethylacetate), C-6 (Butylacetate)
Aldehydes C-5 (n-Pentanal), C-6 (n-Hexanal)
Terpenes C-10 (Limonene)
Chlorinated C-l (Carbon Tetrachloride), C-2
Hydrocarbons (Dichloroethane, Trichloroethane,
Dichloroethylene, Trichloroethylene,
Perch!oroethylene)
xylene, and other hydrocarbons; outgassing from chlorinated water is a source
of trichloroethylene and other halogenated organics (Andelman, 1985); and
perch!oroethylene is emitted from dry-cleaned clothes (Wallace et a!.,
1984a,b,c).
While solvent-based indoor emissions come from an extremely wide variety
of materials and products, other organic vapors can be associated more closely
with specific sources. Table 2-31 lists specific compounds and their asso-
ciated sources. As with Table 2-30, this table is not all inclusive, but
illustrates a variety of organic vapor sources.
2.6.1.2.3 Emission rates. To determine the impact of indoor material/product
sources on indoor concentrations of organic vapors, the emission rates for the
various compound/source combinations are required. To date, such emission rate
data have been developed only for formaldehyde from pressed wood products. The
2-107
-------
TABLE 2-31. SPECIFIC INDOOR SOURCES OF ORGANIC VAPORS
Compound
Material Source(s)
Reference
Paradichlorobenzene
Methylene Chloride
Chloroform
Formaldehyde
Styrene
Toluene Diisocyanate
Phthalic Acid Anhydride,
Trimellitic Acid,
Triethylene Tetraamine
Sodium Dodecyl Sulfate
Benzyl Chloride,
Benzal Chloride
Ethylene Oxide
Moth crystals, Room
deodorants
Paint removers
Chlorinated water
Pressed wood products,
Foam insulation (UFFI),
Textiles, Disinfectants
Plastics, Paints
Polyurethane foam
aerosols
Epoxy resins
Carpet shampoo
Vinyl tiles plasticized
with butyl benzyl
phthlate
Sterilizers (Hospitals)
Nelms et al.
(1987)
Girman and
Hodgson (1986)
Wallace (1986b)
National
Research
Council (1981)
Wallace (1986b)
Carroll
et al. (1976)
Fawcett
et al. (1977)
Kreiss et al.
(1982)
Rittfeldt
et al. (1984)
U.S. Department of Housing and Urban Development (HUD) has developed standards
for formaldehyde emissions from pressed wood products used in mobile homes
(Federal Register, 1984). These standards involve chamber testing of materials
to determine compliance with emission restrictions. As part of the HUD regula-
tion, the Berge equation is used to evaluate the chamber test results. The
Berge equation relates formaldehyde concentration at standard conditions (e.g.,
25°C, 50 percent relative humidity) to the test chamber concentration at
different values of temperature and relative humidity (Godish and Rouch, 1985).
The Berge equation predicts increasing emissions with increasing temperatures
and decreasing relative humidity. Matthews (1986) and his co-workers at Oak
Ridge National Laboratory have developed other models for determining formal-
dehyde emission rates which incorporate the following variables: temperature,
2-108
-------
relative humidity, air exchange rate, product loading, and formaldehyde concen-
tration in the chamber. Matthews' models show that emission rates increase
with increases in air exchange rate, decreases in product loading, and
decreases in chamber concentration. EPA's Air and Energy Engineering Research
Laboratory conducted a cooperative project with Matthews to compare small
chamber testing procedures for determining emissions from particleboard. Table
2-32 provides emission rates for formaldehyde and other organics for various
test conditions (Nelms et al., 1986). All tests were conducted at 23°C and 50
percent relative humidity on particleboard aged approximately eight months.
The emission rates in the bottom row of Table 2-32 are representative of air
exchange rates and loadings found in residential environments. Material newer
than eight months could be expected to have higher rates. The material tested
was low-density particleboard normally used in home construction (e.g.,
flooring material). Medium-density particleboard, such as is used in furni-
ture, has formaldehyde emission rates two to four times higher (Matthews,
1986). Plywood products generally have lower emission rates than particle-
board. Note that the formaldehyde emission rates are related to the air
exchange and loading relationships discussed above. The emission rates for the
other organic compounds do not vary substantially with these two parameters and
appear to be limited by the rate of diffusion to the surface of the particle-
board.
TABLE 2-32. EMISSION RATES FROM PARTICLEBOARD (pg/m2 hr)
Air Exchange
(hr"1)
2.71
0.54
3.61
0.54
Loadi ng
(mVm3)
1.96
1.17
0.78
0.39
Formaldehyde
154
95
230
140
Acetone
37
41
38
37
Hexanal
15
26
20
24
Others*
27
26
31
27
The sum of emission rates for propanol, butanone, benzaldehyde, and benzene.
Limited research has been conducted on the emission rates of organics from
materials other than pressed wood products. Molhave (1982) conducted emission
rate tests on 42 materials using a one cubic meter test chamber with a ventila-
tion rate of one air change per day. Girman et al. (1984) evaluated emissions
2-109
-------
from 15 adhesives using 3.8-liter cans with an air exchange rate of 14 per
hour. EPA has conducted emission tests on caulk, adhesive, floor wax, and wood
stain using a 166 liter chamber at various air exchange rates (Tichenor et al.,
1986). The Saskatchewan Research Council is conducting emissions testing of
caulking compounds. Georgia Tech Research Institute conducts chamber testing
of building materials.
It is difficult to evaluate the limited data on emission rates of organic
vapors from indoor materials/products because standard testing protocols have
not been developed. A number of test conditions are critical to an effective
determination of emission rates, including: temperature, relative humidity,
air exchange rate, and product loading (area of sample/volume of test chamber).
In addition, the effect of chamber concentration and chamber "wall effects"
must be determined. Finally, the age or condition of the sample affects
emission rates. Any emission rate data must be coupled with a clear descrip-
tion of all of the variables and phenomena in order to evaluate and compare
such data. To show the variability in reported emission rates, Table 2-33
compares emission rate data for two materials based on tests by three investi-
gators. No judgment can be made as to which of the values in the table are the
"right answers" without a rigorous evaluation of the variables discussed above.
2.6.1.3 Monitoring of Gas-phase Organics. Previous total human exposure stud-
ies (Pellizzari et al., 1982; Wallace, 1986b) have indicated that indoor air
concentration of many or most volatile organic chemicals are substantially
higher than outdoor levels and constitute a major route of exposure to these
chemicals. The principal VOCs of concern are halogenated aliphatic and
aromatic hydrocarbons and benzene and its homologs. Sources of these VOCs
within residences include building materials and furnishings, cleaning
solvents, fuels, and releases from tap water (especially during showering).
Many halogenated and benzene family VOCs are known or suspected human
carcinogens.
Nearly all monitoring studies for VOCs have depended upon collection on
Tenax®-GC sorbent tubes. This sorbent is known to suffer from many problems
associated with artifact formation (Walling et al., 1986), very limited
capacity for the more volatile compounds (e.g., vinyl chloride and methylene
chloride), and for the polar organics (e.g., acrylonitrile and ethylene oxide).
Background contamination of the Tenax® is variable from batch to batch. Parti-
cular problems occur with toluene, benzene, and to a lesser extent, styrene.
Occasional high background contamination and variabilities are encountered with
2-110
-------
TABLE 2-33. COMPARISON OF ORGANIC EMISSION RATES: SILICONE CAULK AND
FLOOR ADHESIVE
Investigator
Girman et al.
(1984b)
Molhave
(1982)
ii
Tichenor and
Mason (1986)
ii
n
ii
n
Chamber
Volume
(liter)
3.8
1000
n
166
n
n
n
n
Air Exchange
Rate Time*
(per hr) (hr)
14 (1)
0.04 (2)
n n
1.8 0.5
1
5
1
5
Emission
Material
Floor
Adhesive
Floor
Adhesive
Silicone
Caulk
Floor
Adhesive
n
n
Silicone
Caulk
n
Factor
(mg/m2 hr)
140 - 180
271
26
1700
700
100
20
1.6
*The time elapsed since the sample was applied to the substrate.
(1) The sample was dried for 9-14 days prior to testing.
(2) The sample was brought to equilibrium prior to testing.
chloroform and 1,1,1-trichloroethane. Careful preparation and cleanup proce-
dures, coupled with extensive precautions during transport and storage, are
necessary to insure high quality data.
Despite the limitations of Tenax®, a monitoring system was successfully
employed during 1979-85 in the EPA TEAM studies involving 600 residents in
four states (Wallace, 1986b). Battery-operated pumps capable of 12-hour con-
o
tinuous operation at 30 cm /min flow rate were used to sample air through a
cartridge containing 2 g of Tenax-GC. The pump and cartridge were carried in
a specially-designed vest worn by the participants. The sampling system was
found to be highly acceptable to the participants and proved to be very
reliable (less than 1 percent of samples were lost due to pump failure).
Approximately twenty VOCs were monitored with sensitivities in the 10 to 100
3
ng/m range and mean relative standard deviations (RSD) in the 25 to 35 percent
range (exception: benzene, percent/RSD = 45 percent).
2-111
-------
SVOCs from combustion are usually collected in the particulate phase in
which they contribute to the mass loading as RSP. They can be identified and
quantified (subject to errors discussed later) using extraction techniques fol-
lowed by analyses of the extracted material. Some of these materials are vola-
tile and will evaporate from the filter during a 24-hour collection. Unless an
adsorbent tube is placed in series behind the filter this material will be
lost (Jacob et al., 1986). As described above for RSP, VOCs and SVOCs are
heterogeneous and variable mixtures that have component toxicities ranging
from inert to toxic or carcinogenic. Indoor VOCs have been measured as concen-
trations by Lebret et al. (1986), Seifert et al. (1984), and De Bortoli et al.
(1984), and as nominal exposures by Wallace et al. (1986). The TEAM study data
represent true indoor nominal exposures since the subject carrying the monitor
spent 12 hours indoors (Wallace et al., 1986). These data represent 12-hour
averages for a sample of subjects in a given city during a specific season.
For instance, the arithmetic mean is robust with the averaging time, but both
the variance of these data and the higher percentile values will decrease with
averaging time, so that the percentiles of indoor exposure that are greater than
these values will be much less. For example, Lebret et al. (1986) report data
on four homes in a time series of alternating weekly average indoor concentra-
tions, and it can be seen that the variance of these values is considerably less
than the TEAM study values. Because these data contain time periods when the
home was vacant, they should not be compared directly with the TEAM data, even
when allowing for the difference in analytical techniques. Table 2-34 gives
the ranges of 12-hour averages, but for interpretation in relation to possible
irreversible effects of the carcinogenic materials in the mixture, the annual
average exposures need to be predicted.
A passive monitor for VOCs was developed during 1983-85 (Lewis et al.,
1985; Coutant et al., 1985). This device is a small, stainless steel cylinder
(3.8 cm oc x 1.2 cm), weighs 36 g, and contains a small bed of Tenax@-GC (0.4
g) or other sorbent. The dual-faced device is thermally desorbable and has
o
very high efficiency. Effective sampling rates of 75 to 100 cm /min for most
VOCs permit detection limits of 1 ppbv or less with exposure times of 1 to
3 hours. Problems due to reversible sorption, which is analogous to break-
through in active samplers, arise at loadings greater than 1 ug of VOC (usually
after several hours exposure for the more volatile VOCs). Mathematical correc-
tions can be applied if break-through volumes are known, however. The device
2-112
-------
TABLE 2-34. WEIGHTED 12-HOUR MEASUREMENTS FOR BREATH (ug/m3) OF RESIDENTS FOR BAYONNE AND ELIZABETH, NO, COMBINED-TEAM FIRST SEASON
ro
i
Estimated Population Size
Minimum Sample Size:
Maximum Sample Size:
Compound
Chloroform
1,1,1-Trichloroethane
Benzene
Carbon Tetrachloride
Tri chl oroethyl ene
Tetrach 1 oroethyl ene
Styrene
m , p-Di chl orobenzene
Ethyl benzene
o-Xylene
m,p-Xylene
: 128,603
295
339
Mid
Q.L.a
2.08
2.30
0.44
1.70
1.50
4.10
0.97
1.32
0.40
1.10
0.52
Arith.
Mean
3.12
15.0
18.7
1.31
1.77
13.3
1.15
8.10
4.58
3.35
8.95
h
S.E.b
0.34
2.57
1.40
0.26
0.21
1.83
0.13
1.54
0.55
0.36
0.93
Geo..
Meanc
1.30
4.79
8.19
0.60
0.93
7.33
0.72
1.72
2.45
1.99
5.34
H
S.E.
1.13
1.13
1.22
1.15
1.13
1.08
1.11
1.12
1.14
1.12
1.12
Percentiles
Median
1.80
6.60
12.0
0.69
0.88
6.80
0.79
1.30
2.90
2.20
6.35
75
3.70
13.0
24.0
1.06
1.80
12.9
1.25
3.50
5.30
3.70
11.0
90
8.20
30.0
42.0
2.25
3.94
31.0
2.40
21.0
8.90
6.30
19.0
95
11.5
42.0
56.0
2.7
5.9
44.0
3.0
44.0
12.0
9.2
21.0
99
26.0
185
120
20.0
14.0
190
7.2
110
29.0
17.0
53.0
Range
.05-29.0
.06-520
. 02-200
.05-250
.08-30.0
. 12-280
.06-31.0
. 11-158
. 02-290
. 05-220
. 05-350
aMid Q.L. = Median Quantifiable Limit
bS.E. = Standard Error of Arithmetic Mean
cGeo. Mean = Geometric Mean
S.F. = Geometric Standard Error = exp(s) where s is the standard error of the weighted mean of log(x).
Source: Wallace et al. (1986).
-------
was modified to reduce the effective sampling rate by a factor of 20 to 25 to
permit 24-hr exposures for nearly all VOCs. A high-rate version is commer-
cially available from Scientific Instrument Specialists.
Because of increasing concerns over the reliability of Tenax®-based VOC
data, canister-based collection systems have recently been under development.
These systems are based on the use of stainless steel canisters which have
specially electropolished interiors (McClenny et al., 1986). The proprietary
process, known as SUMMA® polishing, passivates the interior walls of the canis-
ters so as to substantially improve the storage capabilities for VOC. Tests
have shown that most VOCs of interest can be stored in the canisters for as
long as 30 days without significant losses (Oliver et al., 1986). Even
reactive species such as ethylene oxide can be kept for at least one week.
The major effort to date has been on pump-based systems which can achieve
24-hour sampling by filling evacuated canisters to 3 atm pressure. A major
comparison of the pump-based system with pumped Tenax® samplers (used in the
distributed air volume mode: 5, 10, 20 and 40 1; see Walling, 1984); was
conducted in a fully furnished, but unoccupied residence in 1986 (Spicer et
al., 1986). The HVAC system of the house was spiked with VOCs spanning a range
of vapor pressures and chemical types so as to produce concentrations of 3 to
o
30 mg/m (low ppbv) of these chemicals inside the air of the living space. The
agreement between Tenax® distributed air volume (DAV) and canisters was very
good. Agreement between the two as characterized by the slope of the linear
regression analysis of the Tenax®-collected sample to canister-collected sample
was: chloroform, 0.802 + 0.032 (standard error); 1,1,1-trichloroethane, 0.857
+ 0.028; tetrachloroethylene, 1.095 + 0.086; bromodichloromethane, 0.862 +
0.018; trichloroethylene 0.916 + 0.019; benzene, 1.081 + 0.141; toluene, 1.030
+ 0.056; styrene, 0.928 + 0.036; p-dichlorobenzene, 0.992 + 0.047; hexachloro-
butadiene, 0.969 + 0.033 (average of four values). The average ratio was
0.953. This comparison represents an ideal situation in which concentrations
are reasonably constant during sampling periods and, for most of the tests, in
which temperature and humidity excursions were minimized. The low- and high-
rate PSDs were also included in this intercomparison test. The data base
resulting from this comparison is currently being examined to determine a
number of secondary matters; for example, the changes in agreement when one- or
two-tube Tenax® is used. Also of considerable interest are the results of a
similar comparisons on ambient air.
2-114
-------
A canister-based sampler designed for intermittent sampling over a week-
long collection period is being tested to determine the likelihood of component
failure. This approach minimizes the need to attend the sampler, at the risk
of misrepresenting the contributions of time-varying VOC sources. Typical
temporal variabilities for indoor air-related sources are currently being
measured. These measurements are possible because of related methods develop-
ment of continuous sequential samplers utilizing SUMMA®-polished stainless
steel syringes. Twelve syringes can be used to sample over periods of as long
as eighteen hours and as short as twelve minutes. Commercially available units
of this type have been modified to allow interaction between a GC/MS and the
sampler by way of GC program commands executed through external electrical
latches. This allows automatic transfer of sample from the syringes to the GC
preconcentrator followed by an automated analysis. All twelve syringes can be
analyzed without manual intervention.
VOC screening procedures for use in reducing the number of canister
samples for which detailed GC/MS analysis is done, and for locating likely
indoor sources, are also undergoing evaluation. Portable gas chromatographs as
well as the devices such as the Photovac TIP that give weighted response to
total trace gases in the air can be applied in these cases.
Based on previous extensive work in monitoring personal exposures in TEAM
studies and on the generally favorable results obtained in the indoor air
intercomparison, single-tube Tenax® and its passive counterpart, the EPA-
developed passive sampling device, remain viable alternatives for VOC sampling,
at least in situations in which environmental conditions are controlled.
Follow-up experimental efforts on the PSDs are in the planning stage. This
will include the alteration of an existing commercially available system to
accommodate PSDs, Tenax® cartridges, and canisters, so as to facilitate direct
comparisons between them. The cause of uncertainties in the sampling rate
applicable to the PSDs will also be studied so that further design modifica-
tions can be implemented if necessary.
Further research is needed to reduce the costs of the canister-based
samplers and to improve the PSDs. New sorbents and mixed-bed sorbent tubes
need to be evaluated for improved collection efficiency and reduced artifact
problems. Badly needed is a sampling/analytical methodology for small polar
organic compounds which are difficult to separate from atmospheric water.
These include ethylene oxide and nitriles.
2-115
-------
2.6.1.4 Health Effects.
2.6.1.4.1 Neurotoxiclty of volatile organic compounds. Monitoring surveys of
homes and public buildings worldwide (Berglund et al. , 1984) indicate that a
bewildering assortment of VOCs is present in such environments. However, the
levels of individual VOCs found are generally orders of magnitude below the
threshold limit values (TLVs®) that is, levels considered to be harmful to
humans exposed for eight hours to any individual compound. Among the VOCs
found are acetone, formaldehyde, methyl ethylketone (MEK), hexane, benzene,
toluene, and xylene, used in building materials, furnishings, and adhesives.
Chlorinated hydrocarbons frequently found include methylene chloride (paint
strippers), trichloroethane (paint), trichloroethylene (type writer correction
fluid and degreasing agents) and p-dichlorobenzene (insect repellents). Some
of these VOCs (e.g., formaldehyde, benzene) are known to be carcinogenic, as
discussed elsewhere in this chapter. Many of the individual VOCs such as
n-hexane, MEK, and toluene are also known to be neurotoxic, but at levels much
higher than those found in typical new buildings. What is striking, however,
in reviewing the known neurotoxicological effects of exposure to many of these
compounds, is the commonality of these effects across compounds (Anger and
Johnson, 1985). Table 2-35 summarizes neurotoxic effects of some common indoor
VOCs. The most frequently observed effects include CNS depression, unconscious-
ness, vertigo, and visual disorders. Tremor, fatigue, anorexia, weakness, and
other neurotoxic effects have been associated with VOCs somewhat less often.
Cognitive effects such as memory impairment and mental confusion have also been
reported. Neurologic symptoms such as these are recognized in several
Scandinavian countries as the "phycho-organic" or "painters" syndrome (World
Health Organization, 1985). Although Anger and Johnson (1985) do not indicate
the exposure levels at which these effects were observed for individual
chemicals, the pattern of documented effects suggests a number of neuro-
behavioral endpoints that should be evaluated in future indoor air studies.
The validity and organic basis of this syndrome are quite controversial (Grasso
et al., 1984), but there is extensive literature on the neurobehavioral
consequences of occupational exposure to industrial solvents that is germane to
the present review. Review of the acute and chronic effects of exposure to
industrial solvents, particularly compounds such as "white spirits", should be
useful in selecting appropriate measures to assess the neurotoxicity of complex
VOC mixtures.
2-116
-------
TABLE 2-35. NEUROTOXIC EFFECTS OF VOLATILE ORGANIC COMPOUNDS COMMONLY FOUND IN
INDOOR ENVIRONMENTS1
Chemical
Acetone
Benzene
*l-Butanol
*2-Butanone
(methyl ethyl ketone)
*n-Butyl acetate
Carbon tetrachloride
Chloroform
Cyclohexane
p-Dichlorobenzene
*1,2 Oichloroethane
*Ethyl benzene
Formaldehyde
*n-Hexane
Methylene chloride
Styrene
Tetrachl oroethy 1 ene
Toluene
1,1,1-Trichloroethane
(methyl chloroform)
Tri chl oroethy 1 ene
(acetylene
trichloride)
*3-xylene
A AN
X
X X
X
X
X X
X
X X
X X
X
X
D E
X
X X
X
X
X
X X
X
X
X
X
X
X
X
X
X
X
X
X X
Neurotoxic Effect2
F I M MI P S T
X X
X
X
XXX
X
X X
XX X
X X
XXX
X X
X XXX
X
XXX X X
XXX X
X XXX X
u
X
X
X
X
X
X
X
X
X
X
X
X
X
X
V
X
X
X
X
X
X
X
X
X
X
X
X
X
vs w
X
X
X
X X
X
X
X X
X X
X X
X X
X
X X
*VOCs used in Molhave et al. (1986) mixture
Adapted from Anger and Johnson (1985).
2See abbreviation key. Neurotoxic effects listed were generally observed at levels con-
siderably higher than the levels of individual VOCs found in typical indoor environments.
A = anesthesia
An = anorexia
D = CNS depression
E = excitement
F = fatigue
I = incoordination
M = mental confusion
MI = memory impairment
P = paresthesia
S = sensory disturbance
T = tremor
U = unconsciousness, stupor, narcosis
V = vertigo
VS = visual disturbances
W = weakness
2-117
-------
Molhave (1985) has hypothesized that the sensory irritant symptoms of
sick-building syndrome (SBS) result from the additive or synergistic effects of
the complex mixture of VOCs, rather than the specific effect of any individual
VOC present in any particular "sick" building. Molhave et al. (1986) have
explored the effects of controlled exposure of humans to complex mixtures of
VOCs. Results of this study suggest that low-level VOC exposure in amounts
comparable to concentrations found in newly constructed Danish homes produced
memory impairment and sensory irritation in subjects known to be sensitive to
VOCs—that is, persons identified by questionnaire as having "sick building
syndrome". These findings are provocative because the available literature on
individual compounds contained in the VOC mixture would not indicate any
adverse effects at such low levels. The literature concerning the health
effects of complex VOC mixtures, however, is negligible. In view of the
widespread sensory irritant complaints of workers in new construction and
renovated buildings and the known presence of numerous VOCs in such environ-
ments, systematic study of the health effects of human exposure to complex VOC
mixtures is needed. In particular, the pioneering work of Molhave et al. needs
to be replicated and clarified as a first step in exploring the neurotoxicity
of VOCs.
2.6.1.4.2 Genotoxicity of volatile organic compounds in relationship to total
organic species. Graedel et al. (1986) have recently catalogued, from pub-
lished sources, compounds found within ambient and indoor environments, their
sources, air media (e.g. aerosol, gas,) in which they have been detected and,
when available, the genotoxicity test results of these compounds. The
chemicals are grouped according to International Union of Pure and Applied
Chemists (IUPAC) guidelines. Under the IUPAC chemical classification scheme, a
compound is listed only under one class. Table 2-36 provides a condensation of
this information for compounds found in indoor air.
Depending upon the source, airborne particle concentration, and humid-
ity, these compounds can be found in the gas phase (e.g., volatile organics),
associated with airborne aerosols, or both. Table 2-37 shows that this
literature review identified 273 chemicals as indoor air compounds and that
218 (or approximately 80 percent) could be detected as volatile compounds.
Of these 218 compounds, 39 were identified as both volatile and aerosol-bound
compounds; therefore, 65 percent (179) of identified indoor air compounds were
identified only as gases. Of the 273 compounds that have been identified within
2-118
-------
TABLE 2-36. PARTITIONING OF INDOOR AIR COMPOUNDS AS INDICATED
BY REVIEWED LITERATURE*
Chemical Class
Inorganics
Hydrocarbon
Ethers
Alcohols
Ketones
Aldehydes
Acid derivatives
Carboxylic acids
0-heterocycles
N-organics
S-organics
Halogenates
Organometallics
Number
Detected
14
110
2
19
10
14
17
19
5
10
2
38
13
273
Gas
8
82
1
17
9
12
7
1
5
9
1
21
6
179
Gas/
Aerosol
1
10
1
1
1
2
1
3
1
1
11
7
39
Aerosol
5
18
1
10
14
1
1
6
55
^Extracted from Graedel et al. (1986). Summarized by whether or not each
compound was found as a gas or within an aerosol or both (Gas/Aerosol).
indoor air samples, 70 have undergone some degree of genetic toxicology testing
(Table 2-37). Among the 70 compounds tested, 31 gave a positive (active)
response in at least one bioassay, 33 were negative in the bioassays used, and
6 had a questionable response in at least one bioassay. Twenty-two have been
tested in a whole-animal carcinogen bioassay, and 16 of these proved to be
carcinogenic in a rodent bioassay. Two chemical classes — hydrocarbons and
halogenated organics -- are worthy of special discussion. Among the 110
hydrocarbons detected within indoor air, only about 10 percent have been tested
for mutagenicity or carcinogenicity. Six of these 13 hydrocarbons have been
tested in whole-animal carcinogen bioassays, and 4 of these were positive
(carcinogenic). Most of these 13 hydrocarbons are formed as a result of some
type of combustion process including tobacco smoking, wood combustion, oil
combustion, and coal combustion. Some of these hydrocarbons (e.g., toluene,
xylene,) are solvents and are associated with adhesives, building materials,
printing processes, and paints. For this class of chemicals, short-term
bioassays appear to be appropriate screening tests for detecting airborne
carcinogens. The other major class of compounds — halogenated organics, is
quite different from the hydrocarbons. Of the 38 halogenated compounds detected
within indoor air, 27 have undergone bioassay testing and 13 of these are
2-119
-------
TABLE 2-37. CHEMICALS DETECTED IN INDOOR AIR SAMPLING:
THEIR OCCURRENCE AND BIOASSAY*
Chemical Class
Inorganics
Hydrocarbon
Ethers
Alcohols
Ketones
Aldehydes
Acid derivatives
Carboxylic acids
0-heterocycles
N-organics
S-organics
Halogenates
Organometallics
Totals
Number
Detected
14
110
2
19
10
14
17
19
5
10
2
38
13
273
Number
Bioassayed
3
13
0
6
2
5
3
1
1
3
0
27
6
70
Positive
Bioassay
2
5
0
1
0
2
2
0
1
2
0
13
3
31
Positive Animal
Carcinogen
0
4
0
0
0
1
2
0
1
1
0
7
0
16
*Extracted from Graedel et al. (1986).
positive in some bioassay. Of the nine halogenates tested for animal carcino-
genicity, seven were carcinogenic and two gave inconclusive results. These
halogenated organics are primarily either solvents (e.g., methylene chloride)
or pesticides (e.g., chlordane). One of the halogenated compounds is tetra-
methyl lead, which is associated almost exclusively with gasoline vapors.
Some of the short-term tests, especially bacterial bioassays, are not effective
screening systems for the carcinogens that fall into this class of compounds.
One must remember, however, that very few gaseous compounds have undergone
genetic bioassay testing. Both the identification of aerosol compounds and the
testing of volatiles may likely be a function of scientific capabilities and
interest rather than a true indication of the distribution of genotoxic
compounds between gases and aerosols. Although one recent effort (Claxton,
personal communication) has shown that approximately one-half of the total
mutagenicity associated with sidestream cigarette smoke is due to the volatile
and semivolatile components. Complex mixtures of gases pose special problems
for genetic toxicologists. Neither the technology for collecting complex
mixtures of gases and returning them to the laboratory for bioassay nor the in
situ bioassay methods for testing indoor gases are fully developed. Future
2-120
-------
efforts are expected to address both of these problems; however, in the mean-
time individual gaseous compounds will have to be identified and then tested as
single components.
2.6.1.5 Mitigation and Control Options.
2.6.1.5.1 Ventilation. The most common technique for reducing the concentra-
tion of organic vapors in indoor environments is modifying the ventilation in
the affected area (Berglund et al., 1982; Moschandreas et al., 1981b; Billings
and Vanderslice, 1982). Increasing the air exchange rate by bringing in
outside air will reduce concentration by dilution and flushing. In some cases,
this same technique will increase the emission rate by enhancing the vapor
pressure driving force for vaporization. Thus, for some compounds higher air
exchange rates can reduce concentrations and exposure time. In many cases,
however, a simple increase in the amount of dilution air to a building may not
be effective. The flow patterns in the affected rooms and the amount of air
actually reaching the contaminated zones must also be considered (Skaaret,
1986). Any changes in the ventilation system design or operation will have
energy and cost implications that will need to be considered.
2.6.1.5.2 Air cleaners.
2.6.1.5.2.1 Adsorption. A number of systems for collecting organic
vapors on adsorbents, primarily activated carbon, are available for indoor air
quality control (Research Triangle Institute, 1986). Most are designed for
commercial and industrial application, although three manufacturers produce
units small enough for residences or single rooms. Units are available both
for induct installation or as "stand alone" devices. Almost no data are
available on the effectiveness of these units at vapor concentrations below 100
ppm. Adsorbers are also used in submarines to control indoor air quality; both
activated carbon and molecular sieves have been employed (Rao, 1986).
2.6.1.5.2.2 Catalytic oxidation. Collins (1986) has reported on a low-
temperature catalyst used in a room air filtration device. While designed to
remove a variety of indoor contaminants (e.g., combustion gases and tobacco
smoke), the device also reduced the concentration of acetaldehyde, acetone,
methylene chloride, and MEK. Benzene was not effectively removed. Rao (1986)
reports on the use of catalytic oxidation for cleaning the air in submarines.
Further evaluations are required to determine the effectiveness of catalysts
in reducing the concentration of organic vapors in indoor environments.
2-121
-------
2.6.1.5.3 Material/product selection. The levels of organic vapors in the
indoor environment can be affected by the selection of materials and products.
As discussed previously, HUD limits the emission rates of formaldehyde from
pressed wood products used in mobile homes to reduce the concentrations in
these residences. Urea formaldehyde foam insulation (UFFI) is no longer used
due to excessive emissions. Denmark requires the labeling of epoxy and poly-
urethane products to inform the consumer of potential health hazards (Andersen
et al., 1982). Judicious selection of low-emission materials and avoidance of
known irritants would limit exposures. Experiments by Molhave et al. (1984)
showed that, for a typical mixture of organic compounds found in indoor air,
people start to become irritated at a total concentration of about 5000 |jg/m .
Since a typical building has many sources that contribute to the total concen-
tration, Tucker (1986a,b) has suggested that it would be prudent to keep the
o
contribution of any one source below about 1000 ug/m . In very rough terms
this means that material sources should be avoided or conditioned if they emit
o
more than 100 mg/hr of organic compounds per 100 m of residential space with
mixed and well distributed air, or sources emitting more than 400 mg/hr per 100
2
m of typical office space with well mixed and distributed air. Unfortunately,
most indoor products have not been sufficiently characterized to allow such
selection to occur.
2.6.1.5.4 Material/product use. The manner in which products are used can
impact indoor organic vapor concentrations. Emissions of formaldehyde from
pressed wood products decline with time (Hawthorne et al., 1984), so aging or
conditioning of such products prior to installation would reduce emissions.
Solvent-containing materials (e.g., paints, adhesives, caulks, paint removers,
and waxes) should be used in well ventilated areas. Manufacturers' instruc-
tions should be followed. Some activities, such as hobbies, woodworking, and
paint stripping, which use high-emission products should be isolated and
separate exhaust fans provided. Solvents and solvent-containing products
should be stored in airtight containers; outside storage is preferred.
2.6.1.5.5 Other measures. Other measures for controlling the levels of indoor
organic vapors include:
building "bake out", the process by which new or renovated
buildings are heated and ventilated prior to occupancy (State
of California, 1984); and
2-122
-------
formaldehyde removal via fumigation. Dement et al. (1984)
report that fumigation with 1000 ppm of ammonia for 24 hours
reduced formaldehyde emissions by more than 70 percent. The
long-term effectiveness of this approach is not known.
2.6.2 Formaldehyde
One noncombustion gas-phase organic compound is HCHO, which is also a
product of combustion. This is a colorless, pungent gas which is highly water
soluble and hence irritating to the mucous membranes of the eyes and respira-
tory tract. It is classed as a VOC, but because of the prevalence of its use
in building materials such as particleboard, carpeting, and cloth finishing, it
is described separately here.
2.6.2.1 Sources of Formaldehyde. Formaldehyde is emitted from UFFI and from
resins used in plywood and particleboard. Emissions from gas stoves and
burning cigarettes also contain formaldehyde (Repace, 1982). Carpeting may
also be a source of formaldehyde (Molhave, 1982; Pickrell et al., 1984).
Anderson and Lundquist (Pepys, 1982) concluded that the use of particleboard
and plywood in furniture construction causes inordinately high levels of
formaldehyde due to the use of urea-formaldehyde glue in construction. Many
textile products such as draperies, rugs, and upholstery fabrics may have a
long-lasting emission of formaldehyde due to the treatment the textiles
receive.
Most of the material source research done to date in this country has been
on formaldehyde emissions from pressed wood products, especially the various
types of particleboard. Regulations for formaldehyde emissions from pressed
wood products were established by HUD in 1984. Prior to that time, research
measurements at Oak Ridge National Laboratory (ORNL) and other laboratories
were being made so as to understand how ventilation parameters like tempera-
ture, relative humidity, and air exchange rates affect emission rates of
formaldehyde. Although such work continues at ORNL, the National Bureau of
Standards (NBS), and Georgia Tech Research Institute, formaldehyde emission
measurements have shifted more toward testing of products to determine compli-
ance with the HUD standard.
2.6.2.2 Monitoring of Formaldehyde. The lack of a sensitive, accurate and
inexpensive method for monitoring formaldehyde in nonoccupational environments
has been a serious obstacle to the performance of exposure studies and has
2-123
-------
prompted the development of improved techniques for the collection and analysis
of formaldehyde.
Three PSDs and two active sampling devices have been considered by EPA as
PEMs for formaldehyde. The PSD which has seen the most use, especially in
Canada, is a dry sulfite device manufactured by Air Quality Research, Inc.
(AQR) (Geisling et al., 1982). The AQR PSD (PF-1) resembles a Palmes tube and
is designed for indoor area monitoring; however, it could be worn on the
person. It has a sensitivity of about 2 ppm-hr; therefore, five to seven days
of exposure are generally necessary for nonoccupational levels of formaldehyde.
In 1985 EPA conducted chamber tests of the AQR PF-1 device and compared it
directly with an active sampler (Snow, 1985). It was recommended that the PF-1
not be used for exposure periods of less than four days because sampling rates
were found to be time-dependent.
A new commercial PSD from the Air Technology Labs. , Inc. (ATL) is the
passive bubbler monitor. The ATL device employs a permeable membrane to
separate the reagent (3-methyl-2-benzothiazolone hydrazone, or MBTH) from the
atmosphere sampled. The manufacturer claims a detection limit as low as 0.1
ppm for a 2-hr exposure, and the device has a distinct advantage in that it can
be developed and read by untrained personnel in the field. To date, however,
EPA has not had funds to test this device.
Prior to the advent of the ATL PSD, a Canadian company known as Crystal
Diagnostic, Inc. (CDI) introduced a prototype PSD containing a film of mono-
dispersed hydrobenzoic acid hydrazide. Upon exposure to formaldehyde, crystal
nuclei form, which after development can be read visually by an untrained indi-
vidual in the field. Attempts to improve the sensitivity of the CDI device
to permit detection of 10 ppb of formaldehyde after 8- to 24-hr exposures
have not been successful. CDI has recently been acquired by Foxboro, which
stated that it intends to continue developing the PSD.
There are two pumped-tube methods for measuring formaldehyde which do have
adequate sensitivity and can be used for either indoor air or personal exposure
monitoring. Both utilize solid sorbents coated with 2,4-dinitrophenylhydrazine
(DNPH) and require analysis by HPLC. Consequently, analytical costs are high.
One method developed by Lipari of General Motors Research Laboratory (Lipari
and Swarin, 1985) is applicable only to formaldehyde and has a detection limit
of less than 2 ppb hr. The other method, developed by EPA (Tejada, 1986), has
comparable sensitivity but offers the added advantage of the capability to
2-124
-------
simultaneously determine other aldehydes and ketones. In 1986 these methods
were compared by EPA for outdoor air monitoring and found to have problems with
blanks used to zero the monitors for measurement of samples. These problems
should, however, be minimal for indoor applications where formaldehyde concen-
trations are generally much higher than for ambient air. It is anticipated
that the blank problems can be eliminated through better packaging for shipment
to and from the laboratory. EPA has selected the Tejada method (Tejada, 1986)
for inclusion in the Toxics Air Monitoring System (TAMS) in 1987.
2.6.2.3 Health Effects. Because of formaldehyde's high water solubility, it
causes irritation to the mucous membranes of the eyes and upper respiratory
tract. A variety of short-term signs and symptoms are commonly accepted as
associated with formaldehyde exposure; some occur at levels that have been mea-
sured in residential air (Table 2-38) (National Research Council, 1981a). How-
ever, other acute and chronic health effects of formaldehyde are more contro-
versial, including human carcinogenicity, effects on the lungs, and neuro-
behavioral impairment.
TABLE 2-38. ACUTE HUMAN HEALTH EFFECTS OF FORMALDEHYDE
AT VARIOUS CONCENTRATIONS*
Formaldehyde Concentration (ppm) Reported
0.0 -
0.05 -
0.05 -
0.01 -
0.10 -
5 -
50 -
>
0.5
1.5
1.0
2.0
25
30
100
100
None reported
Neurophysiologic
Odor threshold
Eye irritation
Upper airway irri
Lower airway and
Pulmonary edema,
Death
Effects
effects
tat ion
pulmonary effects
inflammation, pneumonia
^Adapted from Table 7-2 (National Research Council, 1981a).
As measured by determination of optical chronaxy, electroencephalography,
and sensitivity of dark adapted eyes to light.
The low concentration (0.01 ppm) was observed in the presence of other
pollutants that may have been acting synergistically.
Formaldehyde is primarily deposited in the upper respiratory tract,
including the nasopharynx; and cancer of structures in this region could be a
result of formaldehyde exposure. In 1979, it was reported that rats exposed to
formaldehyde developed nasal cancer, a tumor rarely found in unexposed animals
2-125
-------
(Kerns et al., 1983). This report, subsequently confirmed (Albert et al.,
1982), stimulated the rapid performance of a number of epidemiological investi-
gations.
Occupational exposure to formaldehyde has been examined as a risk factor
for nasal cancer, as well as for several other sites, in both case-control and
cohort studies. Retrospective cohort studies of formaldehyde-exposed workers
have not shown excess nasal cancer, but their statistical power has been
limited (Blair et al., 1986; Stayner et al. , 1986; Acheson et al. , 1984; Levine
et al., 1984; Marsh, 1983). Some case-control studies have shown associations
between measures of formaldehyde exposure and nasal cancer (Olsen et al., 1984;
Hayes, 1986), whereas others have not (Brinton et al., 1985; Hernberg et al.,
1983). The lung, the buccal cavity, and the pharynx have also been examined
as sites for formaldehyde-related malignancy (Blair et al., 1986; Stayner et
al., 1986; Acheson et al., 1984; Harrington and Oakes, 1984; Levine et al. ,
1984).
Based on the results of occupationally and domestically exposed popula-
tions, formaldehyde has been reported to cause excessive respiratory symptoms,
acute and chronic reductions of lung function, and asthma. Questionnaire
surveys on symptoms have been carried out in populations selected because of
complaints about formaldehyde exposure from UFFI (Table 2-39). These surveys
show high symptom prevalences, but their results may have been biased by the
selection of subjects with complaints. A more informative and potentially
less biased design was used in a study of residents of homes insulated with
UFFI and of nonexposed controls (Thun et al. , 1982). During the year before
interview, the prevalence of wheezing and burning skin was significantly higher
for residents of homes with UFFI. Subjects who reported that odor had
persisted for longer than seven days after installation of UFFI had the highest
incidence of symptoms. Excessive respiratory symptoms have also been found in
workers exposed to formaldehyde in mobile homes used as offices (Olsen and
Dossing, 1982; Main, 1983).
Formaldehyde has been shown to be a cause of occupational asthma, although
its mechanism of action is uncertain (Hendrick and Lane, 1977; Imbus, 1985).
Studies of individuals exposed to formaldehyde in their homes have documented
complaints of wheezing, chest tightness, and other symptoms compatible with
asthma (Table 2-40). However, cases of asthma resulting from domestic exposure
to formaldehyde have not been published. In a documented case of a woman who
2-126
-------
TABLE 2-39. SURVEYS OF OCCUPANTS LIVING OR WORKING IN MOBILE HOMES
OR HOMES WITH UFFI
Study Population
Findings
Comments
424 adults, 99 children
living in 334 mobile
homes. Complaint*
investigations,
Washington State
(Breysse, 1979).
256 adults and children
living in 65 mobile
homes or 35 other struc-
tures Complaint* inves-
tigations, Wisconsin
(Dally et al., 1981).
162 residents of 68
homes with UFFI.
Complaint* investiga-
tions Connecticut
(Sardinas, 1979).
Unknown number of
residents in 443
families living in
mobile homes. Com-
plaint* investigations,
Texas (Norsted et al.,
1985).
1396 residents of UFFI
homes; 1395 residents
of non-UFFI homes.
Retrospective cohort,
New Jersey (Thun
et al., 1982).
Adults (A); Children (C)
Eye irritation: A - 58(%)
C - 41
Throat irritation: A - 66
C - 62
Chronic headache: A - 40
C - 16
Chronic cough: A - 9
C - 33
Memory lapse/
drowsiness: A - 24
C - 7
Eye irritation:
Throat irritation:
Headache:
Cough:
Difficulty
sleeping:
Wheezing:
68(%)
57
53
51
38
20
39(%)
Eye irritation:
Nose/throat/lung
irritation: 48
Headache: 17
No apparent relationship
between symptoms and crude
formaldehyde level
No difference in symptom
prevalence in families
living in homes with and
without detectable levels
Exposed more likely to report
wheezing than nonexposed:
Wheezing:
Exposed - 0.6(%)
Nonexposed - 0.1
Burning skin:
Exposed - 0.7
Nonexposed - 0.1
Subgroup, in whose homes odor
persisted >7 days after foam
installed, had higher symptom
incidence
Formaldehyde levels:
0.03 - 1.77 ppm;
No control group;
Exposure-response not
examined
Formaldehyde levels:
0.0 - 3.68 ppm;
No control group;
Exposure-response
not examined
Formaldehyde levels:
0.0-10 ug/1,
with detectable and
nondetectable levels
Formaldehyde levels:
0.0-8 ppm. Compari-
son of homes with
detectable and non-
detectable levels
Population-based study.
Formaldehyde concen-
trations not measured
2-127
(continued on following page)
-------
TABLE 2-39. (continued)
Study Population
Findings
Comments
70 exposed employees
of 7 mobile home care
centers; 34 nonexposed
employees of 3 perma-
nent structures,
Denmark (01 sen and
Dossing, 1982).
21 exposed workers in
mobile home office, 18
nonexposed workers in
another office,
Illinois (Main, 1983).
Exposed reported signifi-
cantly more symptoms than
nonexposed:
Menstrual irregularities
Exposed - 35(%)
Nonexposed - 0
Excessive thirst:
Exposed - 60
Nonexposed - 5
Eye irritation:
Exposed - 55
Nonexposed - 15
Headache:
Exposed - 80
Nonexposed - 50
Exposed reported signifi-
cantly more symptoms:
Eye irritation:
Exposed - 81(%)
Nonexposed - 17
Throat irritation:
Exposed - 57
Nonexposed -
Fatigue:
Exposed - 81
Nonexposed -
Headache:
Exposed - 76
Nonexposed -
No difference
function
Formaldehyde levels
in mobile day care
centers: 0.24 - 0.55
ppm; permanent
structures: 0.05
- 0.11 ppm
Formaldehyde levels
in offices ranged
from 0.12 - 1.6 ppm
22
22
11
in pulmonary
^Complaint investigations were initiated at residents' requests.
developed asthma after installation of UFFI, the offending agent was found to
be UFFI dust rather than formaldehyde (Frigas et al. , 1981). Frigas et al.
(1984) evaluated 13 subjects referred for evaluation of possible asthma
secondary to formaldehyde exposure in the work or home environment. None of
the 13 subjects responded to formaldehyde challenge. There is concern, how-
ever, that it may cause severe allergenic response to as much as 8 percent of
the population.
Surveys of symptoms in subjects concerned about formaldehyde exposure in
their homes have shown a high prevalence of such neuropsychological symptoms as
in headache, memory lapse, fatigue, and difficulty in sleeping (Breysse, 1979;
2-128
-------
TABLE 2-40. STUDIES OF FORMALDEHYDE EXPOSED COHORTS AND CANCER
Study
Findings
Comments
Cohort study of pathol-
ogists, Great Britain
(Harrington and
Shannon, 1975)
Proportional mortality
study of embalmers,
New York (Wairath,
1981)
Proportional mortality
study of embalmers,
California (Wairath
and Fraumeni, 1983)
Cohort study of pathol-
ogists, Great Britain
(Harrington and Oakes,
1984)
Cohort study of anat-
omists, U.S.A. (Stroup,
1984)
Cohort study of under-
takers, Canada
(Levin et al., 1984)
Proportional mortality
study of chemical plant
employees, Massachu-
setts (Marsh, 1982)
Cohort study of chemi-
cal plant employees,
U.S.A. (Marsh, 1983)
Cohort study of chemi-
cal plant employees,
Great Britain (Acheson
et al., 1984)
SMRs elevated for lymphoma
and hematopoietic neoplasms
(211) but not for leukemia
PMRs significantly elevated
for cancers of skin (221)
and colon (143); nonsignifi-
cantly for cancers of brain
(156) and kidney (150), and
leukemia (140)
PMRs significantly elevated
for cancers of colon (188),
brain (191), and prostate
(176), and leukemia (174);
nonsignificantly for bladder
cancer (138)
SMRs significantly elevated
for brain cancer (300) but
not lymphoma
SMRs elevated for brain can-
cer 271, 95% (CI = 130-499)
and leukemia (148, 95%
CI = 71-272)
SMRs nonsignificantly ele-
vated for brain cancer (115)
and leukemia (160)
PMR nonsignificantly ele-
vated for cancers of diges-
tive organs (152) among for-
maldehyde exposed workers.
No data reported on brain
cancer and leukemia
SMR significantly elevated
for cancers of genitouri-
nary tract (169). SMR for
leukemia not elevated. No
data for brain cancer
SMRs for lung cancer signi-
ficantly elevated (124) in
one of six men most highly
exposed
Less than 10% of cohort
deceased. Less than 20
yrs of follow-up
Less than 5% of cohort
deceased. Six yrs of
follow-up
Excess brain cancer
persisted when psychia-
trists used as a refer-
ence group
20 yrs of follow-up
No evidence of trend of
mortality in relation
to exposure
Case-control study with-
in cohort showed no asso-
ciation between GU can-
cer and a general plant
exposure
Retrospective assess-
ment made of level of
exposure
SMR = Standard Mortality Ratio
PMR = Proportional Mortality Ratio
CI = Cancer Incidence
GU = Genitourinary
2-129
-------
Dally et al., 1981; Sardinas et al., 1979). Formaldehyde exposures in these
uncontrolled studies ranged up to about 4 ppm. Schenker et al. (1982) used
standardized respiratory and neuropsychological tests to evaluate 24 residents
of homes insulated with UFFI. Complaints of memory loss were not validated by
formal tests but 11 of the 14 subjects who were tested had a deficit of atten-
tion, and 9 of the 11 had elevated depression scores. Two controlled studies
of the neuropsychological effects of formaldehyde exposure in the occupational
setting also suggest similar effects (01 sen, 1982; Kilburn, 1985).
2.7 RADON
2.7.1 Occurrence and Sources of Radon
Radon-222 (hereafter called radon) is a noble gas with an atomic weight
of 222, which makes it the heaviest known gas. It cannot be detected by the
human senses and it is relatively soluble in water at room temperature. It is
radioactive, emitting alpha particles, with a half-life of 3.8 days (Evans,
1969). Radon gas is a decay product of radium-226, which is itself a decay
product of the uranium-238 series. The first four radon decay products (RDP),
polonium-218, lead-214, bismuth-214, and polonium-214, are relatively short-
lived, each having a half-life of less than 30 minutes. Radon, polonium-218,
and polonium-214 are alpha emitters, whereas lead-214 and bismuth-214 are beta
and gamma emitters. Polonium-214 decays to lead-210 with a half-life of 22
years, which effectively terminates the series as far as lung deposition is
concerned. This decay series actually ends with the stable, nonradioactive
element lead-206.
Trace amounts of naturally radioactive nuclides are present in soil, rock,
and building materials, as well as in living organisms, including human beings.
Radon and its short-lived decay products in indoor air are the greatest contri-
butors to the total dose burden from natural radioactivity. In particular, it
is the alpha-emitting polonium-218 and polonium-214 decay products or radon
daughters that are of most concern (National Council on Radiation Protection
and Measurements, 1984a,b).
Because of the relatively short half-life of the alpha-emitting radon
daughters (<30 min, in aggregate), significant concentrations cannot be main-
tained without a radon source. The daughters will reach equilibrium with radon
rapidly when no separative or removal processes are active. This permits radon
2-130
-------
concentration and its variability to serve as an indicator of possible radia-
tion dose to the lung by short-lived daughters. For dose calculation however,
the alpha particles emitted by both radon and its progeny at equilibrium must
be considered.
Overall estimates of global emanation of radon identify soil as the
greatest contributor of radon, followed by ground waters and the oceans.
Atmospheric concentrations of radon are estimated to average between 100 to 200
3 3
pCi/m (3.7 to 7.4 Bq/m ), with concentrations over uranium ore-grade soil
typically between 500 to 1000 pCi/m3 (18.5 to 37 Bq/m3).
Because uranium-238 and, consequently, radium-226 exist to some extent in
most soils and rocks, radon is also widely distributed in these materials.
Because radon is a gas, it has the potential to diffuse through pores and
cracks in soil and rock to mix with other soil gases and to escape into the
atmosphere. The short-lived RDPs are very small particles which agglomerate
rapidly and readily attach to surfaces, including suspended particles or water
droplets in air. Radon may enter a dwelling directly with the soil gas, in
solution with the water, or by diffusing from construction materials.
Nazaroff (1984) suggests that the infiltration of soil gas directly into
single-family homes is the largest contributor to indoor radon levels. Radon
formed from radium in the soil or rock diffuses into the interstitial spaces
and migrates to the surface with the soil gas. Soil with coarse grain size,
gravel, and sand are highly permeable and allow more migration than muds and
clays which have finer grain sizes and higher moisture contents. Radon reach-
ing the soil surface is diluted in the air. However, it may reach a surface
that is in contact with a structure such as a basement wall or floor or a
slab-on-grade foundation. If pathways such as cracks and holes are present
in the structure, and the air pressure is lower in the structure than in the
soil, radon flows into the structure. Higher than normal concentrations of
radon in the soil or higher than normal permeability of the soil surrounding a
dwelling can contribute to increased rates of radon entry. The combination of
soil permeability and pressure difference between the soil gas and the air
inside the structure largely determine the effective volume of radon that
enters the structure.
Indoor levels of radon have been measured to range from the lower limit of
detection 500 pCi/m3 to 2 x 106 pCi/m3 (ca. 18 to 7 x 104 Bq/m3). This
variability of three orders of magnitude of the radon concentration indoors
2-131
-------
should be considered the working range for commonly needed indoor radon
measurements, although a variability of less than one order of magnitude is
expected on a house-specific basis. Radon daughters in equilibrium with these
levels of radon should vary in the same manner. Equilibrium ratios of the
activity concentration of radon to radon progeny in indoor environments vary
from near 1.0 to 0.3 at higher indoor ventilation rates.
While there is consensus in the radon/radon progeny research area that
radon progeny in indoor environments pose one of the greatest environmental
risks, there is scientific debate as to what indoor conditions and ambient air
characteristics are most closely related to the observed health effects. This
led the National Council on Radiation Protection and Measurements to recom-
mend "that future (radon and radon progeny) measurements should provide the
aerosol data to allow calculation of the specific bronchial dose as well as
documenting exposure . . ."(National Council on Radiation Protection and
Measurements, 1984a,b). In particular, this was a call for determination of
the aerosol size and number to which radon progeny would attach themselves, as
well as a determination of the partitioning between attached and unattached
radon progeny under various aerosol conditions.
Nero (1985) notes that homes served by private wells in areas with high
soil radium content have a high probability of increased radon levels in the
tap water. Water with elevated radium levels has correspondingly high radon
levels, groundwater being more likely to have higher levels than surface water.
Radon can also be dissolved directly in the water as it moves through the soil.
Approximately 50 percent of U.S. homes are served by private wells, lakes, and
streams (Smolen, 1984; U.S. Environmental Protection Agency, 1985b). The radon
is released from the water at elevated temperatures, and when the water is
highly agitated, or in finely divided droplets with large aggregate surface
area, such as occurs in a shower, a dishwasher, or a washing machine.
Certain technological activities are known to cause elevated levels of
indoor radon in some locations. Phosphate industry waste materials in Florida
and uranium mill tailings in Colorado, both of which contain elevated concen-
trations of radium-226, have resulted in elevated indoor radon levels when
these products were placed under and around structures. Waste from a radium
processing facility in New Jersey was also observed to contribute to elevated
radon levels in some structures.
2-132
-------
Radon is also released from many building materials, but normally at very
low levels. Wood materials tend to emanate the least radon, while brick,
cement, and cinder block emanate more. Radon is released from all of these
sources at such a low rate that rarely are these materials an important con-
tributor to elevated radon levels. However, there have been a few cases in
which materials containing significant radium concentrations were used to form
building materials. Examples of these situations are houses that have been
built with materials contaminated with uranium or radium mill tailings and
uraniferous phosphogypsum waste utilized as drywall. Another example is homes
that use radium-containing heat storage rocks (e.g., large pieces of granite)
and circulate large volumes of air into the living areas.
2.7.2 Indoor Concentrations and Exposures
The detection of radon and RDP in air depends upon the interactions of
alpha and gamma radiation from the decay products with some detection media.
For example, alpha particles may interact with zinc sulfide to produce light
pulses, which are counted using a photomultiplier tube assembly. Alternatively,
alpha particles interact within a polycarbonate plastic chip to produce ioniza-
tion tracks, which are enlarged through etching and then visually counted.
Gamma radiation from decay products collected on activated carbon canisters may
interact with sodium iodide crystals to produce light pulses, which are
counted. Other interactions and detection schemes exist and are used in a
variety of instruments and methods depending on the levels to be measured, the
accuracy required, the equipment available, and the convenience of the
operator.
Instrumentation is available to analyze indoor air samples for radon and
radon progeny (with acceptable accuracy and precision). EPA guidance has been
published for the use of continuous monitors (scintillation cell), 3- to 5-day
integrating charcoal canisters, alpha-track detectors, and grab sample scintil-
lation cells (U.S. Environmental Protection Agency, 1986). This instrumenta-
tion has been and is being employed in various local and regional air quality
monitoring and EPA radon mitigation studies. Because of the private sector
involvement in radon mitigation activities and the need to verify reductions in
indoor radon levels, the EPA Office of Radiation Programs (ORP) has maintained
a measurement proficiency program to ensure the quality of both private and
public sector measurement programs.
2-133
-------
A passive radon badge (Terradex Track-Etch) is also available and has
suitable accuracy for monitoring most indoor microenvironments. However,
exposure times of several months are usually required for measurable change to
occur.
Radioactivity is measured in terms of the rate of decay of a particular
unstable nuclide. For instance, the Curie (Ci) is defined to represent 3.7 x
10 disintegrations per second. Radon concentrations are usually measured in
—1 ?
pCi/1 (picoCuries per liter), where a picoCurie is 10 Curies. For occupa-
tional historical reasons, the concentration of RDP is generally expressed in
working levels (WL), where one WL is any combination of RDP in one liter of
air that ultimately releases 1.3 x 10 MeV (million electron volts) of alpha
energy during decay (Holaday et al., 1957). A concentration of 1 pCi/1 of
radon in dynamic equilibrium with its decay products, translates into about
0.005 WL. Since exposure depends upon both concentration and time, the occupa-
tional historical radon exposure unit has been the Working Level Month (WLM).
The WLM was developed to describe exposure sustained during the average number
of hours spent underground by miners, where 1 WLM represents the exposure to
1 WL for 170 hours. The relationship between exposure, measured in WLM, and
dose to the tissues of the lungs, measured in rems (the traditional biological
units) is quite complex and will not be elaborated upon here.
Nearly all structures have some indoor radon which is expected to be at
least as great as that found outdoors; however, concentrations above these
normal levels arise when the above-mentioned sources are present. Naturally
occurring radon in the outdoor air ranges from 0.1 to I pCi/1, where 1 pCi/1
_Q O
corresponds to 6.5 x 10 |jg/m of radon gas. Note that this concentration is
10 orders-of-magnitude smaller than the proposed new ambient air quality
standards for particulates (PM-10 standard). This example serves to illustrate
the extremely small concentrations of radon which are environmentally signifi-
cant. For instance, EPA recommends remedial action for indoor concentrations
_Q 0
of radon which exceed 4 pCi/1 (2.6 x 10 pg/m ).
A few areas in the United States are known to have geological factors
that contribute to elevated indoor radon concentrations. These factors include
above-normal concentrations of radium in the soil and the presence of porous
soil or fractured rock formations affording ready migration of radon into
structures. This combination of conditions exists in the Reading Prong, a
geological area located in parts of Pennsylvania, New Jersey, and New York,
2-134
-------
where extremely high indoor radon levels have been observed. Indoor radon
levels 1000 times the recommended action level have been observed in a few
instances. In these cases, the observed concentrations of radon in the soil
gas were very high.
Most deep water supplies in the United States have radon concentrations of
less than 2000 pCi/1. Some areas, however, such as Maine, have levels exceed-
ing 100,000 pCi/1 (Walsh et al., 1984). On the average, it has been estimated
that drinking water contributes only 2.5 percent to the background radiation
dose (Cothern, 1986).
The average amount of radon in public water supplies is less than 500
pCi/1 and does not pose usually very great risks. Whether or not radon and
its progeny leave the water depends on the temperature and surface area of the
water. If the water is near boiling, in finely divided droplets, and agitated
as in a dishwasher, up to 90 percent of the radon contained in the water can
enter the air. Even in showers, a considerable portion may leave the water,
attach to water vapor, and be inhaled.
There is a need to determine indoor activity concentrations of radon and
radon progeny over a wide range of concentrations for multiple purposes; for
example, for health risks assessments, determining the location and variability
of the radon sources, and determining the effectiveness of indoor radon control
techniques. For this purpose field-hardy instrumentation is needed to deter-
mine aerosol concentration and size distributions and the partition of attached
and unattached radon progeny under various aerosol conditions resulting from
ventilation or the operation of air cleaning devices. A small passive device
for detecting radon progeny is needed, as well as instruments for calibration
of radon/radon progeny measuring devices.
2.7.3 Health Effects Associated with Radon Exposure
The only documented health effect associated with exposure to high concen-
trations of radon is the increased risk of lung cancer. Alpha-emission from
inhaled RDPs in the respiratory tract is thought to produce the tissue injury
that eventually results in malignancy. The alpha particles are presumed to
penetrate into the airway epithelium and damage the genetic material of the
basal cells. The RDPs also release low-energy gamma and beta particles during
decay, but the damage to the lung is almost exclusively due to the alpha parti-
cles released by polonium-218 and polonium-214 (National Council on Radiation
Protection and Measurements, 1984a,b).
2-135
-------
The primary concern when discussing the risks from exposure to radon is
not the exposure to the radon gas itself, but exposure to its decay products,
namely the two alpha-emitting isotopes of polonium. Under normal circumstances,
the RDP rapidly attach to airborne dust particles. The RDP that do not attach
to aerosol particles agglomerate readily to form very small particles in the
size range from about 0.002 p to perhaps as large as 0.02 p. These agglomerates
are referred to as the unattached RDP or unattached radon progeny. The size
of the particles to which the RDP is attached is very important in determining
the location in the respiratory system that will receive the radiation dose.
Particles readily stick the the moist epithelial lining of the bronchi. Most
dust particles are eventually cleared from the bronchi by mucous, but not
quickly enough to keep the bronchial epithelium from being exposed to alpha
radiation from the decay of polonium-218 and polonium-214. The alpha particles
that penetrate the epithelium cells can deposit sufficient energy in the cell
to kill or transform it. The transformed cell has the potential to induce a
lung cancer. Since lung dosimetry models have indicated that the calculated
dose to the lungs from unattached decay products may be greater than from
attached decay products, there may be a significant health risk distinction
between these two cases (James, 1984; Sextro et al., 1984). The health effects
associated with the inhalation of RDPs depend on where in the lung the particle
is deposited. In the upper portion of the respiratory system, particles are
continuously cleared by the ciliated mucous lining. However, particles depo-
sited deep in the unciliated bronchiolar or alveolar regions of the lung have
long residence times and, consequently, may have greater adverse health
effects. The larger particles have a higher probability of depositing in the
upper regions of the lung, whereas only the small particles, such as the
unattached decay products, have a high probability of being deposited deep in
the lungs. Therefore, the mathematical models predict a higher lung dose
associated with unattached RDPs. One implication of this conclusion is that
the use of an air cleaner to remove particles from the air may increase the
fraction of unattached decay products and, consequently, increase the radiation
dose to the lungs. The results of these predictions have not yet been
confirmed by measurements, therefore, the effects of using air cleaners to
reduce the health risks from RDP are not established.
2-136
-------
2.7.4 Risk Estimates
The lung cancer risk associated with exposure to RDP was first recognized
in underground miners. While the lung cancer risk incurred by underground
miners has been recognized for a century, the hazards posed by environmental
RDPs have only recently become a public health concern. Numerous studies of
uranium miners and other underground miners have established a causal relation-
ship between exposure to RDP and lung cancer (Lundin et al., 1971; National
Research Council, 1980). Animal studies confirmed that exposure to RDP alone
causes lung cancer (National Council on Radiation Protection and Measurements,
1984a). The human data come primarily from miners having high exposures,
and the risks of lower exposure levels have not yet been well characterized.
Risks estimates associated with the low exposures typical of low indoor radon
concentrations were obtained by extrapolations and, consequently, are somewhat
uncertain. These uncertainties may soon be removed through the analysis of
more recent data on groups of miners whose exposure levels lie within the range
of typical residential exposures (Muller et al., 1983; National Institute of
Occupational Safety and Health, 1985; Solli et al., 1985; Howe et. al., 1986).
It is also true that data is currently being collected in which residential
exposures were much higher than any from underground mines. These data should
provide good cross checks for risk calculations.
The U.S. Environmental Protection Agency (1985a) estimated that the aggre-
gate health effects over all radon exposure levels range from about 5000 to
20,000 lung cancer deaths per year. The dose to the bronchial epithelium of
the lung is thought to come primarily from the unattached radon daughters
(Shapiro, 1956; National Council on Radiation Protection and Measurements,
1984a,b).
2.7.5 Estimate of Dosage to People Exposed to Radon
Estimating a person's incurred dosage of radiation produced by a measured
concentration of radon in a residence is difficult. Measuring the physical
presence of radon in a volume of air is relatively simple; its concentration
is usually reported in terms of associated disintegrations per second.
Quantifying the amount and significance of the radiation that actually reaches
various body tissues is, however, dependent on several parameters that vary
and interact in complex ways. The link between a given air concentration of
2-137
-------
radon and the consequent radiation dosage involves the following principal
factors, about which there are varying opinions and degrees of consensus:
ratio of indoor radon levels (primarily from soil seepage and
emanations from building materials) to outdoor, background
levels;
presence of radioactive precursors (most important in mining
and industrial settings);
presence of 'radon daughters', the products of further radio-
active decay, and their contributing effects;
proportion of time spend in the indoor environment;
breathing rate (activity level) which determines the amount of
contaminated air brought into direct contact with susceptible
lung tissue cells;
retention rate in the lung of the 'radon daughters'.
W. F. Bale (1980) equates a radon concentration of 1 x 10 Curies per
liter (Ci/1) with an accumulated dose of 22.6 rem (roentgen equivalent man)
for a 40-hr work week. Expressed per 100 picoCuries, this is:
100 pCi/1 = 22.6/40 = 0.565 rem/hr.
Another source, Snihs (1985), equates a radon concentration of 200 Bequerel/m3
2
(Bq/m ) with a yearly dose of 20 milliSieverts (mSv). From Burkart (1986),
o
200 Bq/m can be equated with 5.4 pCi/1 and 20 mSv/yr with 200 mrem/yr. Thus,
100 pCi/1 = 3704 mrem/yr. Assuming 75 percent occupancy (6570 hrs/yr), this
becomes:
100 pCi/1 = 0.564 mrem/hr.
For purposes of estimating dose here, a radon concentration of 100 pCi/1 will
be equated with a dose of 0.5 mrem/hr.
2.7.6 Indoor Air Quality Control Options
Basically, the two options available for controlling the concentration
levels for indoor radon are: 1) removing it once it has entered the structure,
2-138
-------
or 2) preventing it from entering the structure. The removal method that has
been demonstrated to be most effective in reducing indoor radon levels is
forced ventilation. Natural ventilation consists of the exchange of indoor and
outdoor air in response to natural driving forces. The major sources of the
natural driving forces are winds, and temperature and pressure differences
between indoors and outdoors (Perdue et al., 1980). With windows and doors
closed, typical American homes are characterized by a natural annual average
ventilation rate of about 1 ach (air changes per hour). Energy efficient homes
may have exchange rates as low as 0.1 ach whereas older homes may have rates as
high as 2 ach. Ventilation is effective in reducing the concentration of
indoor radon by replacing a certain volume of contaminated indoor air with an
equal volume of relatively uncontaminated outdoor air which then proceeds to
further dilute the remainder of the indoor air. By simple dilution, increasing
ventilation rates over the range from 0.25 to 2.0 ach can yield up to 90
percent reduction in indoor radon concentrations. Even at very high exchange
rates, ventilation will not reduce radon levels below some specific value which
depends on the radon entry rate. Since the absolute effectiveness of ventila-
tion to reduce radon levels decreases with increasing ventilation rate, this
method of reduction will be most cost-effective for tight (low air exchange
rate) homes.
A more cost-effective method for removing indoor radon uses forced venti-
lation with heat recovery. Heat recovery is usually accomplished through an
air-to-air heat exchanger. These are low-pressure drop devices that exchange
heat from the warmer air being exhausted to the incoming cooler air (or vice
versa if an air conditioner is in operation). These devices are sometimes
equipped with auxiliary heaters to compensate for the remainder of the heat
loss. Air-to-air heat exchangers operate with nominally balanced supply and
exhaust flows so the pressure in the building is unchanged. Consequently, both
the natural ventilation and soil gas flows are unchanged, which means the radon
concentration in the building is affected only by the ventilation rate. At
best, radon levels can be reduced by a factor of two or three using forced ven-
tilation with heat recovery (Nazaroff et al., 1981). If the ventilation system
were operated in a manner which leads to a depressurization of the house, the
radon level might actually be increased.
2-139
-------
Although there are several methods for reducing the concentrations of in-
door radon and its decay products, most investigators agree that the most prac-
tical and desirable approach is to prevent its entry in the first place (Nero,
1985; U.S. Environmental Protection Agency, 1986c). Four general approaches
that will prevent radon entry from the common sources are: 1) sealing entry
points, 2) ensuring that the direction of airflow is from the house into the
soil, 3) removing radon from the water supply, and 4) avoiding the use of
building materials that contain significant quantities of radium.
In most instances, the major source of indoor radon is soil gas entering
the house. Consequently, a major first step in preventing the entry of radon
is to seal the major entry points such as exposed earth in basement floors or
drainage sumps. Several studies have shown that sealing all visible cracks and
gaps between floor, walls, and service pipes can significantly reduce radon
concentrations in houses with radon levels in the range of 30 to 70 pCi per
liter (New York State Energy Research and Development Authority, 1985; Holub et
al., 1985). A word of caution concerning sealants is in order. Because radon
entry is usually pressure gradient driven, very small cracks or openings can
provide effective pathways. Therefore, very slight movements of the house
substructure may reopen sealed pathways. As a result, sealing alone does not
provide a great deal of confidence as a long-term solution to radon entry.
However, sealing at least the major openings is almost a necessity for most
other mitigation procedures, which are effective when pressure gradients are
the primary driving forces. Without sealing, for instance, too great an
airflow may be required to reverse the pressure gradient from the soil to the
inside.
Ensuring that the airflow is in the direction from the house into the
soil can be accomplished through a number of methods such as 1) sub-slab
ventilation, 2) block wall ventilation, 3) baseboard ventilation, 4) drain tile
ventilation, or 5) basement pressurization. Active sub-slab ventilation
consists of using a fan to sweep the soil gas out of the aggregate under the
slab in a basement or a slab-on-grade before it can enter the house through
cracks or other pathways. In a typical implementation, one or more pipes
penetrate through the slab into the aggregate and a fan withdraws air at a rate
sufficient to reduce the local pressure to a value lower than that inside the
house. Under these conditions, air flows from the house into the aggregate,
thus preventing the entry of radon-bearing soil gas. If there are major
2-140
-------
unsealed openings in the floor, the energy penalty for heat losses will be
prohibitive. This method will be less effective if the permeability of the
aggregate or soil under the slab is low.
The centers of concrete blocks used to construct many basement walls con-
tain voids, which are generally interconnected both vertically and horizontally
within the wall, forming a network of channels through which air can move.
Soil gas that penetrates the outer surface of the blocks through mortar joint
cracks or pores is able to move throughout the wall to find cracks or openings
to enter the basement. In the active ventilation of hollow-block basement
walls, a fan is used to sweep the soil gas out of the voids in the wall and
reduce the local pressure below that inside the basement so that the direction
of airflow is from the basement into the wall. In this manner, radon is pre-
vented from entering the basement. The typical implementation of this method
is to install one or two pipes in each wall penetrating into a block cavity.
However, there is another approach to implementing this method known as base-
board ventilation. In this case, a sheet metal "baseboard" is installed around
the entire perimeter of the basement (including interior block walls), and
covers the joint between the floor and wall. Holes are drilled through the
interior face of the blocks at fixed intervals inside this baseboard, and the
wall is ventilated by depressurizing the baseboard duct with a fan. This ap-
proach produces more uniform ventilation of the wall but is more expensive than
the method that uses individual pipes. All major openings must be sealed for
this method to be effective.
An alternative method for ventilation under the slab is called drain-tile
soil ventilation. Perforated drain tiles surround part or all of some houses
in the vicinity of the footing to drain moisture away from the foundation. The
water is usually routed to an above-grade soakaway or to a sump in the base-
ment. In many houses, the major radon entry points are near the floor-wall
joint so that pumping on the drain tile near the foundation is often effective
in removing the soil gas containing the radon. In many cases, the pressure
near the foundation can be reduced sufficiently so that the airflow is reversed
to move from the basement into the soil. When the aggregate under the slab is
sufficiently permeable, the soil gas can be extracted from underneath the
entire slab.
Still another method that ensures that the air flows from the basement
into the soil uses basement pressurization. This pressurization can be
2-141
-------
accomplished by either blowing outside air into the basement or by blowing air
from some other part of the house into the basement. This choice depends to
some extent upon the outside temperature extremes and the amount of time the
residents spend in the basement. This method would only be practical for
houses in which the basement is well isolated from the remainder of the house
and when combustion appliances are not adversely affected by the pressuriza-
tion. Unless the basement is very tight, a severe energy penalty for heat loss
is likely to occur.
The previously described techniques of ventilation were designed primarily
for houses with basements or slabs-on-grade. In fact, it is thought that these
types of construction will present the greatest number of problems because they
represent cases in which the house is in intimate contact with the soil,
providing an excellent opportunity for pockets of radon-bearing soil gas to
collect and migrate into the house. However, in the case of crawl spaces that
are not well ventilated, a similar situation could develop. Since the house
tends to be depressurized (especially during winter) radon bearing soil gas in
the crawl space will migrate into the house under the action of the pressure
gradient. One approach to prevent radon entry is to reverse the pressure
gradient through depressurization of the crawl space. An alternative approach
is to pressurize the crawl space, thus preventing the radon from entering the
crawl space and isolating the house from communication with the soil. In
moderate climates, the most practical approach seems to be to adequately
ventilate the crawl space (either actively or passively) by flushing or
diluting the crawl space air with outside air.
Aside from switching to alternative water supplies or reconstructing the
wells, the most effective methods for reducing radon levels in water supplies
seem to involve aeration or adsorption by granular activated carbon. Activated
carbon is quite effective in reducing the radon level, but has two potentially
significant drawbacks. First, the granular beds show a tendency to develop
bacterial growths which may eventually cause health concerns. Second, the
granular bed adsorbs radioactive components (uranium, radium, radon, and RDPs
from the water and, consequently, becomes radioactive. One result is increased
gamma radiation in the vicinity. It presently is not clear whether this
increased gamma radiation creates a significant danger, but the matter should
be studied. Another result is that the accumulation of uranium and radium in
the bed may cause the material to be classified as a low-level radioactive
2-142
-------
waste and, consequently, create a serious disposal problem. Aeration is
another effective means of removing radon from water. The most common method
involves breaking the water into small droplets as a spray. The radon readily
diffuses from the liquid into air.
2.8 BIOLOGICAL CONTAMINANTS
2.8.1 Introduction
Indoor environments are contaminated by a variety of biological pollu-
tants. Among these are molds and their spores and toxins, bacteria, viruses,
protozoans, algae, body parts and excreta of insects, acarids, and arachnids
dander and excreta from animals, and pollens from higher plants.
Airborne molds most commonly found in indoor environments include the
following:
Aspergillus Aureobasidium
PeniciIlium Sporobolomyces
Fusarium Wallemia
Cladosporium Yeasts
Thermophilic Actinomycetes
Aspergillus, PeniciIlium, Fusarium, and Cladosporium are known toxigenie
taxa (Burge, 1986). Among the bacteria that have been identified indoors are
Legionella pneumophila, Clostridium perfringens, Staphyllococcus aureus and S.
dermatitis, Streptococcus pyogenes, and Salmonella typhimurium. Mycobacterium
species have been found in the dust of homes of infected persons. Occasionally
Pseudomonas species and Actinobacter species have been found. A number of
other bacteria associated specifically with factories processing organic
materials from plants or animals have been isolated in those environments
(Solomon and Burge, 1984; Imperato, 1981; Sugarawa and Yoshizawa, 1984;
Lundholm and Laurell, 1984). Among those viruses transmitted by the airborne
route are smallpox (no longer considered to be present as an infectious agent
outside of tightly controlled laboratories), chickenpox, measles, rubella,
influenza, adenovirus 4 and 7, coxsackie A21, and lymptocytic choriomeningitis
viruses (Couch, 1981).
2-143
-------
Protozoa, especially amoeba of the genera Acanthamoeba. Naegleria
gruberi, and N. fowleri, have been identified indoors, on occasion. Algae,
especially algal spores, have been identified in dust.
Microscopic body parts and excreta of insects and acarids tend to be found
in house dust. Dust mites, Dermatophagoides spp., themselves nearly of micro-
scopic size, produce fecal pellets, which disintegrate to form particles in the
\
respirable size range (0.8 to 1.4 u), which can be highly allergenic in sensi-
tized individuals (Andersen and Korsgaard, 1986). Dried excreta of insects
such as cockroaches can also be entrained in dust and be a source of allergy.
Larger animals such as dogs, cats, birds kept as pets, as well as rodents, can
produce a number of body secretions as well as skin flakes, that serve as
allergenic agents.
Pollens from higher plants enter indoor spaces from outside. Pollens from
indoor plants are not generally considered a problem.
2.8.2 Sources of Biological Contaminants
Sources of biological contaminants vary greatly. Many have their origin
out of doors and enter indoor spaces through windows, doors, and cracks, or
through ventilation, and air conditioning or humidifying systems. Animal
dander and excreta, especially from birds such as pigeons roosting near air
inlets of buildings, can also enter the indoors. Among contaminants origi-
nating outside are pollens. These generally are windborne pollens from small,
rather drab, fragrance!ess plants, unlike houseplants grown for showy or
fragrant flowers. Their concentrations vary seasonally, and indoors addi-
tionally with indoor/outdoor pressure differentials, amount and type of
ventilation, wind conditions, temperature, and humidity (Burge, 1985). Algae
and amoeba also enter from outside as do fungi, in similar fashion. These
latter are viable, however, and can colonize suitable indoor environments,
increasing tremendously in number once established in a friendly space indoors.
House dust mites, which feed on skin flakes shed by humans and animals, can
also multiply under ideal conditions and hence increase their output of
allergenic products. Ideal conditions for increased growth among these
organisms are moisture, humid air, and warm temperatures (Burge, 1985). Any
organic material can serve as a medium for fungal growth. Moist surfaces and
water reservoirs for air conditioning and humidifying systems for homes and
buildings can serve as ideal reservoirs, amphifers and disseminators for fungi,
2-144
-------
bacteria, algae and amoeba. Office and public buildings that use chilled water
air conditioners/humidifiers have been identified as one building type that is
subject to such contamination. Legionella ssp. and Clostridium ssp. are soil
bacteria that have been found in the reservoirs of such systems. In addition
to such a water source for bacteria, humans expel saprophytic and pathogenic
bacteria by sneezing, coughing, and speaking, and can introduce such organisms
into indoor spaces. These bacteria may survive for significant periods of time
depending on droplet size, temperature, and relative humidity. Such bacteria,
as well as fungal spores and amoeba droplet nuclei, and sometimes viruses, have
been distributed throughout buildings by ventilating systems, stairway and
elevator shafts through stack effects, and even through building raceways and
plenums. In homes, many of these biological contaminants can likewise be
found, and home air conditioners have sometimes been identified as the source
of the organisms. Growth indoors is promoted by moisture and organic food
sources, so that upholstery, bedding, carpets, plumbing, and food preparation
areas tend to be sources of continuing contamination.
The two most common fungi appearing indoors are Penicillium and Aspergil-
lus (Benson et al., 1972). Although measured concentrations of fungus spores
are generally less indoors than outdoors, under poor hygienic conditions and/or
conditions of high humidity, fungal spore concentrations indoors have been
measured which exceed outdoor concentrations by 200 to 400 percent (Benson et
al., 1972; Nilsby, 1949; Flensborg and Samsoe-Jensen, 1950; Jimenez-Diaz et
al., 1960; Solomon, 1975). Outside sources of fungi, as well as of bacteria,
are humidifiers and cooling towers associated with air conditioning systems
(Miller et al., 1976; Taylor et al., 1978; National Research Council, 1981b).
Indoor concentrations of pollen, known to cause allergic reactions in
sensitive individuals, are driven by outdoor concentrations. The pollen enters
through building cracks and crevices, doors, windows, and the fresh-air intake
of air conditioning systems.
Epidemiological studies have established that many infectious diseases are
communicated by airborne transmission (Repace, 1982). The well-known Legion-
naire's disease is a good example of how outside contamination can affect the
occupants in a building. The disease-causing organism, Legionella pneumophilia,
is unusual among pathogens in that it apparently exists in outdoor natural
reservoirs (soils), and infection is possible through inhalation of contaminated
outdoor air (National Research Council, 1981b).
2-145
-------
2.8.3 Monitoring of Biological Contaminants
Sampling and measurement of biologic pollutants present special compli-
cations not found with less complex chemical agents. Devising universally
suitable means of determining the presence and concentration of biological
contaminants remains a problem.
In the United States, the most common method for sampling biological
particles remains gravity collection, either on culture plates or sticky
slides. This method is never volumetric, produces a qualitatively biased
picture of the air spora, and should never be used for assessment of airborne
particle levels. It is useful only where gravity deposition of particles is
of concern (such as in operating rooms).
Volumetric sampling (where a measured quantity of air is collected and
analyzed) includes two general types: 1) viable or cultural methods where
recovery depends on the viability of collected particles as well as their
ability to grow under given conditions, and 2) particulate methods where
particles are visually counted and identified or biochemically or immunologi-
cally analyzed.
Cultural methods are useful when 1) information on viability is essential
(e.g., for infectious agents); and 2) particles must be cultured to be identi-
fied (e.g., actinomycetes, bacteria, viruses, many small-spored fungi). Choice
of culture media for these samplers is critical. Cultural methods always
underestimate actual spore counts, and the underestimate increases logrithmi-
cally with particle levels (Burge et al., 1977).
Particulate sampling is the method of choice when total biological
particle counts are to be assessed or when biological products (toxins, anti-
gens) are to be measured. Total fungal spore counts can be simply done using
microscopic counting. New methods using fluorescent staining may enable
counting of bacterial particles as well (Palmgran et al., 1986). Immunological
or biochemical analysis of particulate samples is especially useful when known
contaminants are to be analyzed. These methods have been used for airborne
endotoxin (Rylander and Haglind, 1984), mycotoxin (Burg et al., 1981), and a
variety of antigens (e.g., mites (Swanson et al., 1985), cockroaches (Swanson
et al., 1985), and thermophilic actinomycetes (Reed et al., 1983)). Sampling
for particles to be analyzed biochemically or immunologically requires advance
ledge of the compounds of interest. These methods are not useful for surveys
of general bioaerosol contamination.
2-146
-------
Volumetric collection modalities fall into three general categories:
impactors, impingers, and sieve samplers (Burge and Solomon, 1987). Impactors
collect by causing particles to leave the air stream and impact on an adhesive
surface either by rapidly rotating the sampling surface or by accelerating the
air. Commonly used impaction samplers are listed in Table 2-41. The rotating
and centrifugal impactors are efficient only for particles larger than 15 to 20
urn. Remaining samplers are efficient over a wide range of particle diameters.
TABLE 2-41. IMPACTION SAMPLERS USEFUL FOR BIOAEROSOL SAMPLING
Sample Type Viable Particulate
Rotating impactors Rotoroda, Rotoslide
"Centrifugal" samplers RCSb, Wells
Single plate impactors Andarsen-N6c,
SASU, Microban6
Cascade impactors
Slit samplers
Andersen
Slit sampler
Andersen
Burkard^
aTed Brown Assoc., Palo Alto, CA
Biotest Diagnostics, Fairfield, NJ
cAndersen Samplers Inc., Atlanta, GA
Spiral Air Systems, Bethesda, MD
eRoss Industries, Midland, VA
New Brunswick Scientific Co., New Brunswick, NJ
9Burkard Manufacturing Co., Rickmansworth, England
Impingers trap particles in a liquid after impingement on a submerged
surface. The resulting collection fluid can be analyzed immunologically or
biochemically. Liquid impingers are especially useful for sampling for viable
particles in highly concentrated aerosol situations because the fluid can be
dilution cultured. The use of these devices for fungi, toxins, and antigens
has not been well studied.
Sieve sampling utilizes filters of known pore sizes either in portable
filter cassettes or in larger high-volume sampling devices. Filters collect
particles to diameters well below rated pore sizes very efficiently. Samples
can be analyzed by direct counts on the filter, or by cultural, biochemical, or
immunological assay of filter eluates.
2-147
-------
At present, no one sampler is adequate for all bioaerosols. A combina-
tion of a cultural sampler and a particulate collector will cover most situa-
tions. Both filtration and impinger sampling have the potential of being
universal samplers, but much research remains to be done to verify their use
for all bioaerosol types.
2.8.4 Health Effects of Biological Contaminants
2.8.4.1 Infection. Health effects of biological contaminants can be described
primarily as three kinds: pathogenic, toxicogenic, and allergenic. Some
organisms are clearly infective, and their entry into human respiratory systems
presents the possibility of disease. The syndrome for each disease entity is
usually well defined and often leads to identification of the infective agent.
Viral diseases known to be spread by the airborne route, such as measles,
rubella, chickenpox (varicella), are clearly identified. Respiratory diseases
such as colds, influenza, and sore throats may be less clearly defined since
some symptoms are similar to allergic responses. Viruses are different from
bacteria and other infective agents because they cannot replicate outside of
their host and they are highly species-specific.
A fairly common soil bacterium, which grows well in organically enriched
water, and which is disseminated in airborne droplet nuclei, is Legionella spp.
It is the cause of a serious and sometimes fatal pneumonia known as Legionaires
disease. Epidemiologic studies have shown that Legionella causes 10 to 15
percent of pneumonia cases in communities and in hospital situations (Johnson-
Lussenburg, 1986). More commonly it causes a milder, flu-like syndrome termed
Pontiac fever, that disappears without medical treatment.
Other waterborne bacteria such as Acinobacter and Pseudomonas aeruginosa
can become airborne and be disseminated through heating, ventilation, and air
conditioning (HVAC) systems. This can be a hazard in hospitals, where severely
immuno-compromised patients may develop pneumonia from airway colonization by
these bacteria, which are not pathogenic to normal immunelogically competent
persons (Tobin, 1986). Burn patients may also develop severe wound infections
from these organisms.
Mycobacterium tuberculosis is an extremely virulent bacterial organism
which can cause infection if only a single droplet nucleus is inhaled (Riley,
1982). Its source is infected humans. Mycobacterium is encapsulated and may
survive for fairly long time periods in dried sputum, and may become
reentrained in air streams in buildings.
2-148
-------
Some saprophytic fungi can behave opportunistically and become invasive
human pathogens. Systematic studies of such occurrence do not exist, but some
data are available from general surveys of indoor microfungi. Aspergillus
fumigatus, Mucor, and Absidia species are common in indoor areas involving hay
handling (Solomon and Burge, 1984), while Phialophora and Fusarium, which have
taxa that include human pathogens, have been isolated from humidifier fluid.
Yeasts such as Sporothrix and Geotrichum are found colonizing cool-mist vapor-
izers used indoors, and contaminate the air copiously during operation (Solomon
and Burge, 1984).
Human pathogens such as Blastomyces, Crypotococcus, Coccidioides, and
Histoplasma are saprophytic in natural reservoirs of bird and animal droppings
and enter the body by the respiratory route, causing respiratory and systemic
infections. Systemic mycoses begin as lung infections, but the fungus can
migrate to other organs including the heart, brain, and kidneys. For instance,
disseminated blastomycosis may include liver, spleen, and long bones;
coccidiomycoses cases can result in meningitis, spondylitis, and otomycoses;
histoplasmosis can range from flu-like symptoms to ocular histoplasmosis and
acute disseminated histoplasmosis (Day, 1986). Such serious invasive disease
is rare. However, acute disseminated histoplasmosis is usually fatal in
children. Systemic mycoses have been reported in individuals exposed to air
from contaminated ventilation systems. Coccidiodes and Histoplasma are consid-
ered highly infective. Indoor contamination from these fungi is not known
(Solomon and Burge, 1984). Soil contaminated with animal droppings containing
these organisms is readily reentrained, especially during windy conditions in
arid regions, and epidemics of Coccidioides infection have occurred during dust
storms (Ajello et al., 1965; Flynn et al., 1979). Blastomyces and Histoplasma
likewise can be dispersed through reentrained contaminated soils, and their
spores can penetrate indoor spaces (Ajello, 1967).
2.8.4.2 Mycointoxication. Species-specific information regarding fungi is
relatively sparse, but essential if health effects of the organism are to be
described. Some fungi are pathogenic in humans, and their presence alone is
cause for concern. Some fungi also produce potent mycotoxins. Effects of
these poisons are primarily known from their ingestion, and information con-
cerning toxic potential via inhalation is practically unknown. The best known
example is an isolate of Aspergillus flavus which produces aflatoxin. Afla-
toxins are among the most potent liver toxins and carcinogens presently known
2-149
-------
(Rodricks et al., 1977). Three different aflatoxins (B-,, B2, and G-,), are
known animal carcinogens (Nesnow et al., 1986). Aflatoxin B-, is the most
studied of the three. The mutagenic action of Aflatoxin B-, has been studied
and found genotoxic in many microorganisms, plant, and animal test systems
(Ong, 1975). For example, Aflatoxin B-, is mutagenic to Salmonella typhimurium
(Stark et al., 1979), produces micronuclei in bone marrow erythroblasts of
rats and mice (Trzos et al., 1978; Friedman and Staub, 1977), and increases
sister chromatid exchanges (SCEs) in mouse bone marrow cells (Nakanishi and
Schneider, 1979) and chromosome aberrations in cultured human lymphocytes and
Chinese hamster cells (Dolimpio et al., 1968).
The spores of toxigenie fungi contain mycotoxins, often at very high
concentrations. It is reasonable to assume that these toxins would have a
systemic effect when inhaled, since the inhalation route more effectively
allows systemic entry for dissolved substances than ingestion does. Perhaps 30
to 70 percent of isolates tested appropriately are known to produce toxins, and
isolation of a given species does not necessarily mean that isolate is toxi-
genic. While that can only be determined by testing, it is prudent to assume a
toxigenie potential until proven otherwise (Tobin et al., 1986). Other
Aspergillus species, such as A. niger, are not considered a threat to human
health. Yet A. fumigatus and A. parasiticus are pathogenic, toxicogenic, and
allergenic. Toxigenic fungi do not always produce their toxins, especially j_n
vitro, but it is wise to assume that toxigenie strains produce their toxins in
nature.
Molds such as Stachybotrys, Fusarium, Trichothecium, Trichderma, Aero-
monium, Cylindocarpon, and Myrothenium are known to produce trichocene myco-
toxins. Such toxins produce direct toxic effects as well as immunosuppression.
At low concentrations they produce gastrointestinal lesions, hematopoietic sup-
pression, and suppression of reproductive function. Rapidly growing cells seem
to be affected most. Toxicity of the central nervous system produces symptoms
such as anorexia, lassitude, and nausea. Vague symptoms such as those described
in "building related illness" or SBS can be attributed to trichocene intoxica-
tion. Immune suppression can facilitate opportunistic infections by other
molds or bacteria (Day, 1986).
An example of possible mycotoxin intoxication is that associated with the
fungus Stachybotrys. This is a saprophytic fungus that grows on high-cellulose
media low in nitrogen sources. It can grow on nutrient-poor substrates such as
2-150
-------
dust, shower curtains, straw, and wallpaper. It requires moisture and lack of
competition from vigorous fungi (Jarvis, 1986).
Jarvis (1986) describes a case study of an airborne outbreak of Stachybo-
tryotoxicosis in Chicago, IL. A family occupying a brick home in which a
TM
heavily contaminated cold air return duct and a wood fiber board (Celotex ,
Chicago, IL) ceiling insulation material demonstrated heavy growth of
Stachybotrys atra, complained of recurring symptoms. Among these were cold and
flu symptoms, sore throats, diarrhea, headaches, fatigue, dermatitis, inter-
mittent focal alopecia, and generalized malaise. The father experienced severe
leg pains. Some family members showed signs of psychopathy, and one of the
teenaged sons committed suicide. Repeated hospital and clinical examination
revealed no causes of these symptoms.
Air in the home was contaminated with spores from Stachybotrys atra, and
tests for trichocene mycotoxin were positive. Trichocenes were also isolated
from samples of the fungus contaminating the duct and the ceiling insulation
board. Included were the toxins verrucarin B and trichoverrines A and B.
After the Chicago house was thoroughly cleaned of the S. atra - contaminated
ducts, insulation and ceiling material, the family reoccupied the house and no
longer suffered from the earlier complaints. It is possible to correlate their
symptoms with trychocene intoxication, specifically those isolated from the
house air, and known to be produced by S. atra.
2.8.4.3 Allergenic Reactions. More commonly, fungi are associated with
allergenic reactions indoors, which can vary in intensity and type, and can be
described as follows:
2.8.4.3.1 Allergic rhinitis. This is often termed "hay fever" when seasonally
related, which involves nasal air passage obstruction and itching, sneezing,
and oversecretion of mucus. Conjunctivitis, which involves irritation, itch-
ing, and reddening of the eyes, is often associated. Excessive mucus secretion
and blocking of sinus and eustachian passages provide growth reservoirs where
secondary bacterial infections may implant.
2.8.4.3.2 Bronchial asthma. This disease, which involves a recurrent narrow-
ing of bronchioles and hypersecretion of thick mucus that can block airways, is
accompanied by varying degrees of wheezing, shortness of breath, and coughing.
Secondary bacterial infections can result in bronchitis and more sensitive
reactions to irritants and other allergens.
2-151
-------
2.8.4.3.3 Hypersensitivity pneumom'tis (extrinsic allergic alveolitis). This
ailment is the most serious acute immune reaction to sensitizing substances.
It involves the production of large amounts of IgG antibody, cellular hypersen-
sitivity, and the formation of interstitial granulomas. It causes filling and
variable destruction of the alveoli by inflammatory cells. With continued
exposure irreversible pulmonary fibrosis and eventual pulmonary failure, ending
in death, ensues (Reed, 1981; Solomon and Burge, 1981).
2.8.4.3.4 "Monday complaints." This describes a syndrome generally occurring
on return to a work environment after one or more days' absence. It may be
related to hypersensitivity pneumom'tis. Symptoms and signs include air
hunger, wheezing, coughing, fever, and muscle pains. These flu-like symptoms
appear after hours in the sensitizing environment, often last overnight and
recur, with decreasing severity, on sequential daily exposure. Long-term
involvement is associated with chronic bronchitis and with lung scarring
(Solomon and Burge, 1981).
2.8.4.4 Other Allergens. While the sensitizing agents for these described
illnesses frequently are molds, other biological entities can provoke any of
these symptoms. Nonviable agents such as house dust, mite fecal pellets,
cockroach feces, insect and arachnid dried hulks and body parts, animal dan-
ders, nonviable remains of molds and their spores, dried, reentrained animal
excretions such as saliva, sweat, urine and feces, pollens, and biogenic
volatiles, have also been identified as actors. Among the viable organisms
provoking such responses are molds, amoebae, algae, and actinomycetes.
Amoebic species which have been implicated in hypersensitivity pneumom'tis
are Acanthamoeba spp. and Naegleria gruberi, and these have also been isolated
from humidifier baffle plates and from humidifier waters in homes and factories
(Sykora et al., 1982). Acanthamoeba infection of the cornea among contact lens
wearers with resulting sclerokeratitis and epithelial erosion has recently been
reported (Mannis et al., 1986), and may be related to air contamination from
air conditioning/humidifying systems.
2.8.5 Indoor Air Quality Control Options
Many indoor biological contaminants may be greatly reduced by humidity
control and cleanliness. These practices include:
2-152
-------
1. Humidity control in homes and buildings at a level not conduc-
tive to the growth of fungi, molds, and dust mites. This is
generally in the 45 to 50 percent relative humidity (RH) range.
As indicated previously, dust mites will not survive below 45
percent RH.
2. Periodic and thorough cleaning of all places where water is
likely to collect. These places include: drip pans of humidi-
fiers and refrigerators, cooling towers, and in and around
toilets and wash basins.
3. Periodic and thorough cleaning of carpets and fabrics.
4. Extermination of household insect pests.
5. Periodic cleaning or replacement of furnace and air conditioning
filters and other air cleaning devices.
2.9 PESTICIDES
2.9.1 Introduction
Pesticides (including, among others, insecticides, rodenticides, termiti-
cides, and germicides) are used both by professional pest control businesses
and by occupants of indoor spaces to eliminate a wide variety of organisms,
ranging from rodents to insects, fungi, bacteria, and viruses. They are by
definition poisons, but their range of toxicity varies with their target. They
include dicoumarins, organophosphates, carbamates, and chlorinated hydro-
carbons, among others. Their use, storage, and disposal are, for the most
part, regulated, but cautionary information to homeowners, in particular, must
be clarified and emphasized.
2.9.2 Sources of Pesticide Exposure
A national household pesticide usage study involving over 8,000 households
was conducted by EPA (U.S. Environmental Protection Agency, 1980) in 1976 and
1977. It is estimated that 94 percent of pesticide usage is for agricultural
purposes. Pesticides, herbicides, and fumigants used in agriculture may become
airborne and attach themselves to airborne particulate matter, or, through soil
runoff, contaminate rivers, lakes, and streams which may be a source of water
supplies. A study on the role of house dust in, for example, DDT pollution
(Davies, 1972), indicates that house dust can be a principal source of
insecticides. The concentrations of DDT in house dust was higher than that
found in soils in the area. Table 2-42 shows that there are many sources
2-153
-------
TABLE 2-42. SOURCE OF INDOOR PESTICIDE EXPOSURE
Compound (Trade Name)
Type of Household Uses Leading to
Pesticide Potential Human Exposure
Chlorpyrifos (Dursban®)
Pentachlorophenol
Chlordane
Ortho-Phenylphenol
Propoxur (Baygon®)
Resmethrin
Dicofol
Cap tan
Carbaryl (Sevin®)
Lindane (
Dichlorvos (DDVP)
Insecticide
Fungicide
Insecticide
Insecticide
Disinfectant,
Fungicide
Insecticide
Insecticide
Insecticide
Fungicide
Insecticide
Insecticide
Insecticide
Control of mosquitoes, cock-
roaches and other household
insects; turf and ornament-
al insects; fire ants,
termites, and lice
Exterior wood preservative
Subterranean termite control
Household disinfectant;
post-harvest application
to fruits and vegetables
Control of cockroaches,
flies, mosquitoes; lawn and
turf insects
Control of flying and
and crawling insects; fabric
protection; pet sprays and
shampoos; application on
horses and in horse stables;
greenhouse use
Control of mites on fruit,
vegetable, and ornamental
crops
Seed protectant; fungal
control on fruits,
vegetables, and berries
Control of insects on lawns,
ornamentals, shade trees,
vegetables, and pets
Seed treatment; insect
control in soil, on
vegetables, ornamentals
and fruit and nut trees
Household and public
health insect control; flea
collars and no-pest strips
(continued on following page)
2-154
-------
TABLE 2-42. (continued)
Compound (Trade Name)
Type of
Pesticide
Household Uses Leading to
Potential Human Exposure
2,4-D esters
Malathion
Permethrin (cis and trans)
Heptachlor
Aldrin
Dieldrin
Ronnel
Diazinon
Methoxychlor
Atrazine
a-Hexachlorocyclohexane
(a-BHC)
Bendiocarb (Ficam®)
Folpet
Chlorothalonil (Bravo©)
Dacthal
Herbicide
Insecticide
Insecticide
Insecticide
Insecticide
Insecticide
Insecticide
Insecticide,
Nematicide
Insecticide
Herbicide
Insecticide
Insecticide
Fungicide
Fungicide
Herbicide
Post-emergent weed control
Insect control on fruits,
vegetables, ornamentals, and
inside homes
Control of flies, mosquitoes,
ants, cockroaches, garden
insects
Subterranean termite control
Subterranean termite control
Subterranean termite control
Fly and cockroach control
Control of soil and house-
hold insects, grubs and
nematodes in turf; seed
treatment and fly control
Control of insects in garden,
fruit, and shade trees
Weed control
Manufacture and use dis-
continued in U.S.;
ubiquitous in air, residue
from lindane
Household, ornamental, and
turf insect control
Fungus control on flowers,
ornamentals, seeds, plant
beds; paints and plastics
Broad spectrum fungicide;
wood preservative; paint
additive
Selective pre-emergent
weed control on turf,
ornamentals, and vegetable
crops
(continued on following page)
2-155
-------
TABLE 2-42. (continued)
Type of Household Uses Leading to
Compound Pesticide Potential Human Exposure
Oxychlordane — Oxidation product of
chlordane
Heptachlor epoxide -- Oxidation product of
heptachlor
trans-Nonachlor -- Component of chlordane
PCBs (Aroclors 1242 and 1260) — Used in electrical trans-
formers until 1976
Source: Lewis et al. (1986).
leading to indoor exposures of pesticides (Lewis et al., 1986). As indicated
in Table 2-42, many pesticides are used directly in the indoor environment or
in close proximity (e.g., foundation treatment for termites). Chlordane has
been detected in the air of some termiticide-treated homes as long as 14 years
after application (Repace, 1982; Staats, 1980). It is also well recognized
that many insecticides and herbicides find their way into the home through the
food chain.
2.9.2.1 Emission Rates. While many studies have been conducted to determine
the indoor concentrations of pesticides, limited data are available on emission
rates of pesticides to the indoor environment. Leidy et al. (1984) showed that
emissions decreased with time for diazinon applied in cracks and crevices in a
dormitory. Jackson and Lewis (1981) evaluated emissions of propoxur, diazinon,
and chlorpyrifos from pest control strips. Nelms et al. (1987) conducted tests
in environmental test chambers of paradichlorobenzene emissions from moth
crystal cakes at two temperatures and four air exchange rates. Estimates of
the emissions rates are shown in the second and third columns of Table 2-43.
2.9.3 Exposure to Pesticides
To further define the levels of pesticides occurring in indoor air and the
potential health hazard resulting from such exposure, the U.S. Environmental
Protection Agency's Office of Pesticide Programs is developing and implementing
guidelines that will require that pesticide manufacturers submit indoor respi-
ratory exposure and dermal exposure information for those pesticide products
2-156
-------
TABLE 2-43. EMISSION RATES OF PARADICHLOROBENZENE FROM MOTH CRYSTAL CAKES
(mg/cm2 hr)
Air
Exchange Rate (hr-1)
0.25
0.5
1.0
2.0
Temp.
1.
1.
1.
2.
= 23°C
2
3
7
0
Temp. =
4.
4.
5.
6.
35°
5
7
6
3
C
Note that temperature had a marked effect on the emission rates.
Also, the chamber air exchange rate impacted the emission rates due to the
suppression of evaporation at low air exchange rates due to high chamber con-
centrations.
used indoors. Currently, indoor air monitoring data are required only for
those pesticides (e.g., aldrin, dieldrin, chlordane, heptachlor, chlorpyrifos)
registered for termite control. According to the draft Pesticide Assessment
Guidelines, Subdivision U, Applicator Exposure Monitoring of January 2, 1986,
inhalation exposure generally represents a very small component of the total
exposure; however, the application of dusts, aerosols, and fumigants, or the
application of sprays in enclosed spaces can result in significant inhalation
exposure. Therefore, respiratory exposure information will be required for
those pesticide products used indoors for which (a) the toxicological evalua-
tion indicates that the use of the product may pose an acute or chronic hazard
to human health, (b) inhalation exposure is likely to occur during use and/or
for a period of time after use, (c) data are not available for estimating the
magnitude of exposure for a particular work activity and postwork activity with
an acceptable degree of confidence, and (d) depending on the physiochemical
properties of the product, the conditions of use, and the toxicological
effects, it can be expected that exposure by inhalation will be of significant
concern as compared to exposure by dermal absorption. Indoor sites include,
but are not limited to: homes and apartments, greenhouses, barns and other
farm buildings, commercial buildings and manufacturing facilities, restaurants
and food handling and processing facilities; fumigation facilities, warehouses,
railroad boxcars, schools, hospitals and other health care facilities, and
mushroom houses.
2.9.4 Monitoring
Interest in measuring pesticide exposure in indoor air is relatively
recent. Few studies on pesticides were conducted in the 1970s, with perhaps
2-157
-------
the earliest published work being that of Starr et al. (1974), which describes
measurements in Colorado and concludes that household dust is a major reservoir
for pesticides in the indoor environment. Jackson and Wright (1975) evaluated
the variance in pesticide residues following application using air compressors
and aerosol-type sprayers. They concluded that less pesticide movement
occurred following aerosol application. They also determined that food
contamination after insecticide application was appreciable.
Wright et al. (1981, 1984); Leidy et al. (1982, 1984), and Wright and
Leidy (1982) published a series of papers describing measurements following
application of pesticides in dormitories, food-serving areas, buildings,
service vehicles, and homes. In each case, measurable amounts were detected
in the air, on dust particles, and on the walls, and articles in the rooms,
for as long as 35 days after application.
Pellizzari et al. (1981), Jurinski (1984), Ruh et al. (1984), Gebefuegi
and Korte (1984), Livingston and Jones (1981), and Dobbs and Williams (1983)
also reported measurable pesticide exposure after residential or commercial
application of pesticides. Melius et al. (1984) published a survey by the
National Institute for Occupational Safety and Health (NIOSH) of more than 200
buildings in which office workers had complained of poor indoor air quality.
Pesticides may be a major contributor to the sick building syndrome in some of
these cases.
Increased attention is being paid to improving the methodology for measur-
ing pesticides in indoor air, largely because agencies such as EPA are required
to measure and monitor low levels of pesticides in the indoor environment.
Melcher et al. (1978), Lewis et al. (1986), and Jackson and Lewis (1981)
identified methods for increasing sampling sensitivity for a variety of
chemicals. It is now possible to measure some air pesticide concentrations as
3
low as 0.01 ug/m .
A battery-powered, low-volume (4 L/min) sampler suitable for indoor air
and personal exposure monitoring was developed in the early 1980s (Lewis and
MacLeod, 1982). The sorbent found most convenient to use in this sampler was
polyurethane foam (PUF). Sampling efficiencies have been determined for 58
organochlorine, organophosphate, organonitrogen and pyrethroid pesticides, as
well as for several polychlorinated biphenyl (PCB) mixtures. The sampler was
successfully field-evaluated in 1985 in a pilot test by the Agency's Non-
occupational Pesticides Exposure Study (NOPES) methodology (Lewis et al. ,
2-158
-------
1986; Lewis et a"!., 1987). The sampler is currently being used in the two-
city, 300-home NOPES project. Analytical problems were encountered, however,
with two common household pesticides, acephate and glysophate, so these
compounds could not be included in the NOPES program.
PUF may also be used to collect other SVOCs, but breakthrough problems
occur with SVOCs having vapor pressures greater than 10 mm Hg. A combination
sorbent trap in which Tenax GC or other granular sorbent is sandwiched between
two PUF plugs has been demonstrated to efficiently trap the more volatile SVOCs
(Lewis and MacLeod, 1982). However, this system has not been fully validated
in the laboratory or in the field.
The major requirement in a methodology for monitoring low levels of
pesticides exposure is the ability to sample with the precision and accuracy
needed for regulatory purposes. While technology is available to absorb,
elute, and measure some of the pesticides now in use, no field-tested, quanti-
tatively valid methods are available for testing many others. Methods with
known precision and accuracy are needed to establish exposure measurement
guidelines based on field monitoring (i.e., sampling and analysis).
Out of concern for the lack of information on nonoccupational exposure to
pesticides, Congress in 1985 passed a bill that provided funds for a total
pesticide exposure assessment study to be conducted by the U.S. Environmental
Protection Agency.
In response to this initiative, ORD developed the NOPES project. To date,
a pilot study of nine single-family dwellings has been completed, and Phase I
of the NOPES project is under way. The purpose of the pilot study was to
select, validate, and field test sampling methodologies and survey question-
naires to be used in preparation for the NOPES project. Most of the households
selected for the pilot study belonged to retired or semi retired persons who
generally spend an average of 18 hours indoors daily. Thirty pesticides and
related chemicals were selected for monitoring based on present day and
historical usage (Table 2-42). Indoor, outdoor, and personal exposure air
monitoring and tap water analyses were performed at all of the dwellings. Of
the 30 pesticides and related chemicals tested, 22 were detected in indoor
and/or outdoor air samples (Table 2-44). The concentration of pesticides
occurring indoors was generally higher than the levels found outdoors, with
3 3
peak values ranging from 1.7 to 15.0 ug/m , compared to 0.001 to 0.41 ug/m ,
respectively. Table 2-45 lists the most commonly found pesticides and their
2-159
-------
TABLE 2-44. SUMMARY OF PESTICIDES FOUND IN AIR FROM NOPES
Pesticide
Chlorpyrifos
Diazinon
Chlordane
Propoxur
Heptachlor
trans-Nonachlor
o-Phenylphenol
Uieldrin
Lindane
Aldrin
Ronnel
Captan
Bendiocarb
ot-Hexachl orocycl ohexane
Folpet
Malathion
Chlorothalonil
Dichlorvos
Dicofol
Methoxychlor
Pentachlorophenol
trans-Permethrin
Number of
Indoors
9
8
8
7
7
5
5
5
5
4
4
4
3
3
3
3
3
2
1
1
I
0
Households at Which
Detected in Ai
Outdoors
7
7
6
4
5
3
4
3
2
4
1
1
1
2
4
2
3
1
2
1
1
1
Pesticides Were
r
Respiratory
8
6
6
6
6
5
5
5
3
3
3
3
3
3
3
1
1
0
0
0
0
0
Source: Lewis et al. (1986).
TABLE 2-45. SUMMARY OF MONITORING DATA FOR THE FIVE MOST PREVALENT
PESTICIDES (NOPES)
Pesticide
Chlorpyrifos
Diazinon
Chlordane
Propoxur
Heptachlor
Indoor
(|jg/m3)
0.014 to 15
ND to 8.8
ND to 1.7
ND to 0.66
ND to 0.31
Outdoor
((jg/m3)
ND to 0.30
ND to 0.41
ND to 0.21
ND to 0.0039
ND to to 0.048
Personal
Exposure
((jg/m3)
ND to 8.8
ND to 5.1
ND to 4.2
ND to 0.6
ND to 0.18
*ND = Not detected at ca. 0.001 ug/m3.
Source: Lewis et al. (1986).
2-160
-------
monitored exposure levels. In general, these pesticides were ingredients of
pesticide products found in the house or recently applied by the occupants or
professional pest control operators (Lewis et al., 1986).
Phase I of the NOPES project involves indoor and outdoor air monitoring
of 70 dwellings in Jacksonville, FL. The air monitoring is scheduled to last
for three seasons. Phase II of the NOPES project was to be initiated in
December 1986. This portion of the study will encompass indoor and outdoor
air monitoring of 50 dwellings in a northern city for two seasons.
2.9.5 Health Effects Associated with Pesticide Exposure
There is a battery of information available on the health effects of
pesticides, ranging from acute to chronic exposure effects and carcinogenicity.
However, this information is generally based on animal studies and covers those
effects resulting from the oral or dermal routes of exposure. The limited
information on inhalation exposure is mainly in the form of acute toxicity
studies with actual exposure to the pesticide lasting approximately four hours.
Information on human pesticide exposure via inhalation indoors is lacking, as
are data from long-term animal inhalation studies.
2.9.6 Mitigation and Control Options
2.9.6.1 Ventilation. As with other indoor air pollutants, increased ventila-
tion is effective in reducing pesticide concentrations via dilution and flush-
ing, although limit data are available. Levin and Hahn (1986) showed increased
ventilation to be effective in reducing levels of pentachlorophenol in a build-
ing with exposed inside wood beams that had been treated with this compound.
2.9.6.2 Air Cleaners. No references were found on the application of air
cleaning devices for the control of indoor pesticide levels. Depending on the
characteristics of the pesticide vapor, adsorption and/or catalytic oxidation
may be applicable. Specific studies would be required.
2.9.6.3 Material/Product Selection and Use. Levels of pesticide residuals
in the indoor environment can be controlled either by avoiding use or by using
smaller amounts and less frequent applications. When a pesticide must be used,
careful adherence to product label directions is essential. Many pesticides
are controlled via registration and are available only through professional
applicators. The methods of applying pesticides can also affect exposure.
2-161
-------
Wright and Jackson (1975) indicate that aerosol-type sprayers are more effec-
tive than compressed-air sprayers in limiting the application to the target
areas.
2.9.6.4 Other Measures. Indoor exposures to pesticides from treated wood can
be reduced by sealing. Levin and Hahn (1986) report that pentachlorophenol
levels are lowered when exposed beams are sealed with a polyurethane varnish,
although the long-term effectiveness of this approach has not been determined.
2.10 NONIONIZING RADIATION: EXTREMELY LOW FREQUENCY ELECTRIC AND MAGNETIC
FIELDS
2.10.1 Occurrence and Sources of Nonionizing Radiation
Electric power generation, transmission and distribution, and final
utilization in homes, offices and factories is one hallmark of our modern
industrialized society. In the United States, most electric power is in the
form of electric current alternating at a frequency of 60 cycles per second (60
Hz), which falls into the extremely low frequency or ELF portion of the electro-
magnetic spectrum. Because of the high voltages and currents that occur at
various stages from generation to endpoint utilization, there are attendant
electric and magnetic fields that permeate the surrounding environment. Thus
there are magnetic and electric fields generated in or near every electrified
house, office building, and factory in this country, that expose the occupants
to electromagnetic fields at the power line frequency. The same situation
exists throughout the world, except the alternating current may be at a differ-
ent frequency (most likely 50 Hz).
There are many sources of electric and magnetic fields commonly found
indoors. These include any appliance that (1) has an electric motor (refriger-
ator, freezer, clothes washer, hair dryer, shaver, food mixer and chopper,
clock, vacuum cleaner), (2) has electric heating elements (stove/oven, clothes
dryer, coffee maker, iron, hot water heater, electric blanket, heated water
bed), or (3) uses electric light bulbs. In addition to the 60 Hz fields asso-
ciated with these household devices, many generate electromagnetic fields of
various frequencies, especially when the appliance is energized. Fluorescent
lights and light dimmer switches emit electromagnetic fields at other frequen-
cies in addition to 60 Hz. Commonly used appliances such as television sets
and computer terminals generate frequencies in the kHz range that expose
2-162
-------
children and adults from a few minutes to eight or more hours per day. Other
sources of nom'onizing radiation that seem logical to include in this chapter
are devices that operate at microwave frequencies, for example, microwave ovens
and anti-theft devices.
2.10.2 Distribution of Levels and Exposure
Biologically relevant exposures to electric and magnetic fields from power
distribution systems, building wiring, and electric appliances are widespread.
The most obvious exposure occurs in the electric fields beneath high-voltage
transmission lines that carry electric power over long distances. The electric
2
fields generated may be in the 1 to 2 kV/m range. Less obvious are the
magnetic fields emitted from the lower voltage distribution lines in front of
most homes. These fields may be in the 1 to 2 mgauss range. Further, the
electric and magnetic fields due to building wiring are of the order of 0.5
mgauss and 10 V/m, which can peak to over 10 times that amount when electrical
equipment is energized. One obvious example of exposure to biologically-
relevant fields is from electrically heated beds that are used for six to ten
hours a night during the cooler months of the year. The 60 Hz electric field
2
measured 30 cm from various electrical appliances range from 250 V/m for the
2
electric blanket to 2 V/m for the incandescent light bulb. Electric fields at
o
the center of rooms in a typical home range from 0.8 to 13 V/m (Sheppard and
Eisenbud, 1977).
It has been determined that man-made fields can be many orders of magni-
tude stronger than those of natural origin. However, many monitoring questions
remain unanswered about the prevalence of man-made electric and magnetic fields
normally found indoors, including their intensity, periodicity, and relative
orientation with respect to naturally occurring magnetic and electric fields.
All these aspects are relevant to potential biological reactions to the
presence of these fields.
2.10.3 Health Effects of Nonionizing Radiation
The most prominent human health effects so far observed from exposure to
electromagnetic fields, caused by the use of electric power, occur in the areas
of cancer and reproduction. The experimental results that cause the most
concern for human health come from two independent epidemiological studies,
both of which arrive at the same conclusion, namely, that magnetic fields
2-163
-------
from distribution lines in the front of homes are positively correlated with
an increased incidence of childhood cancer (Wertheimer and Leeper, 1979;
Savitz et al., 1986). In another recent epidemiological study conducted in
Sweden (Tomenius, 1986), an increased incidence of nervous system tumors was
found in a population exposed to power-line frequency fields, in this case 50
Hz fields. This report introduces another area of concern, namely, the
interaction of electromagnetic fields with the central nervous system.
Home appliances may be another area of health concern. Wertheimer and
Leeper (1986) found the use of the electric blanket and the electrically heated
water bed to be positively correlated with (1) increased spontaneous abortions
for mothers exposed during the first trimester and (2) a reduction in birth
weight and a lengthening of the gestation period for infants born to exposed
mothers. In addition, there is laboratory research (see 6 below) showing that
electric fields affect brain tissue from chicks exposed during embryogenesis,
organogenesis, and organ maturation at exposure levels that are more than one
order of magnitude below those experienced by the pregnant women studies.
Another laboratory report indicates potential teratogenic effects from exposure
to fields emitted from television sets and computer terminals. Thus some
effect of ELF and higher frequency (kHz) fields on the fetus is possible.
Although no cause-and-effect relationship can be proven in the epidemio-
logical studies, the number and mutually supportive results indicate that more
detailed research is needed. There are laboratory results that are consistent
with these epidemiological reports and that indicate the complexity of the
interplay between the electric and magnetic fields and biological systems.
Some of the more pertinent laboratory reports are cited below:
1. The recent result of a study by Phillips et al. (1986) indicates
that human cells from certain tissues are susceptible to exposure
from 60-Hz fields and can be altered in a way that enhances their
transformation into the cancerous state.
2. Cells from the immune system can be affected by 60 Hz fields in
a way that diminishes their natural function to seek out and
destroy invading organisms. Luben et al. (1982) have shown that
the cytotoxic activity of lymphocytes can be inhibited by
electromagnetic field exposures. A weakened immune system would
not be able to fully respond to the needs of an organism to fend
off infections from outside the body and possibly prevent cells
within the body from changing to a more cancerous state.
2-164
-------
A classical biological system, the Dipteran salivary gland, has
recently been shown to undergo unexpected changes in response to
60 Hz fields (Goodman and Henderson, 1986). These changes
involve the induction of messenger RNA and the production of
proteins from genes that were dormant. This demonstrates that
60 Hz fields can induce changes that affect fundamental
biochemical pathways which, in some cases, could lead to the
induction of oncogenes and ultimately to cancer.
Teratogenic effects were observed in the fetuses of mice exposed
to a level and frequency of magnetic field commonly measured in
the vicinity of television sets and computer terminals. A meet-
ing presentation by Tribukait and colleagues (1986) from the
Karolinska Institute has caused considerable scientific
interest, but no other work has been reported that independently
examines this result.
Brain tissue jri vitro can be affected by 60 Hz fields at levels
commonly found in homes. Blackman et al. (1985) have shown that
certain frequencies of electric and magnetic fields can alter
the interaction of calcium with brain tissue. Since calcium is
important for neurotransmitter release and membrane integrity,
and since it operates as a second messenger in biological
systems, this perturbation may prove to have physiological
significance.
A recent report at the Bioelectromagnetics Society Annual
meeting (Blackman, 1986) indicated that the electric-field
intensities found in homes could be biologically active in
causing changes in the brains of developing organisms; these
changes remained after the chicks were hatched. In this case,
those animals exposed during gestation to the power-line
frequency used in the United States were different from those
exposed to the power-line frequency used in Europe. The conse-
quences for humans are unknown, but may be associated with an
increase in generalized stress or to some more specific ailment,
either alone or in combination with other agents.
The earth's magnetic field was recently discovered to affect the
particular frequencies of fields that cause biological changes.
Blackman et al. (1986) showed that effective frequencies could
be made ineffective, and vice versa, solely by making slight
adjustments in the local value of the earth's magnetic field.
This result has stimulated a number of hypotheses of the
mechanism(s) of interaction and experimental research.
One such test of the influence of the earth's magnetic field
showed that behavior in rodents could be altered by exposure to
60 Hz fields under selected, but widely prevalent, magnetic
conditions (Thomas et al., 1986). The data (1) support the
findings of Blackman and others that 60 Hz fields can cause
changes in biological systems, and (2) show that exposure can
eventually result in changes in the live animal, in this case,
behavioral modification. The underlying mechanism has not yet
been discovered, thus the consequences for human health are
unknown.
2-165
-------
In both the epidemiclogical reports and the laboratory experiments, some
dose response data have been acquired. However, the results are not all
internally consistent. Further, the average time for an effect to occur can
vary widely depending on the initial conditions of the biological system, the
ambient, natural electric and magnetic environments, the prior exposure history,
and the presence of co-stressors. Thus, the results provide strong support for
the need to establish the underlying mechanism(s) of action so that appropriate
judgments can be rendered regarding relative risk from exposure.
2.10.4 Estimate of Population at Risk
The results of the epidemiology studies suggest two principal areas of
concern for human health: reproduction and carcinogenesis. Thus the human
fetus is potentially at risk. Furthermore, the epidemiological evidence sug-
gests that human beings of all ages are potentially at risk for cancer.
Laboratory data that support the latter suggestion show that exposure to elec-
tromagnetic fields may reduce the immune response, alter the utilization of
genomic information, and stimulate biochemical activity involved with trans-
duction of the genetic message. An epidemiological report and additional
laboratory evidence demonstrate that exposure to ELF fields can induce changes
in the nervous systems of animals. The changes include cancer, behavioral al-
terations, and biochemical changes. The latter two responses may add to a
generalized increase in stressful living conditions for the entire population.
2.10.5 Mitigation and Control Options
There are a variety of options to limit exposure to electric and magnetic
fields caused directly or indirectly by electric power usage. The particular
methods to be selected would depend on understanding the mechanism of action
so that safe limits can be ensured. For example, a change in the phasing of
current flow in adjoining wires outside homes and in restructuring the return
ground currents might be sufficient to eliminate the childhood cancer problem.
Appropriate juxtaposition of wires in electric blankets and in water bed
heaters could reduce fields from these sources. Fields from television sets
and from computer terminals might be reduced by appropriate shielding and
counter measures, such as a dummy fly-back transformer to emit the inverse wave
form. Fields from these and other appliances such as light dimmers and
microwave ovens may require individual mitigation techniques designed to reach
certain limits defined by future health effect studies.
2-166
-------
2.10.6 Conclusions
There are data consistent with a possible connection between exposure to
power-line frequency electric and magnetic fields and 1) the appearance of
cancer from basic biological changes in the utilization of genetic information
(Goodman and Henderson, 1986), 2) to enhanced cellular transformations in test
tubes (Phillips et al., 1986), 3) reduced ability to resist immunological
challenges (Luben et al. , 1982), 4) an increase in human childhood cancer
(Wertheimer and Leeper, 1979). In another area, effects have been shown on
brain tissue i_n vitro (Blackman et al., 1985), on brain tissue in animals
exposed before they are born (Blackman, 1986), and on certain behaviors of
animals (Thomas et al., 1986). Thus the specter of potential cancer induction/
promotion and of aberrant behavioral changes have been associated with exposure
to fields in homes caused by the electric power system and the use of home
appliances. These reports do not unequivocally demonstrate that power-line
frequency fields are a human health hazard separate and apart from the known
hazard due to electric shock and burns; however, the combined reports do indi-
cate that a great deal of caution must be exercised before allowing any
increase in exposure of the general population until a better understanding is
obtained of the underlying mechanism of action and possible synergism with
other potentially hazardous agents and stressors in the environment.
There is a growing public awareness of the potential harm that may be
caused by exposure to power-line frequency electric and magnetic fields. For
example, the case of a 1985 settlement in Harris County, Texas which involved
exposure of 4000 children in schools near a 345 kV transmission line. The
jury felt there was a reasonable chance of hazard to the children's health.
(Transmission/Distribution Health and Safety Report, 1986).
2-167
-------
3. BUILDING SYSTEMS
3.1 INTRODUCTION
Previous sections of this report dealt with indoor air on a pollutant
basis. Sources, effects, and IAQ control options were discussed on a pollutant
category basis. In this section we will look at IAQ as a building related
issue.
The interactions of sources of pollution and the building system determine
IAQ. The building system, as used in this report, includes all aspects of the
building. The structural aspects of the building, the arrangement of rooms and
furnishings (including pollution sources), the construction materials used in
the building, the movement of air from room to room via natural and mechanical
ventilation, the heating, ventilating, and air conditioning system (HVAC) are
all included in this definition. The term building system also covers the
activities in the building.
The building system can:
transport pollution (via either the HVAC or natural ventilation),
remove pollution (e.g., forced exhaust to outside or natural
exhaust by open windows or sink effects),
introduce outside air pollution (e.g., via a ventilation inlet
located near a loading dock), and
control pollution (e.g., via air cleaners)
produce pollution (e.g., off-gasing from materials), and
affect ventilation (e.g., barriers can reduce ventilation).
The building environment (e.g., temperature, humidity, and lighting,
affects the behavior of pollutants. For example, pollution emission rates are
often a function of temperature. Interactions between pollutants are also
3-1
-------
affected by the environment. In some cases the effects of pollutants depend on
environmental factors.
The building environment is especially important for biological pollu-
tants. Temperature and humidity play a significant role in determining the
types and quantities of biological pollutants that exist in a building (Morey
et al., 1986).
Control of indoor air pollution requires careful operation of both sources
and the building system. Sources should be removed from the building if
possible. Sources that cannot be removed should be operated to minimize
pollution. Pollution should be confined to areas near sources. Adequate
ventilation throughout the building can prevent pollution build-up.
Although the building system plays a key role in determining IAQ, there
has been little systematic work examining the role of the building. Much of
the research has been ad hoc research related to a specific building.
3.2 THE BUILDING SYSTEM AS A SOURCE OF INDOOR AIR QUALITY PROBLEMS
3.2.1 General
The building system contributes to indoor air problems in a variety of
ways. One of the most obvious ways that the building system affects IAQ is
through air circulation and ventilation. Poor ventilation, low ventilation
effectiveness, and transport of pollutants, are examples of air circulation
effects.
The building system can be a source of pollution emissions. Emissions can
be generated by materials and furnishings in the building or by the HVAC system.
Other portions of the building system, for example, the plumbing system, can
also generate air pollution emissions.
The arrangement of space, equipment, and activities in a building also
affects IAQ. Walls, doors, and furnishings determine circulation patterns and
affect ventilation effectiveness. The location of pollution generating
activities in relation to other activities is important.
3.2.2 Ventilation Problems
Poor ventilation, which is one of the most obvious causes of poor IAQ,
takes several forms, as described below.
3-2
-------
insufficient ventilation, where the amount of fresh air intro-
duced into a building is inadequate
ineffective ventilation where some portions of the building do
not receive adequate ventilation
inadequate exhaust of air, where the amount of air removed from
an area of high emissions is insufficient to transport the
pollutants (Wallingford and Carpenter, 1986; Ferahian, 1986;
Seppanen, 1986)
The building system serves as a highway for moving pollutants from one
area of the building to another. Q fever outbreaks, for example, are caused by
transport of this infectious agent, a virus, through the HVAC system (Bayer,
1982).
Ventilation provides a pathway for polluted outdoor air to enter the
building. An open window near a loading dock, for example, provides an easy
way for pollutants generated by trucks at the dock to enter the building. An
air intake vent located near a source of pollution provides a pathway for
outside pollution to enter the building.
The arrangement of furnishings in the building can disrupt ventilation and
air circulation. This often results in reduced ventilation effectiveness and
poor IAQ.
3.2.3 Source Effects
The building system provides breeding grounds for many biological contami-
nants. The dust and other materials that collect in ventilation duct work often
provide an ideal medium for the growth of various biological contaminants (Morey
et al., 1986). Fiberglass duct work, as it ages and collects a variety of
organic material on its roughened surface and in crevices formed by glass fiber
mats, becomes an excellent growth medium for molds, under appropriately humid
conditions. Fiberglass batting used to line duct work to reduce noise acts
as a filter for dust and other particles, which provide good media for the
growth of molds and/or bacteria. The duct work also provides a pathway for
distributing the biological contaminants throughout the building.
Biological contaminants can also grow on furniture and furnishings (Morey
et al., 1986).
Gaseous organic pollution can be produced by off-gassing from plastic duct
work. Fibers can be generated by abrasion of duct material, and dust and
other materials collected in a duct may be resuspended from time to time. A
3-3
-------
cycle of pollutant collection followed by resuspension, for particles, or
regeneration, for gases, can then be established.
Materials such as minerals and conditioning additives from water used in
humidification systems may also be spread throughout the building (Burge, 1985).
Some types of electrostatic air cleaners may generate ozone when improperly
operated. Dirty air filters can also reduce IAQ (Kress and Fink, 1986). Other
portions of the building system such as the plumbing system, swimming pools,
and construction materials also affect IAQ (Hodgson and Kreiss, 1986; Gamble
et al., 1986; Morey et al., 1984; Wallingford and Carpenter, 1986). As
discussed in other sections of this report, the materials used in the building
can be major sources of pollutants. Once these pollutants are released, they
can be then transported throughout the building.
Pollutant interactions can occur within buildings. For example, organics
and bioagents can adsorb to particles, which are then transported through-
out the building. Other important interactions include particle formation from
condensation and particle evaporation. These processes may occur in the build-
ing space or in the HVAC system. Vaporization of organic vapors from tobacco
particles collected by air cleaners can be a significant problem in situations
where air cleaners are being used to provide "smoke-free" air.
3.2.4 Arrangement of Building Space and Activities
The arrangement of space, doors, walls, and furnishing is important for
IAQ. Walls, partitions, and furnishings affect air circulation and ventila-
tion. Furnishing can be located to block air intakes or air outlets and thus
reduce ventilation effectiveness.
Quite often the original design of a building is based on assumptions of
where specific activities will occur. Activity patterns may be changed, and
pollution producing activities moved into new locations. New activities
and equipment may also be introduced into the building. Unless care is taken
in locating these activities and equipment, an IAQ problem may be created.
Location of people and their activities is of growing importance. This is
especially true with regard to smoking. Many local regulations require that
smoke-free air be provided. Providing such air while allowing smoking will
require careful attention to the location of smokers in a building.
3-4
-------
3.3 MITIGATION OF INDOOR AIR POLLUTION
3.3.1 General
The building system can play a major role in reducing indoor air pollution.
A properly designed and operated building can eliminate or reduce indoor air
pollution by:
confining pollution to areas near sources,
exhausting pollution before it mixes with building air,
diluting pollution with outside air, and
removing pollution with air cleaners.
Indoor air pollution control requires combining source control with good
building system operation. Source control minimizes the amount of pollution
generated. Building system operation can then minimize the impact of the pollu-
tants generated. Details on source control are presented in previous
sections of this report.
A building system approach to IAQ mitigation will result in the most cost-
effective mitigation. Under the building system approach, all aspects of the
problem and control options are analyzed. The effects of various actions on
each other are also analyzed, as are the economics of each option. Finally, the
most cost-effective option is selected.
Because many of the techniques used for IAQ control are discussed in other
sections of this report, most of the discussion in this section will be brief.
3.3.2 Confining Pollution
Confining pollution-producing activities to specific locations is a cost-
effective way of improving IAQ. For confinement efforts to be effective, the
air circulation between the pollution locations and the rest of the building
must be minimized. Thus, separate HVAC systems should be provided for areas
which generate pollution.
3.3.3 Exhausting Pollution
Exhausting pollution before it mixes with the rest of the building is
another common mitigation procedure. Cooking stoves, for example, may have
3-5
-------
hoods and filters to remove pollutants before they can mix with the building
air. Bathroom moisture and odors can be exhausted directly to keep them from
the rest of the house. Exhausting pollution can be combined with confining
pollution in a effective manner.
Local exhaust of pollutants can be a cost effective method of improving
IAQ. This technique is often overlooked. Local air cleaning is another option
that can be used in many situations.
3.3.4 Dilution With Outside Air
Many IAQ problems can be mitigated by increasing building ventilation.
The increase in ventilation reduces pollutant concentrations by dilution with
outside air. However, modern practices of indoor climate control may make
increased ventilation economically unattractive. Increasing the ventilation
rate increases the amount of air to be heated or cooled, and increases the
amount of treated air that is exhausted to the outside. While increasing
ventilation is a theoretical possibility for improving IAQ when outdoor air is
of satisfactory quality, it may not always be practical. Increased ventilation
effectiveness may eliminate or reduce the need for increased ventilation
(Seppanen, 1986). An increase in ventilation effectiveness makes maximum use
of the building ventilation.
3.3.5 Air Cleaners
Air cleaning is one way to gain some of the benefits of increased ventila-
tion without diluting with outside air. High efficiency air cleaners can
remove particulate matter and other pollutants. The major drawbacks of most of
the currently available air cleaners are low efficiency for many pollutants,
high capital costs, and high operating costs. Improperly operated and
maintained air cleaners can even increase pollutant levels.
The efficiency of many types of particulate air cleaners is not well known.
This is especially true for removal of small particles which are important for
defining health effects. The performance of air cleaners for gas, especially
at the low concentrations found indoors, is also largely unknown.
Air cleaner effectiveness is determined by a combination of the efficiency
of the air cleaner and the rate at which air is circulated through the cleaner.
High effectiveness requires high air circulation rates (several changes per
hour) and avoidance of short-circuiting (returning the air cleaner exhaust
directly to its intake).
3-6
-------
Air cleaners can help mitigate a wide range of IAQ problems, for example,
ETS levels can be reduced with high efficiency air cleaners.
3.3.6 Selection of Materials
As discussed in previous sections of this report, the materials used in a
building can be major sources of pollution. Proper selection of materials can
improve IAQ.
3.3.7 Elimination of Entry Routes
In cases where the IAQ problem is caused by entry of pollutants from out-
side the building, entry routes should be eliminated. The actions required to
eliminate the entry routes depend on the nature of the problem. For example,
pollutants entering due to poorly placed air intakes can be blocked by relocat-
ing the air intake.
See the radon (Section 2.7) and pesticides (Section 2.9) discussions for
other techniques for eliminating entry routes.
3.4 COMFORT AND OTHER ISSUES
Comfort is an important part of IAQ. Mitigation measures must address
comfort, as failure to do so can reduce or eliminate the benefits of the
mitigation effort.
There is a danger that comfort will be reduced when increased ventilation
or increased air circulation are used to improve IAQ. High ventilation or
circulation rates may create drafts that cause discomfort. As a result people
may block air circulation ducts to aleviate their discomfort and thus eliminate
many of the benefits for high ventilation rates. Because drafty rooms are per-
ceived as cold, people may also turn up the thermostat to increase heat, which
imposes additional energy penalties.
Building climate (temperature and humidity) can be controlled to minimize
IAQ problems. Control of temperature and humidity is especially important for
controlling biological pollutants such as molds. Many biological pollutants
require special temperature and humidity conditions in order to grow.
Odors are another comfort-related issue. Ventilation standards are based
in part on odor impressions (American Society of Heating, Refrigerating and
Air-Conditioning Engineers, Inc., 1981). Odor is an important component of ETS
and the perception of smoke-free air (Clausen et al., 1986a,b).
3-7
-------
The general importance of comfort is addressed by Davidge (1986), who
discusses the interactions between various factors affecting comfort. Individ-
ual comfort factors may be judged satisfactory, but their interactions may
result in the perception of poor air quality.
3.5 MEASUREMENT AND DIAGNOSIS OF BUILDING SYSTEM FACTORS
3.5.1 General
Solving IAQ problems requires careful diagnosis to identify problems
and solutions. Diagnosis is often hindered by the large number of potential
pollutants (Berglund et al., 1986). The identification of individual pol-
lutants is very difficult, especially since the problem may not be due to any
single pollutant. Some progress has been made in understanding the inter-
actions between pollutants, but more work is needed (Berglund et al., 1986).
Tracing a pollutant to a source is also often difficult. Monitoring
techniques that improve source identification are lacking. Modeling is
necessary to better understand the effects of the building system on IAQ and
it is useful in determining the effects of IAQ on health (Leaderer, 1986).
3.5.2 Air Circulation/Ventilation
Air circulation in a building is a combination of natural and mechanical
circulation. The air circulation in most commercial buildings is dominated by
mechanical circulation. Mechanical circulation is also important in many homes
for much of the year.
Measurement of air circulation falls into three main categories:
a. Measurement of air circulation.
b. Measurement of total ventilation (mixing with outside air).
c. Measurement of local ventilation.
Air circulation measurements can be made in HVAC ducts or in the various
parts of the building. The measurements are made with instruments designed for
low flow rates.
3-8
-------
There are a variety of methods for measuring ventilation (air exchange
with outside). Methods for measuring overall ventilation are fairly simple,
inexpensive, and reliable. Methods for determining ventilation of individual
rooms are expensive. Some of the more common methods are summarized in
Table 3-1.
Methods for relating the various measurements of air circulation and
ventilation to ventilation effectiveness are not well developed.
3.5.3 Pollutant Concentrations and Identification
Indoor air is a complex mixture made up of a host of chemicals. The
chemicals may be present as particles or gases. Quite often the gases can be
sorbed onto the particles. The analysis of this complex mixture is difficult
and requires a variety of instruments.
3.5.3.1 Particles. The particles in indoor air range in diameter from less
than 0.05 to greater than 100 micrometers. The chemical composition of the
particles ranges from simple inorganic compounds to complex organic compounds.
The particles may be of a biological nature such as pollens or molds, bacteria
or viruses, or they may be non-biological particles that are associated with
allergenic molecules of biologic origin.
Analysis of the particulate problem in a building starts with the analysis
of the particle size distribution throughout the building. The instruments
necessary for these measurements are discussed in the particulate section of
this report. The particle size distribution is the most important physical
property of the particles in a building. Most of the behavior of the parti-
cles, such as transport, interactions with other pollutants, and deposition in
lungs depends on the particle size distribution.
Techniques discussed in other sections of this report can be used to
determine the chemical composition of the particles. Biological activity of
the particles should also be determined.
3.5.3.2 Gases. Analysis of the gases found in buildings is discussed in the
various pollutant sections in Chapter 2. Because indoor air is a complex mix-
ture, these measurements can be quite difficult and expensive. In cases in
which the identity of the pollutant is known, the expense and difficulty can be
reduced by limiting measurement to the problem pollutant. Unfortunately, in
most cases the pollutants causing specific problems are not known and a full-
spectrum analysis is necessary (Hedge et al., 1986).
3-9
-------
TABLE 3-1. METHODS OF AIR INFILTRATION MEASUREMENT
Method
Determines
Advantages
Tracer gas (instantaneous)
Tracer gas (continuous)
Tracer gas container sampling
^ Whole house pressurization
i—1
o
Component pressurization
Thermography
Smoke Pencil
Ventilation of who or part
or part of building
Ventilation of whole or
part of building
Ventilation of whole or
part of building
Airtightness of the building
building envelope
Airtightness of the building
envelope
Qualitative detection of
leakage sites
Qualitative detection of
leakage sites
Gives information about ventilation
rate under running conditions.
Gives ventilation rate over long time
under different conditions.
Simple and inexpensive.
Gives information about the leakiness
of the building envelope.
At high pressure differences the
method is relatively insensitive
to weather.
Inexpensive.
Quantifies air leakage through
building components.
Simple equipment.
Provides information about leakage
sites and defects in the thermal
insulation.
Provides Information about leakage
sites and air movements.
Simple and inexpensive.
Indirect method.
Result depends on actual
conditions.
Mixing is difficult.
Needs special training.
Indirect method.
Expensive.
Indirect method.
Low control of taking
samples.
No information about actual
ventilation degree.
Simultaneous air
changes/hr
Air changes at
operating con-
ditions
Air leakage at high
pressured differ-
ences
Time to adjust equipment. Air leakage
Expensive. Leakage sites
Needs temperature difference
of at least 10°C.
Requires special training.
Difficult to find leaks,
especially with internal
pressurization.
-------
3.5.4 Questionnaires
Conventional measurements of pollutant concentrations often fail to uncover
IAQ problems. In these situations, questionnaires are often the only tool for
defining the problem (Hedge et al., 1986). In most cases properly designed
questionnaires play a key role in diagnosis of building problems. Follow-up
questionnaires are used to determine whether or not corrective actions have
solved the problem. The value of a questionnaire is strongly dependent on its
design and on the analysis of the responses.
3.5.5 Diagnosis
The purpose of making the measurements discussed above is to diagnose a
building problem. The diagnosis includes identification of pollutants causing
the problem, identification of the sources of pollution, and identification of
building related factors that allow the pollutant to be a problem. When the
diagnosis is complete, action to mitigate problems can be be implemented.
Relating the measurements to problems and sources is a difficult job.
Although there are case-study examples of the diagnosis process, a systematic
diagnosis procedure has not been developed. Procedure development is hindered
by the fact that indoor air is a complex mixture problem. Quite often there is
no single element or simple combination of elements that separates a pro-
blem building from a nonproblem building (Hedge et al., 1986). Even when the
factors causing the problem are identified, tracing these factors to a source
can be difficult. This process could be improved if signatures for various
sources could be determined.
There are cases where diagnosis has been successful (Hodgson and Kreiss,
1986). In many cases, however, the exact cause of a problem remains unknown.
In these situations it is easier to recommend increased ventilation as a
mitigation approach than it is to fully diagnose the problem.
3.6 INDOOR AIR MODELS
Analysis of pollutants and buildings is necessary for understanding IAQ.
Well-integrated IAQ models are essential for this analysis. The IAQ models
should include all aspects of the building system, including sources and sinks
for pollutants, pollutant interactions, ventilation, air cleaners, and health
effects, among others.
3-11
-------
Some aspects of the IAQ problem have been well-modeled (Walton, 1985;
Versar, 1987; McNall et al., 1985). Most of the available models concentrate
on the ventilation aspects of indoor air. These ventilation models are useful
to those involved in designing ventilation systems. Because of the interest
in energy conservation, models for analysis of energy impacts have been devel-
oped. The U.S. Department of Energy continues to develop such energy impact
models. The available models are research models that describe limited parts
of the IAQ problems in detail. The models have not been developed to the level
of usefulness or ease of use required for general application (Versar, 1987).
While the available models provide an analysis of the specific process being
modeled, they neglect important parts of the indoor air problem. Sources and
sinks for pollutants, pollutant interactions, and air cleaners are a few of the
areas neglected by most IAQ models.
Realistic representation of source and sink terms is essential to under-
standing IAQ. Important data necessary to model source and sink terms realisti-
cally are now being developed through EPA's IAQ program.
Present models tend to neglect the factor of pollutant interactions, which
often determine how pollutant levels will affect people. For example, many
gaseous pollutants (organic and inorganic) adsorb to particles that carry the
pollutant to the lung. In the absence of particles, the gaseous pollutants
might not reach the lung. The interaction of radon and particulate matter is
an example of such a carrier phenomenon.
Little work has been done to include the effects of realistic air cleaner
performance on IAQ in existing models since the effects of air cleaner operation
on IAQ have not been developed. Models of particulate air cleaners do not
include effects of particle diameter on air cleaner performance, yet this is
critical to analyzing the effects of particulate air cleaners on IAQ.
Available models are also weak in correlating the effects on health or
welfare of pollutants with the pollutant concentrations predicted by the model.
Such integration in necessary if one is to estimate exposure reduction as a
result of proposed mitigation efforts. When such models are developed they can
be used to evaluate risks involved in a given indoor situation, and can be used
to evaluate risk reduction offered by various mitigation options.
3-12
-------
4. WELFARE CONCERNS
4.1 INTRODUCTION
Discussions in the previous chapters have focused on aspects relevant to
the evaluation of health effects. To some extent, these considerations are
similar to welfare effect assessment. As noted below, however, there are data
requirements specific to materials damage, soiling, and odors. More important,
however, is the identification of significant exposure scenarios. Of concern
are situations in which materials susceptible to indoor air pollutant damage
must perform an essential function or are valuable in cultural or historical
terms. A frequently discussed example of the former case is telephone switch-
ing equipment, vulnerable to corrosive attack both by gases and soluble ions
in airborne particulate matter (Sinclair and Psota-Kelly, 1985).
As a relatively young nation, the United States has had less occasion than
its European counterparts to be concerned with preserving archives and cultural
properties from atmospheric attack, either in outdoor or indoor exposures. A
report titled Impact of Air Pollutants on Materials, developed for the North
Atlantic Treaty Organization's Committee on the Challenges of Modern Society,
highlighted the member nations' concern for effects on cultural property.
Results of investigations documenting observed damage to objects housed in
museums, galleries, archives, and similar structures were reported. A number
of protective and remedial techniques were discussed.
The U.S. approach to estimating welfare effects of materials damage and
soiling has been almost entirely focused on estimates of cost of damage to
materials exposed to ambient air pollution. For soiling, the perspective has
often been limited to residential or household cleaning. Valuation of aesthe-
tic or historic worth does not lend itself to development of variables for use
in applying physical or economic damage functions. There is, for example, sub-
stantial evidence that "indoor soiling causes significant and, in some cases,
irreversible damage to cultural property" (Baer and Banks, 1985). Dust, soot,
ETS residue, textile fibers and alkaline aerosols from setting concrete have
4-1
-------
all been encountered in the deterioration of works of art, documents, and his-
toric artifacts. Paper, paintings, textiles, and wax objects have been identi-
fied as especially susceptible to particulate pollution; metals and photo-
graphic materials are vulnerable to attack by acid fumes. Protective measures
such as special air conditioning and filtering systems have been developed to
mitigate these effects on works of art and other cultural properties in museums,
galleries, and libraries.
The questions to be answered regarding indoor air pollution and materials
damage, soiling, and odors relate to the need for identification of incremental
effect: In what situations will adverse welfare effects attributable to indoor
air pollution be an important factor relative to other factors commonly en-
countered? To what extent, for example, is the decision to replace, repair, or
take mitigative action solely a function of indoor air pollution? With regard
to soiling of indoor surfaces, is the decision to clean a function of particle
accumulation, or are other considerations (e.g., regular cleaning schedule,
volume and nature of activity) of equal or greater significance? Are materials
damaged at a rate significantly greater by air pollutants than by normal wear
and tear? How much is variability in perceived odor a function of pollutant
concentration, and how much a function of receptor sensitivity?
As noted above, research on the effects of air pollutants on materials has
focused on exposures to ambient air concentrations. Where exposures are
similar in the indoor environment, the research data so developed are appli-
cable to materials exposed to indoor air pollutants. As Baer and Banks (1985)
note, institutions with air conditioning systems include some filtration for
particulate matter, and often are equipped to remove SOy and/or ozone. Oxides
of nitrogen do not appear to be significantly attenuated by building ventila-
tion systems (National Research Council, 1986b). Nazaroff and Cass (1986), for
example, found levels of NO and NO^ in a California museum gallery to be
essentially the same as outdoors. The following summary, excerpted from
previous reviews, gives a general description of the effects of air pollution
on materials that may be found in association with indoor exposures.
Table 4-1 summarizes the types of damage that can be caused by indoor
exposure to air pollution. The important points to be brought out from
Table 4-1 are that a variety of materials are damaged by common air pollutants, that
sulfur oxides figure prominently as a pollutant type that can cause damage to
4-2
-------
TABLE 4-1. AIR POLLUTION EFFECTS ON MATERIALS
Materials
Metals
Type of
Damage
Corrosion,
tarnishing
Principal
Air Pollutants
Sulfur oxides and
other acid gases
Other
Environmental
Factors
Moisture, air, salt,
microorganisms, par-
ti cu late matter
Paint and
organic
coatings
Textiles
Texti1e
dyes
Paper
Magnetic
storage
media
Photographic
materials
Surface erosion,
discoloration,
soiling
Reduced tensile
strength, soil-
ing
Fading, color
change
Embrittlement,
soiling
Sulfur oxides, hydro-
gen sulfide, particu-
late matter
Sulfur oxides, nitro-
gen oxides, particu-
late matter
Nitrogen oxides,
ozone
Sulfur oxides,
particulate matter
Loss of signal Particulate matter
Microblemishes, Sulfur oxides,
"sulfiding" hydrogen sulfide
Moisture, sunlight,
ozone, microorganisms
Moisture, sunlight,
ozone, physical wear
Sunlight
Moisture, physical
wear
Moisture, heat, wear
Moisture, sunlight,
heat, other acid
gases, particulate
matter, ozone and
other oxidants
Rubber
Leather
Ceramics
Cracking
Weakening, pow-
dered surface
Changes sur-
face appearance
Ozone
Sulfur oxides
Acid gases, HF
Sunlight,
wear
Physical
Moisture,
organisms
physical
wear
micro-
Source: Yocom et al. (1982), Baer and Banks (1985), National Research Council
(1986b), Yocom et al. (1986), Murray et al. (1986).
many materials, and that air pollution is only one of the several factors that
can cause damage to materials exposed to indoor atmospheres. In order to adapt
research results produced for other purposes to an assessment of indoor air's
effects on materials, soiling, or odors, several data requirements specific to
the evaluation of welfare effects must be considered.
4-3
-------
Air quality measurement techniques designed for use in air pollution com-
pliance monitoring are not likely to provide the information most useful to the
assessment of effects on materials, because deposition measurements on surfaces
are more relevant. For instance, the use of sampling equipment designed to
represent the experience of the human lung inhaling air cannot also represent
the experience of a surface passively exposed to polluted atmospheres.
Sampling techniques designed to model inhalation could, however, be appropriate
for data gathered in connection with an assessment of malodorous pollutants.
Other aspects of data requirements in which welfare effect considerations
vary markedly from those of health effects have to do with the need for
measurements of such parameters as relative and/or absolute humidity and
temperature. Not only can deposition be strongly affected by such factors, but
most important effects on materials will not occur significantly except under
certain surface conditions. For example, surface moisture is a necessary
prerequisite for significant metal corrosion. The algorithm for exposure,
then, is not usually limited to concentration times time, but surface concen-
tration times the periods of time during which certain physical conditions have
been met. The recent introduction of electrochemical sensors has facilitated
meeting these data requirements in studies of metal corrosion, since they can
provide a continuous documentation of corrosion that can be compared with
time-dependent parameters (e.g., relative humidity, temperature, pollutant
concentration) recorded at the same site (Mansfeld, 1982).
The same conditions that accelerate material degradation by air pollution
also favor growth of biological agents that degrade and discolor, such as
bacteria and fungi (Solomon and Burge, 1984). The effects of the action of
these agents and of the air pollutants that are simultaneously present in an
indoor environment may be very difficult to separate.
The lack of any absolute threshold concentration associated with materials
damage, soiling, or odor is another important concept. In the case of materials
damage, not only must certain environmental conditions be met, but a critical
damage point must be reached before welfare can actually be said to be adversely
affected. Perception thresholds are critical to assessing the impact of either
soiling or odors. There is reason to believe that conditions specific to
indoor spaces may well affect these perceptions (e.g., Engen, 1986; Clausen et
al., 1986a,b).
4-4
-------
4.2 MATERIALS DAMAGE
4.2.1 Introduction
Several fundamental problems in quantifying the extent of damage to mate-
rials from specific pollutants whether exposed indoors or outdoors are listed
below.
Many types of damage associated with air pollutants (e.g.,
corrosion, erosion, fading) tend to occur also in unpolluted
atmospheres and therefore cannot be distinguished from those
caused or enhanced by the presence of air pollutants. Thus,
determining the isolated influence of a given pollutant is not a
straightforward process.
Laboratory studies in which individual pollutants are introduced
to susceptible materials in known exposures independent of other
influences tend to involve unrealistically high concentrations
and otherwise are not representative of real-life outdoor
exposure situations. This problem may be especially magnified
relative to indoor exposures, where, according to Nazaroff and
Cass (1986) "homogenous chemical reactions play an important
role in determining pollutant concentrations."
Changes made during recent years in the formulations of materials
and protective coatings as a result of technological improvements
can have a fundamental effect upon the nature and extent of air
pollution-induced damage to materials and associated damage
estimates.
Determination of the quantities of materials in place in rela-
tion to air pollutant exposures is a difficult task, especially
when data for significant materials (e.g., fabrics, paints and
paper) exposed to indoor environments may not have been collec-
ted in this context.
Data on deposition velocities for gases and particles to indoor
surfaces are quite limited. Available information indicates
that extrapolation from outdoor experience will not be valid
(Nazaroff and Cass, 1986).
The air quality criteria documents for particulate matter and sulfur
oxides (U.S. Environmental Protection Agency, 1982a), nitrogen oxides (U.S.
Environmental Protection Agency, 1982b) and ozone (U.S. Environmental Protec-
tion Agency, 1986a) contain thorough reviews of the literature on material
effects related to each of the pollutant classes. This chapter will rely
heavily upon the parts of those documents that deal with effects on materials
and soiling.
4-5
-------
In a recent review of factors contributing to degradation of materials in
the atmosphere, Graedel and McGill (1986) identified a number of air pollut-
ants that may be present in indoor environments at levels as high or higher
than those outdoors: N0?, HCHO, formic acid (HCOOH), S09, and ozone (Ck).
(~ £3
HCHO and HCOOH are of concern principally because of their effects on exposed
metals. Principal effects and levels associated with NO , SO , and 0,, as
XX o
identified in the EPA's air quality criteria documents for these pollutants,
are summarized below.
4.2.2 Oxides of Nitrogen
The damaging effects of atmospheric oxides of nitrogen have been estab-
lished for a variety of materials, including dyes, fibers, plastics, rubber,
and metals. Field exposures of cotton, viscose rayon, cellulose acetate, and
nylon fabrics colored with representative dyes demonstrate that fading occurs
for specific dyes in air containing NO^, 0,, and $62. These exposures were
carried out in ambient air and protected against sunlight. Chamber studies
using individual pollutants, M^, 0,, and S0?, have shown that some individual
dye-fiber combinations exhibit color fading only in response to NO,, exposure,
whereas others are susceptible to 03, as well as combinations of N0? and 0.,.
S02 introduced an accelerating effect. Disperse dyes used for cellulose
acetate and rayon include vulnerable anthraquinone blues and reds. The
cellulosic fibers, cotton and viscose rayon, dyed with certain widely used
direct dyes, vat dyes, and fiber reactive dyes, suffer severe fading on chamber
3
exposures to 940 ug/m (0.5 ppm) NOp under high humidity (90 percent) and high
temperature (32°F) conditions. Significant fading is observed on 12 weeks
3
exposure to 94 ug/m (0.05 ppm) N02 under high humidity and temperature
conditions (90 percent, 32°F).
o
Acid dyes on nylon fade on exposure to N02 at levels as low as 188 ug/m
(0.1 ppm) under similar conditions. Dyed polyester fabrics are highly resis-
tant to NOp-induced fading. However, permanent press fabrics of polyester
cotton and textured polyester exhibited unexpected fading when first marketed.
The fading was in the disperse dye, which migrated under high-heat conditions
of curing or heat-setting to the reactive medium of resins and other surface
additives.
The yellowing of white fabrics is documented for polyurethane-segmented
fibers (Lycra and Spandex), rubberized cotton, optically brightened acetate,
and nylon. Yellowing is also reported on fabrics finished with softeners or
4-6
-------
antistatic agents. NO,, was demonstrated to be the pollutant responsible for
color change, while 03 and SO,, showed no effect. Chamber studies using N02
concentrations of 376 ug/m (0.2 ppm) for 8 hours showed yellowing equivalent
to that on garments returned to manufacturers.
The tensile strength of fabrics may be adversely affected by the hydrolytic
action of acid aerosols. N02 has been demonstrated to oxidize the terminal
amine group (-NH?) of nylon to the degree that the fiber has less affinity for
acid-type dyes. Nylon 66 may suffer chain scission when exposed to 1,800 to
3
9,400 ug/m (1.0 to 5.0 ppm) N02. Field exposures of fibers have emphasized
the action of acids derived from SO,,, although N02 may also have been present
in high concentrations in urban sites and may well be present at high levels
indoors. Information on the contribution of N02 to degradation is incomplete.
4.2.3 Sulfur Oxides
Though physical damage functions have been developed for the effect of
S02 on a number of materials, even the most reliable damage functions must be
used with caution. More data are required to take into account orientation,
location, and design of materials in use. Those damage functions listed in
Table 4-2 were selected on the basis of their treatment of independent
variables and their inclusion in major literature reviews. Damage functions
vary in form, reflecting different parameters measured and methods of measure-
ment. Time-of-wetness (often expressed as relative humidity above a critical
value) is the most important variable in these damage functions.
Functions for zinc (Zn) or galvanized steel appear to show the best fit,
followed by the functions for oil-based house paint. The field studies by
Haynie and Upham (1970) and Haynie (1980) and the chamber study by Haynie et
al. (1976) incorporated critical variables and provided relatively reliable
damage functions for galvanized steel. The functions selected for weathering
and enameling steel and for oil-based paint also utilized these critical
environmental variables.
While the relationships between S02 exposure and effect are less well
established, effects of sulfur oxides on other materials, including paint,
fabrics, paper, and leather have been identified in the literature. These
effects are of particular concern in the indoor environment when cultural
properties such as archives and works of art are exposed to pollutants.
4-7
-------
TABLE 4-2. SELECTED PHYSICAL DAMAGE FUNCTIONS RELATED TO S02 EXPOSURE
Material
Reference
Exposure-Response Relationships
R2
CXI
Zinc
Galvanized steel
Galvanized steel
Haynie and Upham (1970)
Haynie et al. (1976)
Haynie (1980)
Oil-based house paint Spence et al. (1975)
Y = 0.001028 (RH - 48.8) S02
corr = (0.0187 S02+e41'85-23' 240/RT)t
corr = 2.32 t + 0.0134v0'781S02t
w w
Y = 14.3 + 0.0151 S02 + 0.388 RH
corr = depth of corrosion or erosion, urn
Y = corrosion/erosion rate, ug/yr
S02 - (jg/m3 S02
R = gas constant (1.9872 cal/gm mol K)
RH = percent annual average relative humidity
Note: 1 ppm S02 = 2620 |jg/m3.
Source: U.S. Environmental Protection Agency (1982a).
0.92
0.91
Not provided
by author
0.61
Enameling steel
Weathering steel
Haynie and Upham (1974) corr = 325 Vt e^-^/' ™2 ~ UO^/KM
Haynie et al. (1976) corr = [5.64 V$02 + e^55'44 " 31'150/RT
'J Not provided
by authors
^ Vf 0.91
w
f = fractional time of panel wetness
t = time of wetness in years
v = wind velocity in m/s
T = geometric mean temperature of panels when wet, K
t = time exposure, years
-------
4.2.4 Ozone
More than two decades of research show that 0^ damages certain nonbio-
logical materials; the amount of damage to actual in-use materials, however,
is poorly characterized. Knowledge of I/O 0, gradients, though expanded con-
siderably in recent years, has not been incorporated in materials damage
studies. Moreover, virtually all materials research on photochemical oxidants
has focused on (L.
The materials most studied in 03 research are elastomers and textile
fibers and dyes. Natural rubber and synthetic polymers of butadiene, isoprene,
and styrene, used in products like automobile tires and protective outdoor
electrical coverings, account for most of the elastomer production in the
United States. Little exposure to indoor environments is expected, then, for
these materials.
The effects of 0, on dyes have been known for nearly three decades. In
1955, Salvin and Walker exposed certain red and blue anthraquinone dyes to a
0.1 ppm concentration of 03 and noted fading, which until that time was thought
to be caused by NO^. Subsequent work by Schmitt (1960, 1962) confirmed the
fading action of 0, and the importance of relative humidity in the absorption
and reaction of 0^ with vulnerable dyes. The acceleration in fading of certain
dyes by high relative humidity was noted later by Beloin (1972, 1973) at an
Og concentration of 0.05 ppm and relative humidity of 90 percent. Kamath et
al. (1982) also found that a slight rise in relative humidity (85 to 90 percent)
caused a 20-percent dye loss in nylon fibers.
Both the type of dye and the material in which it is incorporated are im-
portant factors in a fabric's resistance to 03- Haynie et al. (1976) and
Upham et al. (1976) found no effects from 0, concentrations of 0.1 to 0.5 ppm
for 250 to 1000 hr under high and low relative humidity (90 versus 50 percent)
on royal blue rayon-acetate, red rayon-acetate, or plum cotton. On the other
hand, Haylock and Rush (1976, 1978) showed that anthraquinone dyes on nylon
fibers were sensitive to fading from 03 at a concentration of 0.2 ppm at 70
percent relative humidity and 40°C for 16 hr. Moreover, the same degree of
fading occurred in only 4 hr at 90 percent relative humidity. At higher
concentrations, there was a parallel increase in fading. Along with Heuvel et
al. (1978) and Salvin (1969), Haylock and Rush (1976, 1978) noted the
importance of surface area in relation to the degree of fading. In explaining
this relationship, Kamath et al. (1982) found that 0~ penetrated the fiber
4-9
-------
itself and caused most of the fading through subsequent diffusion to the
surface.
Field studies by Nipe (1981) and laboratory work by Kamath et al. (1982)
showed a positive association between 0^ levels and dye fading of nylon mate-
rials at an On concentration of 0.2 ppm and various relative humidities. In
summary, dye fading is a complex function of 0^ concentration, relative humid-
ity, and the presence of other gaseous pollutants. At present, the available
research is insufficient to quantify the amount of material damage attributable
to CL alone. Anthraquinone dyes incorporated into cotton and nylon fibers
appear to be the most sensitive to (L damage.
The degradation of fibers from exposure to 03 is poorly characterized.
In general, most synthetic fibers like modacrylic and polyester are relatively
resistant, whereas cotton, nylon, and acrylic fibers have greater but varying
sensitivities to the gas. 0^ reduces the breaking strength of these fibers,
and the degree of reduction depends on the amount of moisture present. Under
laboratory conditions, Bogaty et al. (1952) found a 20 percent loss in breaking
strength in cotton textiles under high-moisture conditions after exposure to a
0.06 ppm concentration of 03 for 50 days; they equated these conditions to a
500- to 600-day exposure under natural conditions. Kerr at al. (1969) found a
net loss of 9 percent in breaking strength of moist cotton fibers exposed to
DO at a concentration of 1.0 ppm for 60 days. The limited research in this
area indicates that 0~ in ambient air may have a minimal effect on textile
fibers, but additional research is needed to verify this conclusion.
The effects of ozone on paint are small in comparison with those of other
factors. Some studies (Yocom et al., 1986) of air pollutant effects on cultural
properties indicate that ozone may fade or alter pigments used in watercolors
and Japanese prints.
4.2.5 Particulate Matter
Though the most significant welfare effects of particles suspended in
indoor air are associated with soiling, (see Table 4-1 and Section 4.3, below),
there is considerable evidence that water-soluble salts have potential impact
in areas where electronic equipment is located. As electronic equipment is
becoming more common in both office environments and in homes, the significance
of this effect will be greater than at present. The electronics industry has
a special concern for these aerosol components, as hygroscopic elements of
4-10
-------
suspended participate matter can lead to corrosion, current leakage, and sub-
sequent failure of electronic equipment. In reporting the results of a study
to characterize the water-soluble components of size-fractionated aerosols
collected in an office building, Walker and Weschler (1980) suggest that when
electronic equipment is involved, the relative humidity of the indoor space
should be controlled at a level consistent with the deliquescent properties of
the salts.
4.3 SOILING
The previous discussion has dealt with the effects of air pollutants on
materials. In distinguishing between materials damage and soiling, we may say
that materials damage occurs when particle accumulation alters a surface in a
way that is not reversible by a cleaning operation. When the surface altera-
tion can be undone by cleaning, soiling has occurred. A flow chart for analysis
of factors contributing to soiling welfare effects is depicted in Figure 4-1.
Airborne particulate matter is deposited at some rate on a surface. The
resulting particle accumulation produces a physical effect that can be measured
by instrumental means. This physical effect at some level provokes a visual
response. At some level, the visual response is characterized as offensive.
This aesthetic effect at some level provokes a behavioral response either to
remove the offending particles from the affected surface, or to accept the
visual insult. The result of both behavioral responses—either utility lost or
resources expended—may be assigned a dollar value. The magnitude of the
economic effect depends upon the activity necessary to remove the actual
offense.
To evaluate soiling as a welfare effect of airborne particles, the rela-
tionship of airborne particles to physical effect and the physical effect's
relationship to the sensory effect must be established. Table 4-3 illustrates
the relationships between airborne particles and the various stages of particle
surface effects associated with soiling, as developed in literature cited in
the criteria document for sulfur oxides and particulate matter (U.S. Environ-
mental Protection Agency, 1982a). Thus, the vertical columns labeled physical,
sensory, aesthetic, and economic effects show the kind of consideration given
by the cited study to that particle surface effect. It is clear that no single
study has considered all of these effects and their interrelationships.
4-11
-------
PROBABLE INFLUENTIAL FACTORS
PARTICLE SIZE
WINDSPEED
SURFACE ROUGHNESS
SURFACE TEMPERATURE
SURFACE ORIENTATION
PARTICLE OPTICAL PROPERTIES
SURFACE OPTICAL PROPERTIES
CONTRAST
OBSERVER VISUAL ACUITY
OBSERVER COLOR PERCEPTION
OBSERVER DISTANCE FROM SOURCE
OBSERVER SOCIOECONOMIC STATUS
OBSERVER ATTITUDES (CULTURAL)
PURPOSE IN USE OF SURFACE
OBSERVER SOCIOECONOMIC STATUS
OBSERVER ATTITUDES
PURPOSE IN USE OF SURFACE
SEQUENCE OF EVENTS RELATING TO SOILING
YES
OPTICAL EFFECT
NO
PERCEIVED ACCUMULATION
NO
YES
AESTHETIC OFFENSE
YES
Y
ECONOMIC EFFECT
(RESOURCES EXPENDED)
\
ECONOMIC EFFECT
(UTILITY LOSS)
Figure 4-1. Factors contributing to soiling effects of deposited particles.
Source: Bradow et al. (1985).
4-12
-------
TABLE 4-3. SUMMARY OF LITERATURE ADDRESSING FACTORS CONTRIBUTING TO SOILING EFFECTS
Pollutants
Study Units (Term)
O'Connor (1913) Smoke, number of
days/year visible
smoke observed
Michelson and Air pollution, ug/m3
f* Tourin (1966)
t— '
oo
Michelson and Suspended particu-
Tourin (1967) lates, ug/m3
Michelson and Suspended parti cu-
Tourin (1968) lates, ug/m3
Pollutant Data Source
U.S. Weather Bureau
Consumer Reports
Not described; "some
. . .data obtained from
Maryland Department
of Health"
Connecticut State
Department of Health
Index Term, Particle Surface Effect
Physical Sensory Aesthetic
NAa NA "It is safe to
say that if it
were possible
to secure fig-
ures on 'the
disagreeable-
ness of it all'
it would equal
one- fourth of
of the total
of all the
items of cost."
NA NA
NA NA
NA NA NA
Addressed
Economic
$/yr incremental
cleaning costs
associated with
smoke
$/yr incremental
cleaning costs
associated with
parti culates
$/yr incremental
cleaning costs
associated with
particulates
$/yr incremental
cleaning cost
associated with
particulates
Comments
Related increased cleaning
and associated costs to
number of smoke days/year;
Incremental cost of
cleaning for Pittsburgh
area estimated at $9.9
million/year
Cleaning activity fre-
quency and cost data
gathered via survey;
Task frequency related
to particle concentration,
and multiplied by task
cost to estimate incre-
ment
(continued on following page)
-------
TABLE 4-3. (continued)
I
1—>
-p>
Pollutants Index Term, Particle Surface Effect Addressed
Study Units (Term) Pollutant Data Source Physical Sensory Aesthetic Economic
Carey (1959) Atmospheric dust, Calculated and from % area % area % area coverage NA
tons/mi 2/mo the Department of coverage coverage found offensive
Scientific and Indus- notice- by panel
trial Research (U.K.) able by
panel
Comments
% area of white
horizontal surface
covered by dust related
to ambient particles by
calculating deposition
velocity; all particles
contributing to effect
assumed "fine" (<30 um)
and produced from fuel
combustion Panel used to
ascertain % of area
coverage at which panel
a) observed deposit and
b) found deposit offen-
sive; assumed rainfall
every 10 days to accom-
plish outdoor cleansing;
dusting every four days
indoors
Narayanan and
Lancaster
(1973)
Esmen (1973)
Hancock et al.
(1976)
Dustfall,
tons/mi2/mo
Dustfall
Dustfall,
tons/mi2/mo
Newcastle City
(Australia)
Department of Health
NA
Calculated from %
effective area
coverage rate
% effec-
tive area
coverage
NA
NA
% effec-
tive area
coverage
detected
by panel
NA
NA
% effective area
coverage found
offensive by
panel
$/yr incremental
costs of house-
hold upkeep
NA
NA
Compared cleaning and
maintenance costs for
2 areas differing in
dustfall level
Proposed settled dust
photometer for measuring
soiling potential in
terms of percent area
per unit time
Used panels to identify
levels of dust coverage
that determine thresholds
of detection, discrimi-
nation, and identifica-
tion
(continued on following page)
-------
TABLE 4-3. (continued)
study
Booz, Allen
and Hamilton,
Inc. (1970)
Bel oi n and
Haynie (1975)
Pollutants
Units (Term)
Particulate matter,
ug/m3
Total suspended
particulate matter,
Pollutant Data Source
National Air Pollution
Control Administration
Jefferson Co. (AL)
Health Department
Index
Physical
NA
% change
in re-
Term, Particle
Sensory
NA
NA
Surface Effect
Aesthetic
Analysis
included "atti-
tudes toward
cleanliness" as
variables
NA
Addressed
Economic
Cleaning/main-
tenance task
performance
frequency
NA
Comments
Found though frequency of
cleaning in some cases
related to suspended par-
ticulate level, expendi-
tures for cleaning not
related; did not consider
value of homeowner labor
Building materials exposed
for two years at
ug/m
flectance
(change
in haze
for glass)
Watson and
Jaksch (1982)
Particulate matter,
ug/m3
Booz, Allen and
Hamilton
Used
Beloin
and Haynie
and Esmen
data.
NA
Used Booz,
Allen, and
Hamilton
$/year benefit
(welfare gain),
reduction of par-
ticulate level.
Birmingham site; regres-
sion analysis showed time
and particle concentration
significant variables for
reflectance change for
white surfaces; color was
additional significant
factor for brick; poor
correlations found for
concrete, limestone, and
window glass
Used portions of Booz,
Allen and Hamilton data
to construct demand curve
for household cleanliness;
values derived from
Esmen and Beloin and
Haynie used for various
supply curves; case
worked for Philadelphia
SMSA extrapolated to
estimate national welfare
benefit
NA = not addressed in developing relationship between surface particle effect and/or ambient particle concentration.
Source: Bradow et al. (1985).
-------
A necessary condition for soiling of surfaces is accumulation of a percep-
tible mass of contrasting particulate matter on a surface. This condition
involves three physical properties of the deposited particles. Atmospheric
particles can be characterized in terms of (1) deposition velocity, (2) concen-
tration, and (3) color. The rate that particles accumulate on surfaces is
related to deposition velocity and mass concentrations by the following
expression.
Accumulation Rate (ug/m2/sec) = Deposition Velocity (m/sec) x
Concentration (ug/m3)
Deposition to the several types and orientations of surfaces typically
encountered in the indoor environment has been studied very little, although
there is some evidence (Sinclair and Psota-Kelty, 1985) that suggests that
gravitational settling and Brownian diffusion are the dominant deposition
processes. Since the wind speeds are negligible and the turbulence low
indoors, extrapolation to the indoor condition for deposition velocities are
not likely to be correct.
Generally the substances present in atmospheric aerosols have not been
characterized by color, but probably the most intensely colored component of
atmospheric particulate material is black elemental carbon or soot. Stevens
et al. (1980) have found this material almost exclusively in the fine particle
mass, accounting for more than 80 percent of fine particle mass in their analy-
sis; other components, essentially colorless, were ammonium sulfates, nitrates,
organic carbon, and a very small amount of lead. Wolff et al. (1983) found
that elemental carbon concentration accounts for essentially all of the light
absorption in filtered particles. Thus, the most highly colored substance
present in atmospheric particles is present almost exclusively in combustion-
produced fine particles.
Any model of soiling and its effect must interrelate physical, perceptual,
and cost measures. As with any model, each measure must be on a common scale.
Because scales available to physics and perception or behavior are often dif-
ferent, it is important to identify the appropriate scale type, and beyond
that, to identify or at least consider the absolute threshold, the differential
threshold, and context for perception. The significance of these factors for
cost estimation can be illustrated. For example, if measures are extrapolated
4-16
-------
to the zero-zero origin to construct some linear model relating dirt deposited
to dollar cost, costs would be underestimated. Most physical and psychological
scales are not linearly related; in this case, the perceived dirt threshold
begins at a value greater than zero physical dirt. Such a scaling procedure
would produce an overly shallow slope for the prediction equation. A
similar difficulty occurs if perceived dirt and physical amounts of dirt are
assumed to be linearly correlated. These are almost surely related by a power
function having an exponent less than one, in which case extrapolation by a
model that extrapolates beyond data would produce prediction errors in the
other direction (Bradow et al., 1985).
The "rationale for the secondary standards" discussed as part of the pro-
posed NAAQS for particulate matter that appeared in EPA's Federal Register
notice of March 20, 1984, provided a good summary of the state of knowledge
regarding soiling effects on materials: "The available data base provides com-
pelling evidence that elevated levels of particulate matter can produce adverse
welfare effects, but provides little quantitative information on concentration-
effects relationships. Physical damage and economic studies tend to show no
obvious welfare effects "thresholds" for soiling... The available evidence
suggests that the public makes a distinction between concentrations at which
particulate pollution is noticeable and higher levels at which it is considered
a nuisance." The notice goes on to state that no studies have established
unique adverse particulate matter levels, however, and, on behalf of the EPA
Administrator, "seeks the guidance of knowledgeable State and local air
pollution officials with respect to levels at which the public appears to
consider particulate matter a nuisance."
4.4 ODORS
The importance of the role of perception threshold identified in the pre-
ceding discussion of soiling as a welfare effect of indoor air is even more
central to an evaluation of the effects on welfare of odors associated with the
indoor environment. As noted in the discussion, perception threshold is not a
property of the airborne particle. Similarly, the National Academy of Science
(MAS) Committee on Odors from Stationary and Mobile Sources noted that the
limit of detection, or odor threshold, is not a specific property of a sub-
stance, like its color or density. Instead, the threshold depends on the mode
4-17
-------
of presentation of the sample and on the sensitivity and even the expectation
of the recipient. When such factors are carefully controlled, reasonably
reproducible values can be obtained. A summary of olfactory thresholds
identified for major air pollutants is provided in Table 4-4.
Odor intensity (the magnitude of the perceived sensation) can be described
by an ordinal categorization, such as faint-moderate-strong. In more precise
methods, numbering systems are used to estimate the magnitude of one intensity
relative to another. One or more standard substances, in designated concen-
trations, may serve as references.
Attributes of odor other than detectability that are subject to measure-
ment include odor intensity, character (quality), and hedonic tone (pleasantness-
unpleasantness).
Odor character, or quality, is the property of the odor sensation that
permits one to distinguish odors of different substances on the basis of prior
exposure. Various systems of description have been proposed, and there have
been some unsuccessful attempts to categorize all odors in terms of a small
number of "primary" odor types.
The hedonic tone of an odor is the degree to which it is perceived as
pleasant or unpleasant. Such perceptions differ widely from person to person,
and adaptation is definitely a factor in tolerance. Furthermore, these judg-
ments are strongly influenced by the previous associations that a person brings
to the experience and by the emotional context in which the odor is perceived.
Hedonic tone can be measured in terms of preference (dislike very much, like
slightly, etc.), numbers, or pictorial references to facial expressions (smil-
ing, frowning, etc.). Another approach parallels the estimation of intensity
— odors can be numerically rated in accordance with the degree to which they
are more pleasant or unpleasant than other specified odors.
All these sensory methods require careful attention to the acquisition and
preservation of a representative sample of the atmosphere or emission of inter-
est and to the selection of appropriate human judges.
The chemical analysis of mixtures that contain many different chemical
components requires the acquisition of a representative sample and the separa-
tion and identification of the components. To relate such information to the
odor of the mixture, it is also necessary to determine which of the components
are odorous and to assess their contribution to the intensity and character of
the mixture. The analysis must be at least as sensitive as human olfaction.
4-18
-------
TABLE 4-4. MAJOR ODOROUS AIR POLLUTANTS, OLFACTORY THRESHOLDS, AND RELATED DATA
Category
Sulfur
compound
Nitrogen
Compounds
Class
Sulfur oxides
Sul fides
Mercaptans
Thioethers
Inorganic
Aliphatic amines
Aromatic nitro compounds
Heterocyclic amines
Systemic Chemical Name
Sulfur dioxide
Hydrogen sulfide
Carbon di sulfide
Methyl mercaptan
Ethyl mercaptan
Propyl mercaptan
Allyl mercaptan
Benzyl mercaptan
Dimethyl sulfide
Diethyl sulfide
Diallyl disulfide
Ammonia
Dimethyl ami ne
Trimethylamine
2,4,6-Trinitro-t-butyl-
xylene (musk)
Pyridine
Benzo[b]pyrrole (indole)
3-Methylindole (skatole)
Formula
S02
H2S
CS2
CH3SH
C2H5SH
C3H7SH
CHu=CHCH2SH
CeHgCH2SH
(CH3)2S
(C2HS)2S
(CH2=CHCH2S)2
NH3
(CH3)2NH
(CH3)3N
C(C4H9)(CH3)2(N02)3
C5H5N
C8H7N
CgHgN
Mol.
Wt.
64
34
76
48
62
76
74
124
62
90
146
17
45
59
297
79
117
131
Odor
Pungent
Rotten eggs
Rotten
Decayed cabbage
Decayed cabbage
Unpleasant
Garlic
Unpleasant
Decayed cabbage
Foul, garlic
Garlic
Pungent
Fishy
Fishy antebuccal
Musk
Empyreumatic
Fecal
Fecal
Odor Threshold,
ppm (by vol.)
0.47
0.0047-0.18
0.21-0.84
2 x 10"s-0.041
3 x 10"s-0.001
0.0016-0.024
0.003-0.017
0.0026-0.04
0.003
0.0048
1.1 x 10"4-0.012
0.47-54
0.047
0.00021
6 x 10"6-0.005
0.003-0.23
0.05
(continued on following page)
-------
TABLE 4-4. (continued)
Category Class Systemic Chemical Name
Cyanides Hydrogen cyanide
Allylisocyanide
Al ly 1 i sothi ocyanate
Selenium Selenides Hydrogen selenide
Compounds
Ethyl selenomercap tan
Diethyl selenide
.fi. Hydrocarbons, Aliphatic hydrocarbons 2-Butene (butylene)
' Alcohols, and
{•5 Oxygenates 2-Methylpropene
(isobutylene)
Phenol Phenol
Aldehydes Methanal (formaldehyde)
Ethanal (acetaldehyde)
Propenal (acrolein)
4-hydroxy-3-methoxy-
Ketones d-2-Keto-l,7,7,-tri-
methy 1 norcamphene
(camphor)
Formula
HCN
CH2=CHCH2NC
CH2=CHCH2SNC
H2Se
C2H5SeH
(C2Hs)2Se
CH3CH=CHCH3
CH2=C(CH3)2
C6H5OH
H2CO
CH3CHO
CH2=CHCHO
CgHgOs
CioHieO
Mol.
Wt.
27
67
99
81
109
137
56
56
94
30
44
56
152
152
Odor
Bitter almonds
Sweet repulsive
(nauseating)
Mustard oil (nose
and eye irritant)
Putrid
Foul, fetid
Putrid
(nauseating)
Gas-house
Gas-house
Empyreumatic
Pungent
Pungent
Burning fat
Sweet-aromatic
Aromatic-earthy
Odor Threshold,
ppm (by vol.)
0.9
0.18-1.6
0.008-0.42
4 x 10""-0.0012
4 x 10"4-0.0012
0.011
24
20
0.047
1.0
0.066-2.2
0.021-1.8
1.1 x 10"4-2 x 10"7
1.3
(continued on following page)
-------
TABLE 4-4. (continued)
Category
Hal ogen
Compounds
Miscellaneous
Class
Organic acids
Inorganic
Aliphatic halogens
Aromatic halogens
Oxygen
Systemic Chemical Name
Butanoic acid (butyric
acid)
2-Methylbutanoic acid
Butanediene (diacetyl)
Chlorine
Trichloroethylene
Triiodomethane (iodoform)
Benzyl chloride
Chlorohydroxybenzene
(chlorophenol)
Trioxygen (ozone)
Formula
CH3CH2CH2COOH
(CH3)2CHCH2COOH
(CH3CO)2
C12
CHC1=CC12
CHI3
f* H PH PI
C6C1«(OH)C1
03
Mol.
Wt.
88
102
86
71
131
394
126.5
128.5
48
Odor
Rancid,
perspiration
Body odor
Sweet butter
Pungent
Aromatic
Antiseptic
Aromatic
lacrimator
Medicinal
Irritating
Odor Threshold,
ppm (by vol.)
0.001-2.2
0.015
0.025
0.31
0.21
5 x 10~3
0.04-0.31
0.0036-0.03
0.51
Source: National Research Council (1979).
-------
Odors in combination tend to mask each other, so that the perceived
intensity is weaker than sum of the individual odors' intensities (National
Research Council, 1979). With regard to acceptable concentrations of odorous
compounds indoors, in "the prevailing rule in buildings equipped with mecha-
nical ventilation has been that the only acceptable odor is no odor at all"
(National Research Council, 1979).
4.5 ECONOMIC EFFECTS
In evaluating the significance of welfare effects of air pollution, cost-
benefit analysis has always been a major step in formulating regulatory deci-
sions. Figure 4-2 depicts the relationship between pollutant emissions and
economic damage. As shown, one may (1) proceed from ambient pollutant levels
to economic damage estimates directly or (2) estimate damage based on physical
damage functions. The latter route, called the damage function approach, has
been the preferred method, although more recent studies have employed the first
route. The estimation of willingness to pay is common to both choices.
Economic damage (benefit) that results from increased (decreased) pollu-
tant concentrations can be estimated by willingness-to-pay approaches. All
willingness-to-pay approaches try to estimate the aggregate monetary values
that all affected individuals assign to the effects of a change in pollutant
concentration. These approaches can be divided into three classes: damage
function approaches, nonmarket approaches, and indirect market approaches. The
first step of the damage function approach uses the relationship of pollutant
exposure to physical damage. The second step links the physical damage to a
dollar estimate of willingness to pay. Most economic damage estimates using
this approach have not considered substitution possibilities for producers or
consumers; however, proper consideration of these factors can yield good esti-
mates of willingness to pay (via the damage function approach). Nonmarket
approaches generally use surveys to ascertain the monetary values assigned to
the effects. Indirect market approaches use information about the demand for
marketed goods to estimate the willingness to pay for nonmarketed environmental
attributes that are closely related to the marketed good (e.g., property value
studies that estimate the willingness to pay for a change in the level of pol-
lutant concentration through analyses of the changes in price of residential
property). Each of these three approaches has different data requirements.
4-22
-------
PHYSICAL
AND
CHEMICAL
INTER-
ACTIONS
LEGISLATION
EXECUTIVE
ACTS
ECONOMIC
DAMAGE
FUNCTIONS
PHYSICAL
DAMAGE
FUNCTIONS
PROCESSES AND ACTIONS
RESULTANT PRODUCTS
Figure 4-2. Relationship among emissions, air quality, damages and benefits,
and policy decisions. Shaded area represents processes, actions, and resultant
products outside the scope of this chapter.
Source: U.S. Environmental Protection Agency (1982).
Some of the considerations that must be taken into account when the ana-
lyses are used in the public policy decision making process include:
(1) selection of only the most important materials with significant
damage as the subject of extensive damage cost analyses; those
chosen for analysis must be both susceptible to high damage
levels and represent major cost to society,
(2) when it is unlikely that a sufficiently significant reduction in
pollution levels can be achieved, in cases where the smallest
amount of damage is significant (such as in microelectronic
components), elaborate analysis to develop refined data is
unnecessary,
(3) the impact of pollution must be differentiated from the possibly
even greater effects, of natural environmental factors,
4-23
-------
(4) making assumptions that a given percentage of the total main-
tenance and replacement costs is attributable to pollution is a
serious flaw; Many materials wear out or are replaced for other
reasons before they are significantly damaged by pollution,
(5) the calculation of differences between avoidance and preventive
costs is very difficult, and
(6) with works of art and historical monuments, the complex and
aesthetic nature of their value to society renders cost/benefit
studies of a quantitative nature highly questionable.
4.5.1 Radon
In the case of indoor radon, the economic impact is associated with health
risks (real or perceived) rather than material damage. Since the government
does not anticipate regulating levels of indoor radon, it is the decision of
an individual homeowner as to whether or not to mitigate high radon levels.
However, the economic impact on the value of the property may be beyond the
homeowner's control. In areas where high indoor radon levels are prevalent,
potential buyers as well as lending institutions may insist on radon testing.
In some areas, lenders are currently requiring radon testing prior to closing
on a home mortgage. Homes with high radon levels may decrease in value even
if the levels have been reduced by remedial actions. Property values in an
entire region may decrease if the region is perceived as having a potential
radon problem. Once homeowners become aware of high radon levels, they may be
legally liable for future damages if they sell the house without disclosing
the problem. The potential economic effects of a radon scare on property
values are tremendous.
4-24
-------
5. SOURCE CATEGORIZATION
5.1 INTRODUCTION
The preceding chapters have summarized existing knowledge of indoor air
quality. However, because there has been relatively little research attention
given to the health risks of indoor air quality, the existing knowledge is
fragmented, having been collected primarily for other purposes. For example,
there is much known about NOp since it is regulated under the Clean Air Act,
but little is known about N0~ in complex mixtures with other existing pollu-
tants from the same sources. Indeed, some of these indoor sources of NO 2
emissions, are not characterized adequately for the multitude of other emis-
sions that comprise the mixture. Since much of the health information concern-
ing indoor pollutants is derived from animal studies and is highly pollutant-
specific, this information was summarized in Chapter 2 of this document by
pollutant category. With the notable exceptions of ETS and non-ionizing
radiation, most knowledge is pollutant-specific. In contrast, research needs
are source-specific since the source causes the potential health risks that
must be assessed, and the source or effects from the source are ultimately what
must be mitigated. To choose a chemical-specific approach to indoor air
quality research would, for the most part, be more expensive and time-
consuming in nature, and would still leave unresolved the very major problem of
how to assess the risks of complex mixtures from knowledge of a very few of the
components of that mixture. Therefore, this chapter has been written with the
goal of linking the chemical-specific approach of the Information Assessment
document a source-specific approach for establishing research needs.
5.2 ENVIRONMENTAL TOBACCO SMOKE
Thousands of compounds (~3800) which have been identified in either main-
stream (MS) or sidestream (SS) smoke are expected to be present in ETS. The
concentration and phase distribution (particulate or vapor phase) will usually
5-1
-------
be altered in ETS as compared to either MS or SS as a result of dilution in
indoor spaces. Undiluted SS contains a higher concentration of a number of
toxic and carcinogenic substances than MS. Included are nitrosamines, aromatic
amines, nicotine decomposition products, aldehydes, PAHs, CO, ammonia, NO , and
A
many other nitrogen containing aromatics (e.g., pyridines, analines, and
quinolines). ETS contains secondary reaction products not found in SS, and its
composition may differ physically and chemically due to dilution, aging, and
reactions with indoor surfaces such as walls, carpets and furnishings. Stan-
dard laboratory procedures have been established to characterize the properties
of SS and MS. Research is still needed to standardize both the collection and
evaluation of ETS so that the effects of ETS can be studied in laboratories and
in human populations (National Research Council, 1986).
The changes in phase distribution of ETS constituents as the smoke is
diluted and aged in the indoor environment are largely unknown. It is known,
for example, that almost all of the nicotine shifts from the particulate phase
in MS and fresh SS, to the vapor phase in ETS. Consequently, indoor air-
cleaning devices designed to remove particles will not substantially alter the
nicotine exposure, but may alter the concentrations of other toxic components.
The particulate phase of ETS is dominated by small particles (<2.5 microns)
referred to as RSP since they can be inhaled deeply into the lung. The exact
amount of ETS retained in the respiratory tract and body of the nonsmoker is
not known. Based on data from smokers, it is estimated that a nonsmoker
exposed to ETS can retain 0.014 to 1.6 mg of RSP per day from ETS (HiHer,
1984). Studies of smokers have led to the hypothesis that hydrophobia vapor
phase constituents of ETS (e.g., CO) are likely to enter the lung, while
hydrophilic vapor phase constituents (e.g., acetaldehyde) are likely to be
absorbed in the upper respiratory tract. Research is needed to determine
the distribution of constituents in the particulate and vapor phases of aged
ETS. Also, as discussed in chapter 3, the efficiency of air-cleaning systems
in removing the constituents needs to be studied (National Research Council,
1986a).
Indoor radon decays to short-lived radon daughters, which may become bound
to the RSP in ETS, thus affecting the human dosimetry of radon. Tobacco, how-
ever, contains some long-lived radon daughters. Research is needed to elucidate
the possible interactions between ETS and radon daughters, especially as radon
daughters can adhere to RSP and increase the potential hazard of ETS (National
Research Council, 1986a).
5-2
-------
Certain ETS constituents have been measured and used as surrogate indica-
tors of ETS exposure in both personal and indoor air monitoring studies.
Constituent compounds or classes which have been measured include RSP, CO,
nicotine, NO , acrolein, nitroso-compounds, and benzo(a)pyrene. Unfortunately,
X
many of these constituents are not good surrogates because they have a number
of sources in the indoor environment other than ETS. Although nicotine and RSP
have been particularly useful surrogates in the studies reported to date, no
single measure has met the criteria for an ideal ETS surrogate. To facilitate
health effects studies of ETS exposure, an ideal surrogate (marker or tracer)
of exposure to ETS should be unique (or nearly unique) to tobacco smoke, should
be present in sufficient quantities to measure at low ETS concentrations, and
should be emitted in a reasonably constant ratio across brands and types of
cigarettes to other tobacco smoke constituents of interest. Indoor air and
personal exposure monitoring has been handicapped by the lack of a clear
definition of the chemical and physical nature of ETS and the identification of
the target constituents of ETS associated with the health and comfort effects.
Reliable information needs to be obtained on the quantity, transport, and fate
of such chemicals in ordinary indoor environments (National Research Council,
1986a).
The substantial emission rate of RSP from tobacco smoke has led to the
conclusion that ETS is the dominant source of RSP indoors. For this reason a
majority of field and human studies have used RSP as an indicator of exposure
to ETS. In personal monitoring studies, total RSP has been found to be sub-
stantially elevated for individuals who report being exposed to ETS as compared
to those who report no such exposure. Air monitoring and modeling studies
both indicate that RSP levels will be clearly elevated over background levels
in indoor spaces when even low smoking rates occur. In future indoor air re-
search it is recommended that the importance of variation in the input para-
meters such as room size, temperature, humidity, air exchange rate, and numbers
of cigarettes smoked should be noted when interpreting the data on constituents
of ETS obtained from personal monitors and indoor space monitors (National
Research Council, 1986a).
Exposures to ETS have been assessed by questionnaires, air monitoring,
estimation through modeling of concentrations, and biological markers. The
simplest and yet the least precise and reliable method of exposure assessment
has been the use of questionnaires. Such questionnaires have been the basis
5-3
-------
for classifying individuals into broad categories of exposure, however there
are serious difficulties in developing uniform questions that elicit unambig-
uous and correct replies, and even more difficult problems in using these
replies to make quantitative estimates of exposure. Questionnaires are par-
ticularly difficult to use to estimate an integrated exposure over many years,
yet this is the primary method which has been used to approximate such long-
term exposures. The National Research Council (1986a) recommends that future
epidemiologic studies should incorporate into their design several of these
exposure assessment methods in order to assess exposure to ETS more accurately
and to allow estimation of dose. To estimate integrated exposure to ETS,
future studies need to estimate a long-term ETS exposure history, including
which fraction of the day is spent in the presence of ETS and at what ages
these exposures occurred. The data from such a history should be entered into
a specific time-place model, from which cumulative exposure can be estimated.
Personal exposure and dosimetry of ETS is dependent upon so many factors
that the optimal assessment should use biological markers that accurately
indicate uptake and/or dose in physiological fluids, tissues, or cells.
Several chemicals found in body fluids of active smokers have been evaluated as
biological markers of exposure to ETS, and the National Research Council (1986a)
recommends that other markers be investigated as described below. The criteria
for acceptable biological markers are similar to those for measuring ETS in the
external environment.
Nicotine and its metabolite, cotinine, measured in saliva, blood, or
urine, have been the most useful biological markers of recent ETS exposure,
because they are derived virtually exclusively from tobacco products. Urinary
cotinine levels have been shown to increase in nonsmokers with increasing
number of smokers in the home for all age groups. Currently there is diffi-
culty in interpreting the relative cotinine levels in nonsmokers compared to
smokers because of the reported slower clearance of cotinine in nonsmokers and
the lack of good uptake and clearance data for nonsmokers of different ages,
sex, and genetic backgrounds. The National Research Council (1986a) recommends
that absorption, metabolism, and excretion of ETS constituents, including
nicotine or cotinine, need to be carefully studied in order to evaluate whether
there are differences between smokers and nonsmokers in these factors. Further
epidemiologic studies using biological markers are needed to quantify exposure-
dose relationships in nonsmokers.
5-4
-------
Several other potential biological markers that have been evaluated as
indicators of ETS exposure are thiocyanate, COHb, and exhaled CO. These com-
pounds have not been found in sufficient quantities in body fluids at moderate
or low levels of ETS exposure to be generally useful. Since there are several
other sources of CO in the environment that equal or exceed the contribution
made by ETS, this marker is even less useful. Although the use of nicotine
and cotinine measurements in urine, and possibly saliva, are recommended as
the best methods now available for quantifying human exposure to ETS, these
are not ideal markers for all constituents of ETS. In ETS polluted environ-
ments, nicotine is currently thought to be present in the vapor phase as a
free base, thus its uptake by the passive smoker may not be representative of
the uptake of acidic and neutral smoke components from the vapor phase ndr of
any component in the particulate phase. Other suggested biological markers of
exposure include N-nitrosoproline, nitrosothioproline, and some of the aromatic
amines that are present in high concentrations in SS, as well as 3-vinylpyri-
dine, solanesol, and other tobacco specific constituents. Thus, the National
Research Council recommends that future studies should be concerned with
developing techniques to measure the uptake by nonsmokers of various other
types of tobacco-specific ETS components which would be representative of the
particulate organic phase of ETS and the volatile acidic and neutral phases.
Another type of biological marker of ETS exposure is genotoxicity of the
urine. The National Research Council (1986a) concluded that on the basis of
presently available data, it is likely that the exposure of nonsmokers to heavy
ETS increases the potential for mutagenic activity of their urine which is
elevated above that which is observed in the same nonsmokers before, and long
after, ETS exposure. The evaluation of mutagenicity in the urine of nonsmokers
as a result of ETS exposure must consider the possibility of confounding
factors such as dietary constituents, occupational exposures, and other envi-
ronmental exposure factors which may render the findings of elevated mutagen-
icity as nonspecific. Research is needed to clarify the appropriate methods
for estimating mutagenicity (particularly ETS-specific urinary mutagens) and
to isolate and identify the active agents in the body fluids of ETS-exposed
nonsmokers (National Research Council, 1986a).
Highly sensitive methods are now becoming available for determining
protein or DNA-adducts of environmental carcinogens and toxic agents in
circulating blood and tissues. Several constituents which occur in ETS, e.g.,
5-5
-------
benzo(a)pyrene and 4-aminobiphenyl, have been reported as hemoglobin or DNA
adducts, however these chemicals are not specific or unique to ETS. The
development and validation of methods to detect ETS-specific adducts would
provide an ideal marker of human exposure and in some cases (e.g., DNA-
adducts), dose of ETS. The NRC concludes that validation and quantitative
determination of the uptake of tobacco smoke carcinogens is urgently needed.
Studies are needed to develop and apply highly sensitive methods (e.g., immune-
assays or postlabelling) for measuring DNA and protein adducts of tobacco-
specific chemicals.
Evaluations recently completed by the NAS (National Research Council,
1986a) and the Surgeon General of the U.S. (1986) both concluded that consider-
ing the evidence as a whole, exposure to ETS increases the incidence of lung
cancer in nonsmokers. All of the studies from various countries, including the
United States, consistently show a 30 percent increased risk (within 95 percent
confidence intervals) for nonsmoking spouses of smokers. These estimates are
almost all derived from the comparison of persons identified as exposed or
unexposed on the basis of their spouses smoking habits determined by question-
naires. Both reports concluded that better data are needed on the extent and
variability of ETS exposure and the associated dosimetry in order to accurate-
ly estimate the number of deaths or magnitude of risk in the U.S. population.
Because more than 135,000 deaths from lung cancer are expected in the U.S. in
1986 alone, the Surgeon General's report (1986) concluded that the number of
lung cancer deaths attributable to ETS (passive smoking) is substantial and
represents a problem of sufficient magnitude to warrant substantial public
health concern. Both groups concur that laboratory studies are needed to
determine the concentrations of carcinogenic constituents of ETS present in
typical daily environments (e.g., indoor air). The use of biological markers
in epidemiologic studies is recommended to quantify more precisely the dose-
response relationships between ETS exposure and lung cancer occurrence.
Laboratory studies have contributed to a better understanding of the
factors and mechanisms involved in the induction of cancer by tobacco smoke.
There have been many bioassays conducted on MS (and often on cigarette smoke
condensate, CSC) including genotoxicity assays, cardnogenicity assays, and
other bioassays related to either tumor initiation or tumor promotion. Skin
tumorigenesis assays, although using a route of exposure different from the
human respiratory tract, have been especially useful in evaluating the
5-6
-------
carcinogenicity of different tobacco constituents. Similar studies with skin
painting have not been done with ETS and would be of value for assessing the
differential toxicity (e.g., tumor-initiating and tumor-promoting activity) of
ETS and MS (National Research Council, 1986a). Only one study has examined the
carcinogenic potential of the condensate of sidestream cigarette smoke (Wynder
and Hoffmann, 1967). The results of this study indicated that SS was more
tumorigenie than MS.
ETS exposure involves proportionately more exposure to gas-phase than to
particulate phase constituents when compared to MS exposure. There have not
been laboratory studies of the effects of exposure to ETS i_n vivo, and very few
studies have compared MS, SS, and ETS in short-term jjn vitro assays. Both
reports recommend that additional laboratory |n vivo and i_n vitro studies be
conducted to compare the mutagenicity and carcinogenicity of MS and SS, and
that further studies be initiated to evaluate the various constituents of ETS
(e.g., gas phase versus particles, specific fractions) of ETS.
There is no consistent evidence at this time of any increased risk of ETS
exposure for cancers other than lung cancer. Future epidemiologic studies need
to examine the potential risk for nonsmokers exposed to ETS for cancers such as
brain, hematopoietic, nasal sinus, or other cancers consistently related to
active smoking, and cancers of all sites. The NRC recommends that lymphohema-
topoietic neoplasms should be studied in relation to ETS exposure, particularly
as a result of childhood exposure and any potential interrelationship or
interaction with radon exposure.
Acute irritating effects of ETS, especially of the eyes, but also of the
nose and throat, are the most commonly reported effects. These effects have
been documented by studies in which individuals are asked questions regarding
odor or irritation or by measuring eye blink rates. The National Academy of
Sciences (National Research Council, 1986a) recommends that objective physio-
logical or biochemical indices should be sought to validate reports of noxious
reactions and chronic irritation associated with ETS. A number of studies
using standard dermatological tests have shown that tobacco smoke contains
immunogens, and individuals report allergic-like responses to exposure to ETS.
Further research is needed, however, to evaluate the medical importance of
positive skin tests to ETS extracts and to relate immune response on skin tests
to subjective complaints about the noxious, irritating properties of ETS
(National Research Council, 1986a). Although there is some evidence that it
5-7
-------
is the gaseous components of ETS that are the most irritating and objection-
able, more research is needed to determine the specific constituents that are
the irritants and/or allergens in ETS.
Respiratory effects, both acute and chronic, have been shown to be related
to exposure to ETS in children, particularly those exposed to ETS during the
first two years of life. These effects include pulmonary symptoms (e.g.,
wheezing, coughing, and sputum production) as well as respiratory infections
(e.g., manifested as bronchitis and pneumonia). Children whose parents smoke
have measurable effects on lung function; however, it is not clear whether
these children exposed to ETS will be at increased risk for the development of
chronic obstructive lung disease. Both the Surgeon General (1986) and the
National Research Council (1986a) recommend that ETS be eliminated from the
environments of small children. Because much of the evidence for adverse
respiratory effects of ETS on infants and young children is derived from
studies of children whose parents smoke, it is difficult to conclusively
demonstrate whether these effects result from indoor air exposure to ETS,
because the children were often exposed j_n utero as well. More research is
needed to clarify the respiratory effects of ETS exposure postnatally. Studies
are also needed to distinguish the effects of ETS on asthmatic and hyper-
responsive individuals. The National Research Council (1986a) recommends that
a combination of animal toxicological studies and human studies be used to
address these unresolved issues.
Although cardiovascular disease is associated with active smoking, both
the National Academy of Sciences and Surgeon General conclude that further
studies are needed to establish a causal relationship between either acute or
long-term ETS exposure and such disease. The connection between cardiovascular
disease indicators such as cardiac function, blood pressure, and angina in
nonsmokers and ETS exposure must be made before an effect of ETS on the etiol-
ogy of cardiovascular disease can be established.
Several other health effects of ETS have been studied in children, partic-
ularly effects on growth and development. In most of these studies it has not
been possible to differentiate the J_n utero exposure from the indoor air ETS
exposure during infancy and childhood. Effects on growth and development
need to be investigated before conclusions can be drawn about such effects from
ETS exposure. A number of studies consistently report that chronic middle ear
infections are more common in children whose parents smoke than in children of
5-8
-------
nonsmoking parents. The association between chronic middle ear infections and
ETS exposure has not yet been causally established, however.
5.3 BIOLOGICAL CONTAMINANTS
Biological contaminants such as viruses, bacteria, fungi, and protozoans
can originate as infectious agents shed by humans and animals. These same
groups of organisms, as well as insects, arachnids, and acarids, can originate
outdoors, but colonize indoor spaces under proper conditions of nutrition,
moisture, and temperature. Pollens, too, can enter indoor spaces through
cracks. Data concerning sources and health effects, as well as mitigation
efforts are detailed in Chapter 2.
At present, significant information about levels of pollution by biological
contaminants is not available. There is no baseline data of what constitutes
"normal" levels of flora, such as bacteria and fungi. Monitoring instruments
and techniques are primitive. Techniques for monitoring biologic contaminants,
especially viable agents, have not been standardized, so that comparison of
monitored levels by different researchers is not possible. Seasonal and
geographic patterns and variations in levels of organisms are not known. Many
fungi produce mycotoxins with very high toxic potentials that in many cases are
also carcinogenic. The conditions under which these toxins are produced and
the levels and kinds of toxins found indoors are generally not known, but some
neurotoxic symptoms from some mycotoxins mimic those of "sick building syn-
drome". Some bacteria produce endotoxins, for which the health effects in
indoor environments are not known.
The virulence of viable organisms varies tremendously, so that a single
organism can be infective for some diseases, while others require many thou-
sands of organisms per liter to be infective. Again, baseline data for viru-
lence of viable contaminants is not known.
Almost all biologic contaminants have allergenic potential. This includes
dander and excreta from humans and animals, as well as body parts and secre-
tions from insects, arachnids, and acarids. Pollens and fungal spores also
have allergenic potential. The numbers and sensitivities of people sensitized
to such agents is not well defined.
5-9
-------
5.4 PERSONAL ACTIVITIES
Pollutants generated from personal activities include N0?, particles, bio-
logical substances, and gas-phase organic compounds. Further descriptions of
these pollutants can be found in Chapter 2 of this document. Indoor NOp levels
are invariably high in the presence of smokers and cooking with gas relative to
outdoor values. Currently, integrated exposure to NCL can be measured by two
passive devices: the Palmes tube and the Yanagisawa badge, and by two types of
commercial personal monitors. Animal toxicology studies support the belief
that chronic exposure to N02 may result in chronic lung disease. Moreover, the
concentration of N0? has more of an effect on overall health than the duration
of exposure.
Particles and biological substances are generated by personal activities
such as vacuum cleaning and sweeping where the particles are reentrained from
floors and rugs. House dust, especially that collected in carpet fibers,
serves as an important sink for air pollutants, as well as soil particles
brought into the home by foot traffic. Extensive particulate sampling has been
conducted in the ambient environment, but generally these methods and samplers
are not well suited to indoor monitoring. A particulate exposure monitor
currently being developed by EPA maintains the integrity of organic chemicals
in house dust. There is insufficient knowledge of personal activity patterns
indoors to quantify the relationship between particulate matter exposure and
dose. Validated, standardized methods for biological aerosols are needed for
indoor use.
Gas-phase organic compounds originate as products from personal use acti-
vities involving hair spray, paint solvents, and cleaning fluids. The TEAM
studies conducted by the U.S. EPA utilized personal monitors with Tenax®
sampling cartridges and miniature air pumps to collect VOCs. Generally it is
thought that the levels of individual VOCs are well below threshold limit
values considered to be harmful for any compound.
The contribution of pollutants generated by personal activities to total
indoor exposure is largely unknown. Additional research is needed on determin-
ing personal activity patterns indoors, with subsequent measurement of spatial
and temporal gradients of pollutants associated with personal activities.
Moreover, existing measurement methodology may not be adequate to meet the
special demands of personal exposure monitoring. The development and valida-
tion of quiet, unobtrusive devices is needed to accurately assess exposure to
many indoor pollutants.
5-10
-------
5.5 BUILDING SYSTEMS
In spite of the recognized importance of the building system, there is
little information on building system performance. Most IAQ studies are either
limited to studies of sources or to individual building studies. These build-
ing studies clearly demonstrate the difficulties of diagnosing IAQ problems.
These studies also demonstrate that many IAQ problems are mixture problems.
That is, the problems are caused by a mixture of pollutants and by the inter-
actions of these mixtures with the building system.
The role the building system plays in determining IAQ must be better
defined. The research necessary to do this falls into six categories. These
are:
• ventilation and ventilation effectiveness
• component performance
• building system source and sink effects
• effects on sources and sinks
• effects of building systems on health effects
• integration of data from diverse sources into understanding of whole
system.
Each of these needs is discussed briefly below. Ventilation is a major
theme in all IAQ work. Increased ventilation is offered as a mitigation method
for nearly all problems, yet there is no clear understanding of how to make
maximum use of ventilation to improve IAQ. In many cases, gross building venti-
lation is adequate (i.e., standards are met), but IAQ is poor due to ineffective
utilization of the ventilation. Work is needed to develop measurement methods
to define ventilation and ventilation effectiveness. Research is also needed
to improve ventilation effectiveness.
A building system is made up of numerous components. The performance of
many of these components (e.g., air cleaners) is not well understood—especially
under conditions of interest in IAQ. Work to quantify the performance of major
components of the building system is essential.
The building system can act as a source or a sink for many pollutants, yet
there are no good data describing the role of the building system as a source
or a sink. Such data are essential to a clear understanding of IAQ. A coordi-
nated research effort to develop these data is necessary. This research program
5-11
-------
must be coordinated with other source type research to avoid overlap. Research
on various sources and sinks must be conducted to ensure that these effects
are quantified.
The building environment can also have a major impact on the health
effects from various pollutants. However, there is insufficient information on
the exact nature of these impacts. Thus, research is needed to define the
effects of building environment on health effects of various pollutants.
A full understanding of IAQ requires that the building system and the
various separate sources and sinks be treated as a whole. This integration of
the various pieces of data into a complete understanding of IAQ requires
research to develop models and simulators of buildings. The building simula-
tors should include methods to predict health impacts.
5.6 MATERIALS AND FURNISHINGS
Formaldehyde is the only major organic vapor for which data are adequate
to characterize sources and associated emission rates from building materials
and furnishings. Major uncertainties exist for other organics and their
sources. One area which has received little attention is that of transporta-
tion. Since the public spends significant time in motor vehicles and air-
planes, data on organic vapor sources in these environments is needed. These
uncertainties lead to the following research needs for source characterization.
Testing protocols are needed for determining emission rates from vehicle
sources via environmental test chambers, including consideration of tempera-
ture, relative humidity, air exchange, air exchange, product loading, chamber
vapor concentration, and chamber wall effects.
Emission factors are needed for a wide variety of building materials,
furnishings, household chemicals, and consumer products; such emission factors
should include consideration of environmental variables and should be appli-
cable for total organics and major organic species. The time variability of
emissions should also be examined, and materials and products used in transpor-
tation microenvironments should be included.
Source models should be developed which relate emission factors to indoor
concentrations; such models would be inputs to more general IAQ models discussed
elsewhere. Model testing and validation would be necessary in test house and
field studies.
5-12
-------
Material and product rankings should be developed based on their emission
characteristics (e.g., emission rates and compounds emitted); these rankings
could be coupled with health effects information to develop rankings based on
health hazards.
While the effects of ventilation on reducing indoor concentrations of
organic vapors by dilution and flushing are fairly well documented, the effects
of changing air exchange rates on source emission rates is unclear for many
sources and compounds. Emission rates for some compounds can actually increase
with increased ventilation. The effectiveness of building "bake out" to
achieve emission reductions has yet to be determined. Air cleaning devices
employing adsorption or catalytic oxidation are well understood for single
compounds at concentrations exceeding 100 to 1000 ppm, but performance data are
needed for such devices at lower concentrations (e.g., 1 to 100 ppb, or roughly
q
5 to 500 ug/m ), and for mixtures of organic vapors. The following research
needs for organic vapor control are suggested.
The influence of material and product age on emission rates should be
determined; such information would be developed via long-term testing in envi-
ronmental test chambers. The influence of ventilation practices on emission
rates for individual compounds needs to be determined. The effectiveness of
building "bake out" via environmental chamber testing at elevated temperatures
and air exchange rates should be evaluated; test house and field studies are
necessary for validation. Finally, the effectiveness of air cleaning devices,
including absorbers and catalytic oxidizers, should be determined.
5.7 COMBUSTION APPLIANCES
Combustion appliances such as gas stoves, heaters, and water heaters,
kerosene heaters, woodburning stoves, furnaces, and fireplaces are in common
use across the United States. All these sources involve combustion at some
stage and produce both indoor and outdoor pollution, the specific magnitude
being dependent on the degree of venting to the outside and/or the degree of
leakage indoors.
Indoor combustion of all organic material is incomplete, resulting in the
production of NO , CO, CO,, and a very broad class of substances called "prod-
s\ £.
ucts of incomplete combustion" (PICs). For some organic fuels, combustion
emissions are enriched in SOp (e.g., kerosene or high-sulfur gas, oil, or coal).
5-13
-------
It is not possible to assess adequately the health risks associated with
exposure to combustion emissions at the present time for several major reasons.
Exposure assessment to PICs is not yet possible. A significant number of
studies have monitored emissions from combustion appliances, especially gas
stoves and kerosene heaters, but the monitoring has focused on the criteria
pollutants within the emissions (i.e., NOp, CO, SOp, PM). Even so, there is
significant uncertainty of what the exposure patterns are on an hourly (or
other relatively short-term) basis. Some research has focused on PICs, since
several are carcinogenic, but this work has been relatively small in scope and
methods are not adequate presently to characterize the majority of PICs.
Dosimetry provides a linkage between exposure and hazard assessment, (i.e.-, the
relationship between exposure level and delivered dose which produces the
health effects), but it is insufficiently studied to be of impact in risk
assessment. Knowledge of dosimetry is reasonably advanced and/or part of an
ongoing effort for N0~ and for insoluble particles. However, insufficient
research attention is being focused on dosimetry of PICs, including the organ-
ics absorbed onto particles. Hazard assessment is likewise deficient. There
is substantial knowledge of the health effects of individual gaseous pollutants
such as NCL, CO, and 862, but even if there were adequate information on NOp
indoor exposure patterns, for example, the risk of exposure to the source would
be unknown due to toxicological interactions within complex mixtures. Current
health information on PM is specific to the outdoor environment. Of the known
PICs, many have been studied for mutagenicity and a few for carcinogenicity,
but many have not been studied at all and still more PICs have not even been
identified. A more specific discussion of these issues as related to indi-
vidual combustion appliances follows.
Of gas stoves and heaters, gas stoves have been the most frequently
studied. The research has focused on NOp, since relatively high levels of NO^
(compared to outdoors) are emitted by combustion appliances, and N0? is a NAAQS
pollutant. Numerous monitoring and epidemiological studies of gas stoves have
been completed (see Section 2.2.4.4). The results of the studies are contro-
versial, primarily due to the conflicting results which might be traceable to
the variation in methods and study approaches applied. On balance, these
studies do indicate that there is a health risk from cooking with unvented gas
stoves, but this health risk is inadequately characterized from both exposure
assessment and hazard assessment perspectives.
5-14
-------
Gas heaters have long posed an "accidental" health risk since high levels
of CO can be produced which have resulted in significant illness and even
death. These "accidental" risks have been known and mitigation procedures
established. However, risks from routine use are not adequately characterized.
Kerosene heaters have been more recently studied to determine emissions
and the mutagenic (and therefore possible carcinogenic) risks. Results indi-
cate that relatively high levels of SQ? are emitted, causing concern that the
pulmonary function of asthmatics might be affected. Organic fractions of the
emissions of a few of the kerosene heaters studied are mutagenic in HI vitro
bioassays. The information available suggests the possibility that some car-
cinogens such as nitro-PAHs and HCHO may be emitted under some combustion
conditions. There are several types of kerosene heaters: convective, radiant,
radiant/convective, and wickless, which are also categorized as blue flame and
white flame heaters. Further complicating risk assessment is the variation in
fuel type, wick height, length of operation, and design of the unit. Because
of these variables, there is a wide variety of exposure situations, none of
which have been adequately characterized. There is also a wide variety of
emitted pollutants, of which only a few have had health effects characterized.
Those that are known are sufficient to cause concern over health risks. The
major uncertainties for both carcinogenic and noncarcinogenic risks need to be
reduced to enable hazard assessment. Research to address these issues can
focus on specific alleviation or mitigation recommendations to be applied either
by engineers or homeowners. For example, if one fuel type is particularly
hazardous, a less toxic fuel might be used.
Wood burning has become increasingly common. For example, in the Pacific
Northwest, up to 50 percent of homes use wood fuel to meet some of their
heating needs. Woodburning appliances are vented to the outside, but there is
an indoor pollution component because even "airtight" wood heaters are a
statistically significant source of indoor PM and PAHs, several of which are
carcinogens or have carcinogenic potential. NO and CO, each having noncar-
/\
cinogenic health risks, are also emitted. Non-airtight combustion units
(especially fireplaces) contribute more to indoor pollution than the airtight
units. As is the case for other classes of combustion units, there is suffi-
cient evidence to be concerned about the health risks, especially carcinogenic
risks, for wood stoves and fireplaces, but concern alone does not enable risk
5-15
-------
assessment and does not advise manufacturers or the public about design and/or
use of units to mitigate that risk.
The increasing public use of combustion appliances is caused principally
by economic considerations, since use of electrical units with similar func-
tions is more expensive. There is a counterpart economic decision, namely the
impact of health risks, that needs to be incorporated in decision-making by the
public. It is also important to advise manufacturers about redesign options
that might reduce health risks. Generally, there is no doubt that there is a
health risk, the doubt resides in the characterization and degree of health
risk. Such uncertainties must be clarified to permit informed decision-making
by the public. To these ends, a research program has been proposed which
begins with a preliminary comparative assessment of health risks from major
types of combustion appliances. This approach produces information for pre-
liminary decision-making by the public and, equally important, provides guid-
ance as to whether further research is necessary and if so what its direction
should be.
5.8 OUTDOOR SOURCES
5.8.1 Radon
It is generally agreed that the principal source for high indoor radon
levels is the soil under the structure. The radon enters the building as part
of the soil gas which is driven by the pressure differential between the inside
of the building and the soil. In a few locations, the radon level may be
increased by the release of dissolved radon brought into the home in the
potable water supply. However, in the case of private water supplies (which is
the only category with a significant problem), less than three percent of the
average background indoor radon comes from the water. If potable water is
likely to be a significant source of radon, a number of mitigative techniques
have been proposed (Lawrence Berkeley Laboratory, 1982; U.S. Environmental
Protection Agency, 1984b). These include: 1) alternate water supplies, 2)
reconstruction of wells, 3) radon removal treatments, 4) aeration facilities,
and 5) the use of granular activated carbon beds.
In a few special instances, significant radon contributions come from
building materials. Avoidance of building materials made from uranium or
phosphate mining wastes is recommended.
5-16
-------
In order to develop procedures for diagnosing radon entry points and
factors affecting entry, both before and after radon reduction techniques have
been employed, it is necessary to carry out a comprehensive measurement pro-
gram in a variety of contaminated houses. Without careful measurements to
characterize the mechanisms which influence radon entry, the mitigation process
will remain an art form practiced only by a select group of building contrac-
tors rather than the science it could become. To illustrate that house mitiga-
tion is not yet a science, we need only point out that common mitigation
techniques which work well in some situations often fail in other apparently
similar situations, and the reasons for failure are not well understood. With
careful measurements it should be possible to understand why one application is
successful while another is not. In fact, with a little knowledge of the
driving mechanisms, the differences might prove to be predictable.
Several advances in controlling indoor radon concentrations have been made
in recent years. The EPA has done a significant amount of work on radon miti-
gation in existing homes and a guidance document for homeowners has been issued
(U.S. Environmental Protection Agency, 1986d). Some of the mitigation methods
under study include: 1) Natural and forced air ventilation, 2) avoidance of
structure depressurization, 3) sealing of potential soil gas entry routes, 4)
forced air ventilation with heat recovery, 5) drain-tile soil ventilation, 6)
hollow-block basement wall soil ventilation, and 7) sub-slab ventilation. Air-
cleaning systems that use air filtration or electrostatic precipitation have
also been studied. However, because their effect on the actual radon dose to
the lung is unclear, air-cleaning techniques are less desirable radon control
methods. Source reduction or increased ventilation are more favored means of
reducing indoor radon concentrations. The majority of the mitigation tech-
niques deal with reducing soil gas entry because this is known to be the most
significant radon source. .
Since soil gases are the major sources of indoor radon, there is a compel-
ling need to understand the mechanisms by which radon collects and migrates in
specific soils and geologic formations as well as the factors which are rate
controlling. The ability to predict the emanation characteristics from differ-
ent types of soils would be quite valuable in locating houses with very high
radon levels as well as locating potential new building sites. Unfortunately,
this predictive capability does not yet exist, but it should have priority on
the list of research needs.
5-17
-------
In many respects, designing a "radon-resistant" home should be much easier
than retrofitting an existing one against radon entry from soil sources.
Building a new house on low radon-emanating soils would be the obvious first
choice, however predicting the soil-emanation characteristics is not so easily
accomplished and significantly more work needs to be done toward this goal
(American ATCON, 1986). The knowledge which the EPA has gained by studying
existing houses is now being applied to the design of new houses. The choice
of construction methods for new homes in radon-prone areas should be given
careful consideration. For instance, hollow concrete blocks should not be used
for basement walls or in the foundation, and all cracks and openings should be
sealed. Gravel drainage beds with perforated drainage pipes should be in-
stalled on all four sides of the house and all drains should be sealed with a
water trap. It is also advisable to put good aggregate under any concrete
slab (American ATCON, 1986). An important aspect of EPA's continuing research
is identifying the most cost-effective radon-reduction methods for particular
structures and soil characteristics and transferring the technology to both the
private and the public sectors.
The major aspects of current research needs in the reduction of indoor
radon levels are summarized in the following statements. Development of proce-
dures for diagnosing radon entry points as well as pertinent factors affecting
its entry, both before and after the installation of reduction techniques, are
urgently needed. Development and demonstration of cost-effective methods for
reducing radon concentrations in existing houses are needed, as are development
and demonstration of cost-effective techniques to prevent high radon levels in
new houses. And finally, the transfer of technical information on the design,
installation, operation, performance, and cost of radon reduction methods and
equipment to the Federal, State, and Local governments, as well as to the
private and public sectors, is essential.
5.8.2 Pesticides
While present data on the occurrence and sources of indoor pesticides
is sparse, the ongoing NOPES investigation will provide substantial data.
Research is needed to determine emission rate characteristics, including:
1) development of testing protocols for determining pesticide emission rates
via environmental test chambers, 2) development of pesticide emission factors,
and 3) development of pesticide source models, including testing and validation.
5-18
-------
Only limited data are available on the effects of ventilation on indoor
pesticide concentrations. No data were found on the use of air cleaners to
control pesticide levels. Limited data are available on the influence of ap-
plication practices on indoor pesticide exposure. These data gaps lead to the
following research needs:
determining the influence of ventilation practices on pesticide
emissions to the indoor environment,
evaluating the impact of pesticide application practices on the
indoor residuals,
determining the applicability and effectiveness of air cleaners
for removing or destroying indoor pesticide vapors, and
evaluating radon reduction techniques for the control of indoor
chlordane levels (Chlordane, from soil treatment for termites,
has the same entry routes as soil radon; thus, techniques
presently being demonstrated for radon mitigation should be
effective for chlordane control).
5.9 CONCLUSION
The goal of reducing risk by reducing total exposure to air pollutants
can be approached by source control and mitigation. Table 5-1 indicates some
of the relationships between chemical-specific pollutants and their source-
specific origins. The Research Needs Document, which follows, addresses
indoor air pollution abatement from a source specific approach, as well as a
more generic approach which encompasses several source categories.
5-19
-------
TABLE 5-1. RELATIONSHIP BETWEEN POLLUTANT AND SOURCE CATEGORIES
Pollutants
Tobacco Smoke Components
Radon
Parti cul ate Matter
TSP
RSP
N0x
so!
CO
Oxidants
Pb
C02
en Metals
' Fibers
o Asbestos
Synthetics
Biological
Bacterial
Fungi
Plant Spores
Insects
Combustion Appliances
Passive Vented and Unvented
Smoking Al A2 A3 A4
X
0
X X
X X X X X
X X X X X
0 X X X X
0
X X X X X
X
0
0
Materials and Sources of
Furnishings Biological Personal Activities
Bl 62 B3 B4 Contamination Cl C2
X X
X XX
X
X
X
X
X X
X
XX XX
X
C3 C4 C5
0 0
0 X X
X
0 0
0 X X
0 0
0
0 0 X
0 X X
Outdoor
Environment
X
X
X
X
X
X
X
X
X
Pesticides
Organic Compounds (Non-ETS)
PAHs
Aldehydes
Alcohols
VOCs
CFCs
Hydrocarbons
Al - Gas Heaters
A2 - Kerosene Heaters
A3 - Wood/Coal Stoves and Fireplaces
A4 - Gas Cooking Stoves
X = Direct Source
Bl - Building Materials
B2 - Household Chemicals
83 - Furnishings
B4 - Stored Materials
Cl - Personal/Office Products
C2 - Transportation
C3 - Athletics
C4 - Maintenance (Cleaning, Painting)
C5 - Misc. (e.g., Photocopying)
-------
6. REFERENCES
Acheson, E. D. ; Gardner, M. J.; Pannett, B. ; Barnes, H. R.; Osmond, C.;
Taylor, C. P. (1984) Formaldehyde in the British chemical industry.
Lancet (8377): 611-616.
Ahmed, T. ; Marchette, B. ; Danta, I.; Birch, S. ; Dougherty, R. L.; Schreck, R.;
Sackner, M. A. (1982) Effect of 0.1 ppm N02 on bronchial reactivity in
normals and subjects with bronchial asthma. Am. Rev. Respir. Dis. 125:
152.
Ahmed, T.; Danta, I.; Dougherty, R. L.; Schreck, R.; Sackner, M. A. (1983)
Effect of N02 (0.1 ppm) on specific bronchial reactivity to ragweed
antigen in subjects with allergic asthma. Am. Rev. Respir. Dis. 127: 160.
Akiba, S.; Kato, H. ; Blot, W. J. (1986) Passive smoking and lung cancer among
Japanese women. Cancer Res. 46: 4804-4807.
Akland, G. G. ; Hartwell, T. D. ; Johnson, T. R. ; Whitmore, R. W. (1985)
Measuring human exposure to carbon monoxide in Washington, D.C., and
Denver, Colorado, during the winter of 1982-1983. Environ. Sci. Technol.
19: 911-918.
Albert, R. E.; Sellakumar, A. R. ; Laskin, S.; Kuschner, M.; Nelson, N.;
Snyder. C. A. (1982) Gaseous formaldehyde and hydrogen chloride induction
of nasal cancer in the rat. JNCI J. Natl. Cancer Inst. 68: 597-603.
Alfheim, I.; Ramdahl, T. (1984) Contribution of wood combustion to indoor air
pollution as measured by mutagenicity in Salmonella and polycyclic
aromatic hydrocarbon concentration. Environ. Mutagen. 6: 121-130.
Alzona, J. ; Cohen, B. L. ; Rudolph, H. ; Jow, H. N. ; Frohliger, J. 0. (1979)
Indoor-outdoor relationships for airborne particulate matter of outdoor
origin. Atmos. Environ. 13: 55-60.
Amdur, M. 0. (1960) The response of guinea pigs to inhalation of formaldehyde
and formic acid alone and with a sodium chloride aerosol. Int. J. Air
Pollut. 3: 201-220.
American ATCON. (1985) A review of potential low-cost mitigative measures for
soil-generated radon: 'phase I1 report draft. Report no. 1401/1334.
American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Inc. (1981) ASHRAE standard: ventilation for acceptable indoor air
quality. Atlanta, GA: The American Society of Heating, Refrigerating, and
Air-Conditioning Engineers, Inc.; ASHRAE report no. ASHRAE 62-1981.
6-1
-------
Ammann, H. A.; Berry, M. A.; Childs, N. E.; Mage, D. T. (1986) Health effects
associated with indoor air pollutants. In: Proceedings of the ASHRAE
conference IAQ '86: managing indoor air for health and energy conservation;
April; Atlanta, GA. Atlanta, GA: American Society of Heating, Refrigerating
and Air-Conditioning Engineers, Inc.; pp. 53-70.
Andelman, J. B. (1985) Inhalation exposure in the home to volatile organic
contaminants of drinking water. Sci. Total Environ. 47: 443-460.
Andersen, I.; Korsgaard, J. (1986) Asthma and the indoor environment:
assessment of the health implications of high indoor air humidity. In:
Berglund, B. ; Berglund, U. ; Lindvall, T.; Spengler, J.; Sundell, J., eds.
Indoor air quality: papers from the third international conference on
indoor air quality and climate; August 1984; Stockholm, Sweden. Environ.
Int. 12: 121-127.
Andersen, I.; Seedorff, L.; Skov, A. (1982) A strategy for reduction of toxic
indoor emissions. Environ. Int. 8: 11-16.
Anderson, E. W.; Andelman, R. J.; Strauch, J. M.; Fortuin, N. J. ; Knelson, J.
H. (1973) Effect of low-level carbon monoxide exposure on onset and
duration of angina pectoris: a study on 10 patients with ischemic heart
disease. Ann. Intern. Med. 79: 46-50.
Anderson, H. R. (1978) Respiratory abnormalities in Papua New Guinea children:
the effects of locality and domestic wood smoke pollution. Int. J.
Epidemiol. 7: 63-72.
Anderson, H. R. (1979) Respiratory abnormalities, smoking habits and
ventilatory capacity in a highland community in Papua New Guinea:
prevalence and effect on mortality. Int. J. Epidemiol. 8: 127-135.
Anger, W. K.; Johnson, B. L. (1985) Chemicals affecting behavior. In:
O'Donoghue, J. L., ed. Neurotoxicity of industrial and commercial
chemicals, v. I. Boca Raton, FL: CRC Press; pp. 51-148.
Aronow, W. S. (1978) Effect of passive smoking on angina pectoris. N. Engl. J.
Med. 299: 21-24.
Baer, N. S.; Banks, P. N. (1985) Indoor air pollution: effects on cultural and
historic materials. Int. J. Museum Manage. Curatorship 4: 9-20.
Bale, W. F. (1980) Memorandum to the files, March 14, 1951: hazards associated
with radon and thoron. Health Phys. 38: 1061-1066.
Bauer, M. A.; Utell, M. J. ; Morrow, P. E. ; Speers, D. M.; Gibb, F. R. (1984)
0.30 ppm nitrogen dioxide inhalation potentiates exercise-induced
bronchospasm in asthmatics. Am. Rev. Respir. Dis. 129: A151.
Bauer, M. A.; Utell, M. J.; Morrow, P. E.; Speers, D. M. ; Gibb, F. R. (1986)
Inhalation of 0.30 ppm nitrogen dioxide potentiates exercise-induced
bronchospasm in asthmatics. Am. Rev. Respir. Dis. 134: 1203-1208.
Beloin, N. J. (1972) A field study: fading of dyed fabrics by air pollution.
Text. Chem. Color. 4: 77-82.
6-2
-------
Beloin, N. J. (1973) A chamber study: fading of dyed fabrics exposed to air
pollutants. Text. Chem. Color. 5: 128-133.
Beloin, N. J.; Haynie, F. H. (1975) Soiling of building materials. J. Air
Pollut. Control Assoc. 25: 399-403.
Bender; Gothert, W. W.; Malorny, G. ; Sebbesse, P. (1971) Effects of low carbon
monoxide concentrations in man. Arch. Toxicol. 27: 142-158.
Benignus, V. A.
of carbon
1007-1014.
; Otto, D. A.
monoxide on
; Prah, J. D.; Benignus, G. (1977) Lack of effects
human vigilance. Percept. Mot. Skills 45:
Benson, F. B.; Henderson, J. J.; Caldwell, D. E. (1972) Indoor-outdoor air
pollution relationships: a literature review. Research Triangle Park, NC:
U. S. Environmental Protection Agency, National Environmental Research
Center; EPA publication no. AP-112. Available from: NTIS, Springfield,
VA; PB-214102.
Berglund, B.; Johansson, I.; Lindvall, T. (1982) The influence of ventilation
on indoor/outdoor air contaminants in an office building. Environ. Int.
8: 395-399.
Berglund, B.; Lindvall, T. ; Sundell, J. , eds. (1984a) Indoor air: proceedings
of the 3rd international conference on indoor air quality and climate, v.
2, radon, passive smoking, particulates and housing epidemiology; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research.
Berglund, B.; Lindvall, T.; Sundell, J. , eds. (1984b) Indoor air: proceedings
of the 3rd international conference on indoor air quality and climate, v.
3, sensory and hyperreactivity reactions to sick buildings; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research. Available from: NTIS, Springfield, VA; PB85-104206.
Berglund, B.; Lindvall, T. ; Sundell, J. , eds. (1984c) Indoor air: proceedings
of the 3rd international conference on indoor air quality and climate, v.
5, buildings, ventilation and thermal climate; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research. Available from:
NTIS, Springfield, VA; PB85-104222.
Berglund, B.; Berglund, U. ; Lindvall, T. (1986) Assessment of discomfort and
irritation from the indoor air. In: Proceedings of the ASHRAE conference
IAQ '86: managing indoor air for health and energy conservation; April;
Atlanta, GA. Atlanta, GA: American Society of Heating, Refrigerating and
Air-Conditioning Engineers, Inc.; pp. 138-149.
Bergman, H.; Edling, C. ; Axelson, 0. (1986) Indoor radon daughter
concentrations and passive smoking. In: Berglund, B. ; Berglund, U. ;
Lindvall, T. ; Spengler, J. ; Sundell, J., eds. Indoor air quality: papers
from the third international conference on indoor air quality and
climate; August 1984; Stockholm, Sweden. Environ. Int. 12: 17-19.
6-3
-------
Berkey, C. S. ; Ware, J. H.; Dockery, D. W.; Ferris, B. G., Jr.; Speizer, F. E.
(1986) Indoor air pollution and pulmonary function growth in preadolescent
children. Am. J. Epidemiol. 123: 250-260.
Berwick, M.; Zagraniski, R. T. ; Leaderer, B. P.; Stolwijk, J. A. J. (1984)
Respiratory illness in children exposed to unvented combustion sources.
In: Berglund, B.; Lindvall, T. ; Sundell, J., eds. Indoor air: proceedings
of the 3rd international conference on indoor air quality and climate, v.
2, radon, passive smoking, particulates and housing epidemiology; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 255-260.
Biller, W. F. et al. (1981) A general model for estimating exposure associated
with alternative NAAQS. Presented at: 74th annual meeting of the Air
Pollution Control Association; June; Philadelphia, PA. Pittsburgh, PA:
Air Pollution Control Association; paper no. 81-18.4.
Billings, C. E.; Vanderslice, S. F. (1982) Methods for control of indoor air
quality. Environ. Int. 8: 497-504.
Binder, R. E. ; Mitchell, C. A.; Hosein, H. R. ; Bouhuys, A. (1976) Importance
of the indoor environment in air pollution exposure. Arch. Environ.
Health 31: 277-279.
Binder, S. D.; Sokal, D.; Maughn, D. (1986) Estimating the amount of soil
ingested by young children through tracer elements. In: East Helena,
Montana child lead study, summer, 1983. Helena, MT: Montana Department of
Health and Environmental Sciences; app. 23.
Blackman, (1986) Presented at: Bioelectromagnetics Society annual meeting;
June; abstract K-l, p. 36.
Blackman, C. F. ; Benane, S. G.; Joines, W. T. (1985) Effects of ELF (1-120 Hz)
and modulated (50 Hz) RF fields on the efflux of calcium ions from brain
tissue in vitro. Bioelectromagnetics 6: 1-11.
Blackman, C. F.; Benane, S. G.; Rabinowitz, J. R.; House, D. E.; Joines, W. T.
(1986) A role for the magnetic field in the radiation-induced efflux of
calcium ions from brain tissue i_n vitro. Bioelectromagnetics 6: 327-337.
Blair, A.; Stewart, P.; O'Berg, M.; Gaffey, W.; Walrath, J.; Ward, J.; Bales,
R. ; Kaplan, S. ; Cubit, D. (1986) Mortality among industrial workers
exposed to formaldehyde. JNCI J. Natl. Cancer Inst. 76: 1071-1084.
Bland, M. ; Bewley, B. R.; Pollard, V.; Banks, M. H. (1978) Effect of
children's and parents' smoking on respiratory symptoms. Arch. Dis.
Child. 53: 100-105.
Bock, F. G. (1982) Nonsmokers and cigarette smoke: a modified perception of
risk. Science (Washington, DC) 215: 197.
Bogaty, H.; Campbell, K. S.; Appel, W. D. (1952) The oxidation of cellulose by
ozone in small concentrations. Text. Res. J. 22: 81-83.
6-4
-------
Booz, Allen and Hamilton, Inc. (1970) Study to determine residential soiling
costs of participate air pollution. Raleigh, NC: U. S. Department of
Health, Education and Welfare, National Air Pollution Control
Administration; EPA report no. APTD-0715. Available from: NTIS,
Springfield, VA; PB-205807.
Bradow, F. V. P.; Lockhead, G. R.; Bradow, R. L. (1985) Implications of
perception thresholds for soiling.
Brinton, L. A.; Blot, W. J. ; Fraumeni, J. F., Jr. (1985) Nasal cancer in the
textile and clothing industries. Br. J. Ind. Med. 42: 469-474.
Brunekreef, B.; Fischer, P.; Remijn, P.; van der Lende, R.; Schouten, J.;
Quanjer, P. (1985) Indoor air pollution and its effect on pulmonary
function of adult non-smoking women: III. Passive smoking and pulmonary
function. Int. J. Epidemiol. 14: 227-230.
Burchfiel, C. M., III. (1984) Passive smoking, respiratory symptoms, lung
function and initiation of smoking in Tecumseh, Michigan [Ph. D.
dissertation]. Detroit, MI: University of Michigan.
Burg, W. R.; Shotwell, 0. L.; Saltzman, B. E. (1981) Measurements of airborne
aflatoxins during the handling of contaminated corn. Am. Ind. Hyg. Assoc.
J. 42: 1-11.
Burge, H. A. (1985) Indoor sources for airborne microbes. In: Gammage, R. B. ;
Kaye, S. V., eds. Indoor air and human health: proceedings of the seventh
life sciences symposium; October 1984; Knoxville, TN. Chelsea, MI: Lewis
Publishing, Inc.; pp. 139-148.
Burge, H. A. (1986) Toxigenic potential of indoor microbial aerosols.
Presented at: Fifth symposium on the application of short-term bioassays
in the analysis of complex environmental mixtures; October; Durham, NC.
Burge, H. A.; Boise, J. R.; Rutherford, J. A.; Solomon, W. R. (1977)
Comparative recoveries of airborne fungus spores by viable and non-viable
modes of volumetric collection. Mycopathology 61: 27-33.
Burge, H. A.; Solomon, W. R. (1987) Sampling and analysis of biological
aerosols. In: Symposium on the characterization of contaminant emissions
from indoor sources; Chapel Hill, NC. Atmos. Environ. 21: 451-456.
Burkart, W. (1986) An estimation of radiation exposure and risk from
airtightening of homes in an alpine area with elevated radon source
strength. In: Berglund, B.; Berglund, U. ; Lindvall, T.; Spengler, J. ;
Sundell, J. , eds. Indoor air quality: papers from the third international
conference on indoor air quality and climate; August 1984; Stockholm,
Sweden. Environ. Int. 12: 49-53.
Bylin, G. ; Lindvall, T.; Rehn, T.; Sundin, B. (1985) Effects of short-term
exposure to ambient nitrogen dioxide concentrations on human bronchial
reactivity and lung function. Eur. J. Respir. Dis. 66: 205-217.
Carey, W. F. (1959) Atmospheric deposits in Britain - a study of dinginess.
Int. J. Air Pollut. 2: 1-26.
6-5
-------
Carroll, K. B.; Secombe, C. J. P.; Pepys, J. (1976) Asthma due to
non-occupational exposure to toluene (tolylene) di-isocyanate. Clin.
Allergy 6: 99-104.
Chan, W. C. ; Fung, S. C. (1982) Lung cancer in non-smokers in Hong Kong. In:
Grundmann, E., ed. Cancer campaign, v. 6, geographical pathology in
cancer epidemiology. Stuttgart, West Germany: Gustav Fischer Verlag; pp.
199-202.
Chan, W. C. ; Colbourne, M. J.; Fung, S. C.; Ho, H. C. (1979) Bronchial cancer
in Hong Kong 1976-1977. Br. J. Cancer 39: 182-192.
Chapman, R. S.; Calafiore, D. C. ; Hasselblad, V. (1985) Prevalence of
persistent cough and phlegm in young adults in relation to long-term
ambient sulfur oxide exposure. Am. Rev. Respir. Dis. 132: 261-267.
Charlton, A. (1984) Children's coughs related to parental smoking. Br. Med. J.
288: 1647-1649.
Chen, Y.; Li, W.; Yu, S. (1986) Influence of passive smoking on admissions for
respiratory illness in early childhood. Br. Med. J. 293: 303-306.
Christensen, C. L. ; Gliner, J. A.; Horvath, S. M.; Wagner, J. A. (1977)
Effects of three kinds of hypoxias on vigilance performance. Av. Sp. Env.
Med. 48: 491-496.
Chuang, J. C.; Hannan, S. W.; Wilson, N. K. (1987a) Environ. Sci. Technol. 21:
in press.
Chuang, J. C.; Mack, G. A.; Koetz, J. R.; Petersen, B. A. (1987b) Project
report. U. S. Environmental Protection Agency; EPA report no.
EPA-600/2-87-
Clausen, G. H.; Fanger, P. 0.; Cain, W. S. ; Leaderer, B. P. (1986a) Stability
of body odor in enclosed spaces. In: Berglund, B.; Berglund, U.;
Lindvall, T.; Spengler, J.; Sundell, J., eds. Indoor air quality: papers
from the third international conference on indoor air quality and
climate; August 1984; Stockholm, Sweden. Environ. Int. 12: 201-205.
Clausen, G. H.; Moller, S.; Fanger, P. 0.; Leaderer, B. P.; Dietz, R. (1986b)
Background odor caused by previous tobacco smoking. In: Proceedings of
the ASHRAE conference IAQ '86: managing indoor air for health and energy
conservation; April; Atlanta, GA. Atlanta, GA: American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc.; pp. 119-125.
Coburn, R. F. ; Forster, R. E.; Kane, P. B. (1965) Considerations of the
physiological variables that determine the blood carboxyhemoglobin
concentration in man. J. Clin. Invest. 44: 1899-1910.
Cohen, A. F.; Cohen, B. L. (1980) Protection from being indoors against
inhalation of suspended matter of outdoor origin. Atmos. Environ. 14:
183-184.
Cole, P.; Goldman, M. B. (1975) Occupation. In: Fraumeni, J. F. , Jr., ed.
Persons at high risk of cancer: an approach to cancer etiology and
6-6
-------
control: proceedings of a conference; December 1974; Key Biscayne, FL.
New York, NY: Academic Press; pp. 167-184.
Col ley, J. R. T. (1974) Respiratory symptoms in children and parental smoking
and phlegm production. Br. Med. J. 2: 201-204.
Collins, M. F. (1986) Characterization of the LTC catalyst: performance
against common air pollutants. Presented at: 79th annual meeting of the
Air Pollution Control Association; June; Minneapolis, MN. Pittsburgh, PA:
Air Pollution Control Association; paper no. 86-43.3.
Colome, S. D.; Spengler, J. D.; McCarthy, S. (1982) Comparison of elements and
inorganic compounds inside and outside of residences. Environ. Int. 8:
197-212.
Comstock, G. W.; Meyer, M. B.; Helsing, K. J.; Tockman, M. S. (1981) Respiratory
effects of household exposures to tobacco smoke and gas cooking. Am. Rev.
Respir. Dis. 124: 143-148.
Consumer Products Safety Commission. (1983) Summary report on testing of 17
kerosene heaters associated with consumer incidents. Engineering
Laboratory, Directorate for Engineering Sciences; ES DOC #543968.
Correa, P.; Fontham, E. ; Pickle, L. W.; Lin, Y.; Haenszel, W. (1983) Passive
smoking and lung cancer. Lancet (8350): 595-597.
Couch, R. B. (1981) Viruses and indoor air pollution. Bull. N. Y. Acad. Med.
57: 907-921.
Coutant, R. W. ; Lewis, R. G.; Mulik, J. (1985) Passive sampling devices with
reversible adsorption. Anal. Chem. 57: 219-223.
Coutant, R. W. ; Brown, L.; Chuang, J. C.; Riggin, R. M. ; Lewis, R. G. (1987)
Phase distribution and artifact formation in ambient air sampling for
polynuclear aromatic hydrocarbon. Atmos. Environ. 21: in press.
Cox, B. G.; DeWitt, D. S. ; Whitmore, R. W.; Singletary, H. M.; Howard, C. E.;
Starner, K. K. (1985) A survey design for a national study of indoor
radon concentrations in residential dwellings. Research Triangle Park,
NC: Exposure Monitoring System Laboratory; EPA contract no. 68-01-6826.
Available from: NTIS, Springfield, VA;
Dalager, N. A.; Pickle, L. W. ; Mason, T. J.; Correa, P.; Fontham, E.;
Stemhagen, A.; Buffler, P. A.; Ziegler, R. G.; Fraumeni, J. F. , Jr.
(1986) The relation of passive smoking to lung cancer. Cancer Res. 46:
4808-4811.
Dally, K. A.; Hanrahan, L. P.; Woodbury, M. A.; Kanarek, M. S. (1981)
Formaldehyde exposure in nonoccupational environments. Arch. Environ.
Health 36: 277-284.
Davidge, R. 0. C. (1986) ASHRAE standards: a guarantee of occupant satisfaction?
In: Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
6-7
-------
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 171-177.
Davidson, D. I. (1986) Pollutant emission factors for gas stoves: a literature
survey. EPA report no. EPA-600/x-xx-xxx.
Davies, J. E. (1972) The role of house dust in human DDT pollution. W. Indian
Med. J. 21: 172.
Day, J. (1986) UFFI-fungal interaction. In: Significance of fungi in indoor
air: report of a working group. Ontario, Canada: Health and Welfare
Canada Working Group on Fungi and Indoor Air.
De Bortoli, M.; Knoeppel, H. ; Pecchio, E.; Peil, A.; Rogora, L.; Schauenburg,
H.; Schlitt, H.; Vissers, H. (1984) Integrating "real life" measurements
of organic pollution in indoor and outdoor air of homes in northern
Italy. In: Berglund, B.; Lindvall, T. ; Sundell, J., eds. Indoor air:
proceedings of the 3rd international conference on indoor air quality and
climate, v. 4, chemical characterization and personal exposure; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 21-26. Available from: NTIS, Springfield, VA; PB85-104214.
De Bortoli, M.; Knoeppel, H. ; Pecchio, E. ; Peil, A.; Rogora, L.; Schauenburg,
H.; Schlitt, H. ; Vissers, H. (1986) Concentrations of selected organic
pollutants in indoor and outdoor air in northern Italy. In: Berglund, B.;
Berglund, L). ; Lindvall, T.; Spengler, J.; Sundell, J. , eds. Indoor air
quality: papers from the third international conference on indoor air
quality and climate; August 1984; Stockholm, Sweden. Environ. Int. 12:
343-350.
De Koning, H. W. ; Smith, K. R.; Last, J. M. (1984) Biomass fuel combustion and
health. World Health Organization; WHO EFP/84-64.
Dement, J. M. ; Smith, N. D. ; Hickey, J. L. S.; Williams, T. M. (1984) An
evaluation of formaldehyde sources, exposures and possible remedial
actions in two office environments. In: Berglund, B. ; Lindvall, T.;
Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 3, sensory and
hyperreactivity reactions to sick buildings; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 99-104.
Dobbs, A. J. ; Williams, N. (1983) Indoor air pollution from pesticides used in
wood remedial treatments. Environ. Pollut. Ser. B: 271-296.
Dockery, D. W.; Spengler, J. D. (1977) Personal exposure to respirable
particulates and sulfates versus ambient concentrations. Boston, MA:
Harvard School of Public Health.
Dockery, D. W.; Spengler, J. D. (1981) Personal exposure to respirable
particulates and sulfates. J. Air Pollut. Control Assoc. 31: 153-159.
Dockery, D. W.; Ware, J. H.; Ferris, B. G., Jr.; Speizer, F. E.; Cook, N. R.;
Herman, S. M. (1982) Change in pulmonary function in children associated
with air pollution episodes. J. Air Pollut. Control Assoc. 32: 937-942.
6-8
-------
Dodge, R. (1982) The effects of indoor pollution on Arizona children. Arch.
Environ. Health 37: 151-155.
Dodge, R. ; Solomon, P.; Moyers, J.; Hayes, C. (1985) A longitudinal study of
children exposed to sulfur oxides. Am. J. Epidemiol. 121: 720-736.
Dolan, M. C. (1985) Carbon monoxide poisoning. Can. Med. Assoc. J. 133:
392-399.
Dolimpio, D. A.; Jacobson, J. C.; Legator, M. (1968) Effect of aflatoxin on
human lymphocytes. Proc. Soc. Exp. Biol. Med. 127: 559.
Duan, N. (1982) Models for human exposure to air pollution. Environ. Int. 8:
305-309.
Ekblom, B. ; Huot, R. (1972) Response to submaximal and maximal exercise at
different levels of carboxyhemoglobin. Acta Physiol. Scand. 86: 474-482.
Ekwo, E. E.; Weinberger, M. M.; Lachenbruch, P. A.; Huntley, W. H. (1983)
Relationship of parental smoking and gas cooking to respiratory disease
in children. Chest 84: 662-668.
Engen, T. (1986) Perception of odor and irritation. In: Berglund, B.;
Berglund, U. ; Lindvall, T.; Spengler, J. ; Sundell, J. , eds. Indoor air
quality: papers from the third international conference on indoor air
quality and climate; August 1984; Stockholm, Sweden. Environ. Int. 12:
177-187.
Esmen, N. A. (1973) A direct measurement method for dustfall. J. Air Pollut.
Control Assoc. 23: 34-36.
Eudy, L. W.; Thome, F. A.; Heavner, D. L.; Green, C. R.; Ingebrethsen, B. J.
(1986) Studies on the vapor-particulate phase distribution of
environmental nicotine by selective trapping and detection methods.
Presented at: 79th annual meeting of the Air Pollution Control
Association; June; Minneapolis, MN. Pittsburgh, PA: Air Pollution Control
Association; paper no. 86-38.7.
Evans, R. D. (1969) Engineers' guide to the elementary behavior of radon
daughters. Health Phys. 17: 229-252.
Evans, R. D. ; Harley, J. H.; Jacobi, W.; McLean, A. S.; Mills, W. A.; Stewart,
C. G. (1981) Estimate of risk from environmental exposure to radon-222
and its decay products. Nature (London) 290: 98-100.
Fawcett, I. W. ; Newman Taylor, A. J.; Pepys, J. (1977) Asthma due to inhaled
chemical agents - epoxy resin systems containing phthalic acid anhydride,
trimellitic acid anhydride and triethylene tetramine. Clin. Allergy 7:
1-14.
Federal Register. (1980) Carbon monoxide; proposed revisions to the national
ambient air quality standards. F. R. (August 18) 45: 55066-55083.
Federal Register. (1984) Manufactured home construction and safety standards.
F. R. 49: 31996-32003, 32011-32012.
6-9
-------
Federal Register. (1985) Carbon monoxide and human health effects. F. R.
(September 13) 50: 37486-37487.
Ferahian, R. H. (1986) Contravention of Building Bylaws for HVAC systems and
bad maintenance as causes of indoor air pollution. In: Proceedings of the
ASHRAE conference IAQ '86: managing indoor air for health and energy
conservation; April; Atlanta, GA. Atlanta, GA: American Society of
Heating, Refrigerating and Air Conditioning Engineers, Inc.; pp. 251-257.
Fergusson, D. M.; Hons, B. A.; Norwood, L. J. (1985) Parental smoking and
respiratory illness during early childhood: a six-year longitudinal
study. Pediatr. Pulmonol. 1: 99-106.
Fine, D. H. ; Rounbehler, D. P. (1975) Trace analysis of volatile n-nitroso
compounds by combined gas chromatography and thermal energy analysis. J.
Chromatogr. 109: 271-279.
Fischer, P.; Remijn, B.; Brunekreef, B.; van der Lende, R.; Schouten, J. ;
Quanjer, P. (1985) Indoor air pollution and its effect on pulmonary
function of adult non-smoking women: II. Associations between nitrogen
dioxide and pulmonary function. Int. J. Epidemiol. 14: 221-226.
Fisk, W. J. , et al. (1985) Indoor air quality control techniques: a critical
review. LBL-16493; DOE/BP-00098-2.
Flachsbart, P. G.; Ott, W. R. (1984) Field surveys of carbon monoxide in
commercial settings using personal exposure monitors. Washington, DC: U.
S. Environmental Protection Agency, Office of Monitoring Systems and
Quality Assurance; EPA report no. EPA-600/4-84-019. Available from: NTIS,
Springfield, VA; PB84-211291.
Flensborg, E. W.; Samsoe-Jensen, T. (1950) Studies in mold allergy: 3. Mold
spore counts in Copenhagen. Acta Allergologica (Copenhagen) 3: 49-65.
Fletcher, R. A. (1984) A review of personal/portable monitors and samplers for
airborne particles. J. Air-Pollut. Control Assoc. 34: 1014-1016.
Florey, C. du V.; Melia, R. J. W.; Chinn, S.; Goldstein, B. D. ; Brooks, A. G.
F.; John, H. H.; Craighead, I. B.; Webster, X. (1979) The relation between
respiratory illness in primary school children and the use of gas for
cooking. III. Nitrogen dioxide, respiratory illness and lung infection.
Int. J. Epidemiol. 8: 347-353.
Friedman, M. A.; Staub, J. (1977) Induction of micronuclei in mouse and
hamster bone-marrow by chemical carcinogens. Mutat. Res. 43: 255-262.
Frigas, E.; Filley, W. V.; Reed, C. E. (1981) Asthma induced by dust from
urea-formaldehyde foam insulating material. Chest 79: 706-707.
Frigas, E.; Filley, W. V.; Reed, C. E. (1984) Bronchial challenge with
formaldehyde gas: lack of bronchoconstriction in 13 patients suspected of
having formaldehyde-induced asthma. Mayo Clin. Proc. 59: 295-299.
6-10
-------
Fugas, M. (1975) Assessment of total exposure to an air pollutant. Presented
at: International conference on environmental sensing and assessment, v.
2; September; Las Vegas, NV. New York, NY: Institute of Electrical and
Electronic Engineers; paper no. 38-5.
Gamble, J. ; Morey, P.; Richards, T.; Petersen, M. ; Castellan, R. M. (1986)
Building-related respiratory symptoms: problems in identification. In:
Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 16-30.
Gammage, R. B.; White, D. A.; Gupta, K. C. (1984) Residential measurements of
high volatility organics and their sources. In: Berglund, B. ; Lindvall,
T. ; Sundell, J., eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 4, chemical
characterization and personal exposure; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 157-162.
Available from: NTIS, Springfield, VA; PB85-104214.
Garfinkel, L. (1981) Time trends in lung cancer mortality among nonsmokers and
a note on passive smoking. JNCI J. Natl. Cancer Inst. 66: 1061-1066.
Garfinkel, L.; Auerbach, 0.; Joubert, L. (1985) Involuntary smoking and lung
cancer: a case-control study. JNCI J. Natl. Cancer Inst. 75: 463-469.
Garland, C.; Barrett-Connor, E. ; Suarez, L.; Criqui, M. H.; Wingard, D. L.
(1985) Effects of passive smoking on ischemic heart disease mortality of
nonsmokers. Am. J. Epidemiol. 121: 645-650.
Gas Research Institute. (1985) Characterization of emission sources into the
indoor environment. IITRI project CO 8675; pp. 138-144.
Gebefuegi, I.; Korte, F. (1984) Indoor contamination of household articles
through pentachlorophenol and lindane. In: Berglund, B.; Lindvall, T.;
Sundell, J., eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 4, chemical
characterization and personal exposure; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 317-322.
Available from: NTIS, Springfield, VA; PB85-104214.
Geisling, K. L. ; Tashima, M. K.; Girman, J. R.; Miksch, R. R. (1982) A passive
sampling device for determining formaldehyde in indoor air. Environ. Int.
8: 153-158.
GEOMET Technologies, Inc. (1976) The status of indoor air pollution research
1976. Gaithersburg, MD: U. S. Environmental Protection Agency; final
report for contract no. 68-02-2294.
Gillis, C. R.; Hole, D. J.; Hawthorne, V. M.; Boyle, P. (1984) The effects of
environmental tobacco smoke in two urban communities in the west of
Scotland. Eur. J. Respir. Dis. 65(Suppl. 133): 121-126.
6-11
-------
Ginsberg, M. D. (1985) Carbon monoxide intoxication: clinical features,
neuropathology and mechanisms of injury. J. Toxicol. Clin. Toxicol. 23:
281-288.
Girman, J. R. ; Hodgson, A. T. (1986) Source characterization and personal
exposure to methylene chloride from consumer products. Presented at: 79th
annual meeting of the Air Pollution Control Association; June;
Minneapolis, MM. Pittsburgh, PA: Air Pollution Control Association; paper
no. 86-52.7.
Girman, J. R.; Apte, M. G.; Traynor, G. W.; Allen, J. R.; Hollowell, C. D.
(1982) Pollutant emission rates from indoor combustion applicances and
sidestream cigarette smoke. Environ. Int. 8: 213-221.
Girman, J. R.; Allen, J. R.; Lee, A. Y. (1984a) A passive sampler for water
vapor. In: Berglund, B.; Lindvall, T.; Sundell, J. , eds. Indoor air:
proceedings of the 3rd international conference on indoor air quality and
climate, v. 4, chemical characterization and personal exposure; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 227-232. Available from: NTIS, Springfield, VA;
PB85-104214.
Girman, J. R.; Hodgson, A. T. ; Newton, A. S. ; Winkes, A. W. (1984b) Volatile
organic emissions from adhesives with indoor applications. In: Berglund,
B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 4,
chemical characterization and personal exposure; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
271-276. Available from: NTIS, Springfield, VA; PB85-104214.
Godish, T.; Rouch, J. (1985) An assessment of the Berge equation applied to
formaldehyde measurements under controlled conditions of temperature and
humidity in a mobile home. J. Air Pollut. Control Assoc. 35: 1186-1187.
Goldstein, B. D. ; Melia, R. J. W.; Chinn, S.; Florey, C. V.; Clark, D.; John,
H. H. (1979) The relations between respiratory illness in primary
schoolchildren and the use of gas for cooking: II - factors affecting
nitrogen dioxide levels in the home. Int. J. Epidemiol. 8: 339-345.
Goodman, R.; Henderson, A. S. (1986) Sine waves enhance cellular
transcription. Bioelectromagnetics 7: 23-29.
Gortmaker, S. L.; Walker, D. K. ; Jacobs, F. H.; Ruch-Ross, H. (1982) Parental
smoking and the risk of childhood asthma. Am. J. Public Health 72:
574-579.
Graedel, T. E.; McGill, R. (1986) Degradation of materials in the atmosphere:
common materials are vulnerable to atmospheric attack. Environ. Sci.
Technol. 20: 1093-1100.
Graedel, T. E.; Hawkins, D. T. ; Claxton, L. D. (1986) Atmospheric chemical
compounds: sources, occurrence, and bioassay. Orlando, FL: Academic
Press, Inc.
6-12
-------
Grant, L. D. (1984) Review of scientific studies on NOx available since CASAC
closure on the criteria document (EPA-600/8-82-026F) and OAQPS staff
paper [memo to Gerald E. Emison]. Research Triangle Park, NC: U. S.
Environmental Protection Agency, Environmental Criteria and Assessment
Office; November 13.
Grasso, P.; Sharratt, M. ; Davies, D. M.; Irvine, D. (1984) Neurophysiological
and psychological disorders and occupational exposure to organic
solvents. Food Chem. Toxicol. 22: 819-852.
Haider, M.; Groll-Knapp, E.; Hoeller, H. ; Neuberger, M.; Stidl, H. (1976)
Effects of moderate CO dose on the central nervous system -
electrophysiological and behaviour data and clinical relevance. In:
Finkel, A. J.; Duel, W. C. , eds. Clinical implications of air pollution
research, air pollution medical research conference, American Medical
Association; December 1974; San Francisco, CA. Acton, MA: Publishing
Sciences Group, Inc.; pp. 217-232.
Hammond, S. K.; Leaderer, B. P.; Roche, A. C.; Schenker, M. (1987) Collection
and analysis of nicotine as a marker for environmental tobacco smoke. In:
Symposium on the characterization of contaminant emissions from indoor
sources; Chapel Hill, NC. Atmos. Environ. 21: 457-462.
Hancock, R. P.; Esmen, N. A.; Furber, C. P. (1976) Visual response to
dustiness. J. Air Pollut. Control Assoc. 26: 54-57.
Harley, N. H. (1984) Comparing radon daughter dose: environmental versus
underground exposure. Radiat. Prot. Dosim. 7: 371-375.
Harrington, J. M.; Oakes, D. (1984) Mortality study of British pathologists
1974-80. Br. J. Ind. Med. 41: 188-191.
Harrington, J. M.; Shannon, H. S. (1975) Mortality study of pathologists and
medical laboratory technicians. Br. Med. J. 4(5991): 329-332.
Hasselblad, V.; Humble, C. G.; Graham, M. G. ; Anderson, H. S. (1981) Indoor
environmental determinants of lung function in children. Am. Rev. Respir.
Dis. 123: 479-485.
Hawley, G. G. (1981) The condensed chemical dictionary. 10th rev. ed. New
York, NY: Van Nostrand Reinhold Company.
Hawthorne, A. R.; Gammage, R. B.; Dudney, C. S. (1984) Effect of local geology
on indoor radon levels: a case study. In: Berglund, B.; Lindvall, T.;
Sundell, J., eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 2. radon, passive
smoking, particulates and housing epidemiology; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
137-148.
Haylock, J. C. ; Rush, J. L. (1976) Studies on the ozone fading of
anthraquinone dyes on nylon fibers. Text. Res. J. 46: 1-8.
6-13
-------
Haylock, J. C. ; Rush, J. L. (1978) Studies on the ozone fading of
anthraquinone dyes on nylon fibers. Part II: in-service performance.
Text. Res. J. 48: 143-149.
Haynie, F. H. (1980) Theoretical air pollution and climate effects on
materials confirmed by zinc corrosion data. In: Sereda, P. J. ; Livitan,
G. G. , eds. Durability of building materials and components: proceedings
of the 1st international conference; August 1978; Ottawa, Canada.
Philadelphia, PA: American Society for Testing and Materials; pp.
157-175. (ASTM special technical publication 691).
Haynie, F. H.; Upham, J. B. (1970) Effects of atmospheric sulfur dioxide on
the corrosion of zinc. Mater. Prot. Perform. 9: 35-40.
Haynie, F. H.; Upham, J. B. (1974) Correlation between corrosion behavior of
steel and atmospheric pollution data. In: Coburn, S. K., ed. Corrosion in
natural environments; Philadelphia, PA: American Society for Testing and
Materials; pp. 33-51. (ASTM special technical publication 558).
Haynie, F. H.; Spence, J. W. ; Upham, J. B. (1976) Effects of gaseous
pollutants on materials - a chamber study. Research Triangle Park, NC: U.
S. Environmental Protection Agency, Office of Research and Development;
EPA report no. EPA-600/3-76-015.
Hazucha, M. J. ; Ginsberg, J. F.; McDonnell, W. F.; Haak, E. D. , Jr.; Pimmel,
R. L. ; Salaam, S. A.; House, D. E.; Bromberg, P. A. (1983) Effects of 0.1
ppm nitrogen dioxide on airways of normal and asthmatic subjects. J.
Appl. Physiol.: Respir. Environ. Exercise Physio!. 54: 730-739.
Hedge, A.; Sterling, E. M.; Sterling, T. D. (1986) Building illness indices
based on questionnaire responses. In: Proceedings of the ASHRAE
conference IAQ '86: managing indoor air for health and energy
conservation; April; Atlanta, GA. Atlanta, GA: American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc.; pp. 31-43.
Helsing, K. J. ; Comstock, G. W. ; Meyer, M. B. ; Tockman, M. L. (1982)
Respiratory effects of household exposures to tobacco smoke and gas
cooking on nonsmokers. Environ. Int. 8: 365-370.
Hendrick, D. J. ; Lane, D. J. (1977) Occupational formalin asthma. Br. J. Ind.
Med. 34: 11-18.
Hernberg, S. ; Collan, Y. ; Degerth, R.; Englund, A.; Engzell, U.; Kuosma, E.;
Mutanen, P.; Nordlinder, H.; Hansen, H. S. ; Schultz-Larsen, K.; Sogaard,
H.; Westerholm, P. (1983) Nasal cancer and occupational exposures:
preliminary report of a joint Nordic case-referent study. Scand. J. Work
Environ. Health 9: 208-213.
Heuvel, H. M. ; Huisman, R. ; Schmidt, H. M. (1978) Ozone fading of disperse
blue 3 on nylon 6 fibers. The influence of physical fiber properties.
Text. Res. J. 48: 376-384.
Hirayama, T. (1981) Non-smoking wives of heavy smokers have a higher risk of
lung cancer: a study from Japan. Br. Med. J. 282: 183-185.
6-14
-------
Hirayama, T. (1984) Cancer mortality in nonsmoking women with smoking husbands
based on a large-scale cohort study in Japan. Prev. Med. 13: 680-690.
Hodgson, M. J.; Kreiss, K. (1986) Building-associated diseases: an update. In:
Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 1-15.
Hoek, G.; Brunekreef, B.; Meijer, R.; Scholten, A. (1984a) Indoor N02 and
respiratory symptoms of Rotterdam children. In: Berglund, B. ; Lindvall,
T.; Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 3, sensory and
hyperreactivity reactions to sick buildings; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 227-232.
Available from: NTIS, Springfield, VA; PB85-104206.
Hoek, G.; Brunekreef, B.; Meijer, R. ; Scholten, A.; Boleij, J. (1984b) Indoor
nitrogen dioxide pollution and respiratory symptoms of schoolchildren.
Int. Arch. Occup. Environ. Health 55: 79-86.
Hoek, G. ; Meijer, R. ; Scholten, A.; Noij, D. ; Lebret, E. (1984c) The
relationship between indoor nitrogen dioxide concentration levels and
personal exposure: a pilot study. Int. Arch. Occup. Environ. Health 55:
73-78.
Holaday, D. A. et al. (1957) Control of radon and daughters in uranium mines
and calculations on biologic effects. Washington, DC: Public Health
Service publication no. 494.
Holland, D. M. (1983) Carbon monoxide levels in microenvironment types of four
U.S. cities. Environ. Int. 9: 369-378.
Holub, R. F.; Droullard, R. F.; Borak, T. B.; Inkret, W. C.; Morse, J. G.;
Baxter, J. F. (1985) Radon-222 and 222Rn progeny concentrations measured
in an energy-efficient house equipped with a heat exchanger. Health Phys.
49: 267-277.
Honicky, R. E. ; Akpom, C. A.; Osborne, J. S. (1983) Infant respiratory illness
and indoor air pollution from a woodburning stove. Pediatrics 71:
126-128.
Honicky, R. E.; Osborne, J. S., III; Akpom, C. A. (1985) Symptoms of
respiratory illness in young children and the use of wood-burning stoves
for indoor heating. Pediatrics 75: 587-593.
Horstman, D.; Roger, L. J.; Kehrl, H. ; Hazucha, M. (1986) Airway sensitivity
of asthmatics to sulfur dioxide. Toxicol. Ind. Health 2:
Horstman, D. H.; Roger, L. J.; McDonnell, W. F.; Kehrl, H.; Seal, E.; Chapman,
R. ; Massaro, E. (1987) Pulmonary effects in asthmatics exposed to 0.30
ppm nitrogen dioxide during repeated exercise. Submitted to ARRD.
6-15
-------
Horwood, L. J.; Fergusson, D. M. ; Shannon, F. T. (1985) Social and familial
factors in the development of early childhood asthma. Pediatrics 75:
859-868.
Howe, G. R. ; Nair, R. C.; Newcombe, H. B.; Miller, A. B.; Abbatt, J. D. (1986)
Lung cancer mortality (1950-80) in relation to radon daughter exposure in
a cohort of workers at the Eldorado Beaverlodge uranium mine. JNCI J.
Natl. Cancer Inst. 77: 357-362.
Hytonen, S.; Alfheim, I.; Sorsa, M. (1983) Effect of emissions from
residential wood stoves on SCE induction in CHO cells. Mutat. Res. 118:
69-75.
Imbus, H. R. (1985) Clinical evaluation of patents with complaints related to
formaldehyde exposure. J. Allergy Clin. Immunol. 76: 831-840.
Imperato, P. J. (1981) Legionellosis and the indoor environment. Bull. N. Y.
Acad. Med. 57: 922-935.
International Agency for Research on Cancer. (1986) IARC monographs on the
evaluation of the carcinogenic risk of chemicals to humans, v. 38:
tobacco smoking. Lyon, France: World Health Organization.
Jackson, M. D. ; Lewis, R. G. (1981) Insecticide concentrations in air after
application of pest control strips. Bull. Environ. Contam. Toxicol. 27:
122-125.
Jackson, M. D.; Wright, C. G. (1975) Diazinon and chlorpyrifos residues in
food after insecticidal treatment in rooms. Bull. Environ. Contam.
Toxicol. 13: 593-595.
James, A. C. (1984) Dosimetric approaches to risk assessment for indoor
exposure to radon daughters. Radiat. Prot. Dosim. 7: 353-366.
Jarvis, B. B. (1986) Potential indoor air pollution problems associated with
macrocyclic trichothecene producing fungi. In: Significance of fungi in
indoor air: report of a working group. Ontario, Canada: Health and
Welfare Canada Working Group on Fungi and Indoor Air.
Jimenez-Diaz, C.; Ales, J. M. ; Ortiz, F.; Lahoz, F.; Garcia Puente, L. M. ;
Canto, G. (1960) The aetiologic role of molds in bronchial asthma. Acta
Allergol. Suppl. 7: 139-149.
Johnson, D. ; Billick, I.; Moschandreas, D.; Relwani, S. (1984) Emission rates
from unvented gas appliances. In: Berglund, B. ; Lindvall, T.; Sundell,
J., eds. Indoor air: proceedings of the 3rd international conference on
indoor air quality and climate, v. 4, chemical characterization and
personal exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish
Council for Building Research; pp. 367-373. Available from: NTIS,
Springfield, VA; PB85-104214.
Johnson, T. ; Paul, R. F. (1983) The NAAQS exposure model (NEM) applied to
carbon monoxide. Research Triangle Park, NC: U. S. Environmental
Protection Agency; EPA report no. EPA-450/5-83-003.
6-16
-------
Jones, J. R. ; Higgins, I. T. T.; Higgins, M. W.; Keller, J. B. (1983) Effects
of cooking fuels on lung function in nonsmoking women. Arch. Environ.
Health 38: 219-222.
Jurinski, N. B. (1984) The evaluation of chlordane and heptachlor vapor
concentrations within buildings treated for insect pest control. In:
Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of
the 3rd international conference on indoor air quality and climate, v. 4,
chemical characterization and personal exposure; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
51-56. Available from: NTIS, Springfield, VA; PB85-104214.
Kabat, G. C. ; Wynder, E. L. (1984) Lung cancer in nonsmokers. Cancer
(Philadelphia) 53: 1214-1221.
Kagawa, J. ; Tsura, K. (1979) Respiratory effects of 2-hour exposure to ozone
and nitrogen dioxide alone and in combination in normal subjects
performing intermittent exercise. Ninon Kyobu Shikkan Gakkai Zasshi 17:
765-774.
Kauffmann, F. ; Tessier, J.-F.; Oriol, P. (1983) Adult passive smoking in the
home environment: a risk factor for chronic airflow limitation. Am. J.
Epidemic!. 117: 269-280.
Keller, M. D.; Lanese, R. R.; Mitchell, R. I.; Cote, R. W. (1979a) Respiratory
illness in households using gas and electricity for cooking. I. Survey of
incidence. Environ. Res. 19: 495-503.
Keller, M. D.; Lanese, R. R.; Mitchell, R. I.; Cote, R. W. (1979b) Respiratory
illness in households using gas and electricity for cooking. II. Symptoms
and objective findings. Environ. Res. 19: 504-515.
Kentner, M.; Triebig, G.; Weltle, D. (1984) The influence of passive smoking
on pulmonary function—a study of 1,351 office workers. Prev. Med. 13:
656-669.
Kerns, W. 0.; Oonofrio, D. J.; Pavkov, K. L. (1983) The chronic effects of
formaldehyde inhalation in rats and mice: a preliminary report. In:
Gibson, J. E. , ed. Formaldehyde toxicity. New York, NY: Hemisphere
Publishing Corporation; pp. 111-131.
Kerr, N.; Morris, M. A.; Zeronian, S. H. (1969) The effect of ozone and
laundering on a vat-dyed cotton fabric. Am. Dyest. Rep. 58: 34-36.
Kilburn, K. H.; Seidman, B. C.; Warshaw, R. (1985) Neurobehavioral and
respiratory symptoms of formaldehyde and xylene exposure in histology
technicians. Arch. Environ. Health 40: 229-233.
Klein, J. P.; Forster, H. V.; Stewart, R. D.; Wu, A. (1980) Hemoglobin
affinity for oxygen during short-term exhaustive exercise. J. Appl.
Physiol. 48: 236-242.
6-17
-------
Kleinman, M. T. ; Bailey, R. M. ; Linn, W. S. ; Anderson, K. R.; Whynot, J. D.;
Shamoo, D. A.; Hackney, J. D. (1983) Effects of 0.2 ppm nitrogen dioxide
on pulmonary function and response to bronchoprovocation in asthmatics.
J. Toxicol. Environ. Health 12: 815-826.
Koo, L. C.; Ho, J. H.-C.; Saw, D. (1984) Is passive smoking an added risk
factor for lung cancer in Chinese women? J. Exp. Clin. Cancer Res. 3:
277-283.
Kossove, D. (1982) Smoke-filled rooms and lower respiratory disease in
infants. South Afr. Med. J. 61: 622-624.
Kreiss, K.; Gonzalez, M. G. ; Conright, K. L.; Scheere, A. R. (1982)
Respiratory irritation due to carpet shampoo: two outbreaks. Environ.
Int. 8: 337-341.
Lawrence Berkeley Laboratory. (1982) Control of formaldehyde and radon in the
built environment - a survey of the literature. U. S. Department of
Energy contract no. DE-AC03-76F00098.
Leaderer, B. P. (1986) Assessing human exposure to indoor air contaminants.
In: Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 208-212.
Leaderer, B. P.; Stolwijk, J. A. J.; Zagraniski, R. T.; Qing-Shan, M. (1984) A
field study of indoor air contaminant levels associated with unvented
combustion sources. Presented at: 77th annual meeting of the Air
Pollution Control Association; June; San Francisco, CA. Pittsburgh, PA:
Air Pollution Control Association; paper no. 84-33.3.
Lebowitz, M. D.; Corman, G.; O'Rourke, M. K.; Holberg, C. J. (1984)
Indoor-outdoor air pollution, allergen and meteorological monitoring in
an arid Southwest area. J. Air Pollut. Control Assoc. 34: 1035-1038.
Lebret, E.; van de Wiel, H. J.; Bos, H. P.; Noij, D.; Boleij, J. S. M. (1984)
Volatile hydrocarbons in Dutch homes. In: Berglund, B.; Lindvall, T.;
Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 4, chemical
characterization and personal exposure; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 169-174.
Available from: NTIS, Springfield, VA; PB85-104214.
Lebret, E.; van de Wiel, H. J.; Bos, H. P.; Noij, D. ; Boleij, J. S. M. (1986)
Volatile organic compounds in Dutch homes. In: Berglund, B.; Berglund,
U.; Lindvall, T.; Spengler, J.; Sundell, J. , eds. Indoor air quality:
papers from the third international conference on indoor air quality and
climate; August 1984; Stockholm, Sweden. Environ. Int. 12: 323-332.
Lee, S. D. ; Chen, C. W. ; Parker, J.; Anderson, L. D.; Bayliss, D. L.;
Cogliano, J.; Sakai, C.; Mohin, T. (1986) Health and risk assessment of
chloroform. Research Triangle Park, NC: U. S. Environmental Protection
Agency, Office of Health and Environmental Assessment.
6-18
-------
Leeder, S. R.; Corkhill, R.; Irwig, L. M.; Holland, W. W.; Col ley, J. R. T.
(1976) Influence of family factors on the incidence of lower respiratory
illness during the first year of life. Br. J. Prev. Soc. Med. 30:
203-212.
Leidy, R. B. ; Wright, C. G. ; Dupree, H. E., Jr. (1982) Concentration and
movement of diazinon in air. J. Environ. Sci. Health Part B 17: 311-319.
Leidy, R. B.; Wright, C. G.; Dupree, H. E., Jr.; MacLeod, K. E. (1984)
Concentration and movement of diazinon in air. II. vertical distribution
in rooms. J. Environ. Sci. Health Part B 19: 747-757.
Levin, H.; Hahn, J. (1986) Pentachlorophenol in indoor air: methods to reduce
airborne concentrations. In: Berglund, B.; Berglund, U.; Lindvall, T.;
Spengler, J.; Sundell, J. , eds. Indoor air quality: papers from the third
international conference on indoor air quality and climate; August 1984;
Stockholm, Sweden. Environ. Int. 12: 333-341.
Levine, R. J.; Andjelkovich, D. A.; Shaw, L. K. (1984) The mortality of
Ontario undertakers and a review of formaldehyde-related mortality
studies. JOM J. Occup. Med. 26: 740-746.
Lewis, R. G.; Jackson, M. D. (1982) Modification and evaluation of a
high-volume air sampler for pesticides and semi volatile industrial
organic chemicals. Anal. Chem. 54: 592-594.
Lewis, R. G.; MacLeod, K. E. (1982) Portable sampler for pesticides and
semi volatile industrial organic chemicals in air. Anal. Chem. 54:
310-315.
Lewis, R. G.; Mulik, J. D.; Coutant, R. W.; Wooten, G. W.; McMillin, C. R.
(1985) Thermally desorbable passive sampling device for volatile organic
chemicals in ambient air. Anal. Chem. 57: 214-219.
Lewis, R. G. ; Bond, A. E.; Fitz-Simons, T. R.; Johnson, D. E.; Hsu, J. P.
(1986) Monitoring for non-occupational exposure to pesticides in indoor
and personal respiratory air. Presented at: 79th annual meeting of the
Air Pollution Control Association; June; Minneapolis, MN. Pittsburgh, PA:
Air Pollution Control Association; paper no. 86-37.4.
Lewis, R. G. ; Bond, A. E.; Johnson, D. E.; Hsu, J. P. (1987) Measurement of
atmospheric concentrations of common household pesticides: a pilot study.
Environ. Sci. Technol. 21: in press.
Lewtas, J. (1985) Combustion emissions: characterization and comparison of
their mutagenic and carcinogenic activity. In: Stich, H. F. , ed.
Carcinogens and mutagens in the environment, v. IV: the workplace. Boca
Raton, FL: CRC Press; pp. 59-74.
Lewtas, J.; Goto, S.; Williams, K.; Chuang, J. C.; Petersen, B. A.; Wilson, N.
K. (1987) The mutagenicity of indoor air particles in a residential pilot
field study: application and evaluation of new methodologies. In:
Symposium on the characterization of contaminant emissions from indoor
sources; Chapel Hill, NC. Atmos. Environ. 21: 443-449.
6-19
-------
Lijinsky, W. ; Shubik, P. (1964) Benzo(a)pyrene and other polynuclear
hydrocarbons in charcoal-broiled meat. Science (Washington, DC) 145:
53-55.
Linn, W. S. ; Solomon, J. C. ; Trim, S. C.; Spier, C. E.; Shamoo, D. A.; Venet,
T. G.; Avol, E. L.; Hackney, J. D. (1985) Effects of exposure to 4 ppm
nitrogen dioxide in healthy and asthmatic volunteers. Arch. Environ.
Health 40: 234-239.
Lionel, T. ; Martin, R. J.; Brown, N. J. (1986) A comparative study of
combustion in kerosene heaters. Environ. Sci. Technol. 20: 78-85.
Lipfert, F. W. (1982) A national assessment of the air quality impacts of
residential firewood use. In: Proceedings of the residential wood and
coal combustion specialty conference; March; Louisville, KY. Pittsburgh,
PA: Air Pollution Control Association; pp. 226-239.
Livingston, J. M. ; Jones, C. R. (1981) Living area contamination by chlordane
used for termite treatment. Bull. Environ. Contam. Toxicol. 27: 406-411.
Love, G. J.; Lan, S.-P.; Shy, C. M.; Riggan, W. B. (1982) Acute respiratory
illness in families exposed to nitrogen dioxide ambient air pollution in
Chattanooga, Tennessee. Arch. Environ. Health 37: 75-80.
Luben, R. A.; Cain, C. D.; Chen, M. C.-Y.; Rosen, D. M.; Adey, W. R. (1982)
Effects of electromagnetic stimuli on bone and bone cells in vitro:
inhibition of responses to parathyroid hormone by low-energy
low-frequency fields. Proc. Natl. Acad. Sci. U. S. A. 79: 4180-4184.
Ludlin, F. D., Jr.; Wagoner, J. K. ; Archer, V. E. (1971) Radon daughter
exposure and respiratory cancer, quantitative and temporal aspects.
Washington, DC: Public Health Service; PHS joint monograph no. 1.
Mack, G. A.; Howes, J. E. ; Rodes, C. E. (1987) Carbon monoxide exposures of
Washington commuters. J. Air Pollut. Control Assoc. 37: in press.
Mannis, M. J.; Tamaru, R.; Roth, A. M.; Burns, M.; Thirkill, C. (1986)
Acanthamoeba sclerokeratitis: determining diagnostic criteria. Arch.
Ophthalmol. 104: 1313-1317.
Marsh, G. M. (1982) Proportional mortality patterns among chemical plant
workers exposed to formaldehyde. Br. J. Ind. Med. 39: 313-322.
Marsh, G. M. (1983) Mortality among workers from a plastics producing plant: a
matched case-control study nested in a retrospective cohort study. JOM J.
Occup. Med. 25: 219-230.
Master, K. M. (1974) Air pollution in New Guinea: cause of chronic pulmonary
disease among stone-age natives in the highlands. JAMA J. Am. Med. Assoc.
228: 1653-1655.
Matthews, T. G. (1986) Modeling and testing of formaldehyde emission
characteristics of pressed-wood products. Report XVIII to the U. S.
Consumer Product Safety Commission. Oak Ridge National Laboratory report
no. ORNL/TM-9867.
6-20
-------
Mazumdar, S.; Schimmel, H.; Higgins, I. T. T. (1982) Relation of daily
mortality to air pollution: an anlysis of 14 London winters,
1958/59-1971/2. Arch. Environ. Health 37: 213-220.
McClellan, R. 0.; Bice, D. E.; Cuddihy, R. G. ; Gillett, N. A.; Henderson, R.
F.; Jones, R. K.; Mauderly, J. L.; Pickrell, J. A.; Shami, S. G.; Wolff,
R. K. (1986) Health effects of diesel exhaust. In: Lee, S. D. ; Schneider,
T.; Grant, L. D. Verkerk, P. J., eds. Aerosols: research risk assessment
and control strategies. Chelsea, MI: Lewis Pub., Inc.; pp. 597-615.
McClenny, W. A.; Lumpkin, T. A.; Pleil, J. D. ; Oliver, K. D.; Bubacz, D. K.;
Faircloth, J. W. ; Daniels, W. H. (1986) Cannister-based VOC samplers. In:
Proceedings of the 1986 EPA/APCA symposium on measurement of toxic air
pollutants; April; Raleigh, NC; pp. 402-407.
McCurdy, T. (1987) Using exposure analyses in setting national ambient air
quality standards (NAAQS). Presented at: APCA international specialty
conference on regulatory approaches for control of air pollutants;
February; Atlanta, GA. Pittsburgh, PA: Air Pollution Control Association.
McFarland, R. A. (1973) Low level exposure to carbon monoxide and driving
performance. Arch. Environ. Health 27: 355-359.
McFarland R. A.; Roughton, F. J. W.; Halperin, M. H.; Niven, J. E. (1944) The
effects of carbon monoxide and altitude on visual thresholds. J. Aviat.
Med. 15: 381-394.
McKenzie, R. L.; Bright, D. S. ; Fletcher, R. A.; Hodgeson, J. A. (1982)
Development of a personal exposure monitor for two sizes of inhalable
particulates. Environ. Int. 8: 229-233.
McNall, P.; Walton, G.; Silberstein, S.; Axley, J. ; Ishiguro, K. ; Grot, R.;
Kusuda, T. (1985) Indoor air quality modeling: phase I report framework
for development of general models. Gaithersburg, MD: U. S. Department of
Commerce, National Bureau of Standards; report no. NBSIR 85-3265.
Melcher, R. G.; Garner, W. L. ; Severs, L W.; Vaccaro, J. R. (1978) Collection
of chlorpyrifos and other pesticides in air on chemically bonded
sorbents. Anal. Chem. 50: 251-255.
Melia, R. J. W.; Florey, C. du V.; Chinn, S. (1979) The relation between
respiratory illness in primary school children and the use of gas for
cooking. I - Results from a national survey. Int. J. Epidemic!. 8:
333-338.
Melia, R. J. W.; Florey, C. du V.; Morris, R. W.; Goldstein, B. D.; John, H.
H. ; Clark, D.; Craighead, I. B.; Mackinlay, J. C. (1982a) Childhood
respiratory illness and the home environment. II. Association between
respiratory illness and nitrogen dioxide, temperature and relative
humidity. Int. J. Epidemic!. 11: 164-169.
Melia, R. J. W. ; Florey, C. du V.; Morris, R. W.; Goldstein, B. D.; Clark, D.;
John, H. H. (1982b) Childhood respiratory illness and the home
environment. I. Relations between nitrogen dioxide, temperature and
relative humidity. Int. J. Epidemic!. 11: 155-163.
6-21
-------
Melia, J. ; Florey, C. ; Sittampalam, Y. ; Watkins, C. (1983) The relation
between respiratory illness in infants and gas cooking in the UK: a
preliminary report. In: Air quality Vlth world congress: [proceedings of
the International Union of Air Pollution Prevention Association]; May;
Paris, France. Paris, France: SEPIC (APPA); pp. 263-269.
Melius, J. ; Wallingford, K.; Carpenter, J.; Keenlyside, R. (1984) Indoor air
quality: the NIOSH experience (evaluation of environmental office
problems). Am. Conf. Ind. Hyg. Report 10: 3-7.
Michelson, I.; Tourin, B. (1966) Comparative method for studying costs of air
pollution. Public Health Rep. 81: 505-511.
Michelson, I.; Tourin, B. (1967) Report on study of validity of extension of
economic effects of air pollution damage from upper Ohio river valley to
Washington, DC area. Environmental Health and Safety Research Associates.
Michelson, I.; Tourin, B. (1968) The household cost of air pollution in
Connecticut. Report to the Connecticut State Department of Health; New
Rochelle, NY: Environmental Health and Safety Research Associates.
Miller, 0. P. (1987) Analysis of nitrite in N02 diffusion tubes using ion
chromatography. U. S. Environmental Protection Agency; EPA report no.
EPA-600/2-87-
Miller, M. M. ; Patterson, R. ; Fink, J. N. ; Roberts, M. (1976) Chronic
hypersensitivity lung disease with recurrent episodes of hypersensitivity
pneumonitis due to a contaminated central humidifer. Clin. Allergy 6:
451-462.
Molhave, L. (1982) Indoor air pollution due to organic gases and vapours of
solvents in building materials. Environ. Int. 8: 117-127.
Molhave, L. (1985) Volatile organic compounds as indoor air pollutants. In:
Gammage, R. B. ; Kaye, S. V., eds. Indoor air and human health:
proceedings of the seventh life sciences symposium; October 1984;
Knoxville, TN. Chelsea, MI: Lewis Publishing, Inc.; pp. 403-413.
Molhave, L. ; Bach, B. ; Pedersen, 0. F. (1984) Human reactions during
controlled exposures to low concentrations of organic gases and vapors
known as normal indoor air pollutants. In: Berglund, B.; Lindvall, T.;
Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 3, sensory and
hyperreactivity reactions to sick buildings; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 431-436.
Available from: NTIS, Springfield, VA; PB85-104206.
Molhave, L. ; Bach, B.; Pedersen, 0. F. (1986) Human reactions to low
concentrations of volatile organic compounds. In: Berglund, B.; Berglund,
U.; Lindvall, T.; Spengler, J. ; Sundell, J. , eds. Indoor air quality:
papers from the third international conference on indoor air quality and
climate; August 1984; Stockholm, Sweden. Environ. Int. 12: 167-175.
6-22
-------
Morey, P. R. ; Hodgson, M. J. ; Sorenson, W. G. ; Kullman, G. J.; Rhodes, W. W.;
Visvesvara, G. S. (1984) Environmental studies in moldy office buildings:
biological agents, sources and preventive measures. Ann. Am. Conf. Gov.
Ind. Hyg. 10: 21-35.
Morey, P. R. ; Hodgson, M. J. ; Sorenson, W. G. ; Kullman, G. J. ; Rhodes, W. W. ;
Visvesvara, G. S. (1986) Environmental studies in moldy office buildings.
ASHRAE Trans. 92: 399-419.
Morin, R. S. (1985) The bacterial mutagenicity of selected indoor air
particulate emissions arising from combustion processes [doctoral].
Chapel Hill, NC: University of North Carolina.
Moschandreas, D. J. ; Pelton, D. J. ; Berg, D. R. (1981a) The effects of
woodburning on indoor pollutant concentrations. Presented at: 74th annual
meeting of the Air Pollution Control Association; June; Philadelphia, PA.
Pittsburgh, PA: Air Pollution Control Association; paper no. 81-22.2.
Moschandreas, D. J. ; Zabransky, J. ; Pelton, D. N. (1981b) Comparison of
indoor-outdoor pollution. Electric Power Research Institute; EPRI report
no. EPRI-EA-1733.
Muller, J. ; Wheeler, J. F.; Gentleman, G. S.; Kusiak, R. A. (1983) Study of
mortality of Ontario miners: 1955-1977, part 1. Toronto, Canada: Ontario
Ministry of Labour.
Mumford, J. L. ; He, X. Z. ; Chapman, R. S.; Cao, S. R.; Harris, D. B.; Li, X.
M.; Xian, Y. L.; Jiang, W. Z.; Xu, C. W.; Chuang, J. C.; Wilson, W. E. ;
Cooke, M. (1987) Lung cancer and indoor air pollution in Xuan Wei, China.
Science (Washington, DC) 235: 217-235.
Murray, D. R. ; Atwater, M. A.; Yocom, J. E. (1986) Assessment of the costs of
materials damage from air pollution in Los Angeles, California. Presented
at: 79th annual meeting of the Air Pollution Control Association; June;
Minneapolis, MN. Pittsburgh, PA: Air Pollution Control Association; paper
no. 86-85.9.
Mustafa, M. G. ; Elsayed, N. M.; von Dohlen, F. M.; Hassett, C. M.;
Postlethwait, E. M. ; Quinn, C. L. ; Graham, J. A.; Gardner, D. E. (1984) A
comparison of biochemical effects of nitrogen dioxide, ozone, and their
combination in mouse lung. I. Intermittent exposures. Toxicol. Appl.
Pharmacol. 72: 82-90.
Nakanishi, Y. ; Schneider, E. L. (1979) In vivo sister-chromatid exchange: a
sensitive measure of DMA damage. Mutat. Res. 60: 329-337.
Narayanan, R. L.; Lancaster, B. W. (1973) Household maintenance costs and
particulate air pollution. Clean Air (Melbourne, Australia) 7: 10-13.
National Council on Radiation Protection and Measurements. (1984a) Evaluation
of occupational and environmental exposures to radon and radon daughters
in the United States: recommendations of the National Council on
Radiation Protection and Measurements. Bethesda, MD: National Council on
Radiation Protection and Measurements; NCRP report no. 78.
6-23
-------
National Council on Radiation Protection and Measurements. (1984b) Exposures
from the uranium series with emphasis on radon and its daughters.
Bethesda, MD: National Council on Radiation Protection and Measurements;
NCRP report no. 77.
National Institute of Occupational Safety and Health. (1985) Evaluation of
epidemiologic studies examining the lung cancer mortality of under ground
miners. Cincinnati, OH: Centers for Disease Control.
National Research Council. (1977) Carbon monoxide. Washington, DC: National
Academy of Sciences; EPA report no. EPA-600/1-77-034. Available from:
NTIS, Springfield, VA; PB-274965.
National Research Council. (1979) Odors from stationary and mobile sources.
Washington, DC: National Academy of Sciences.
National Research Council. (1981a) Formaldehyde and other aldehydes.
Washington, DC: National Academy Press.
National Research Council. (1981b) Indoor pollutants. Washington, DC: National
Academy Press.
National Research Council. (1986a) Environmental tobacco smoke: measuring
exposures and assessing health effects. Washington, DC: National Academy
Press.
National Research Council. (1986b) Preservation of historical records.
Washington, DC: National Academy Press.
National Research Council. (1986c) Indoor pollutants. Washington, DC: National
Academy Press.
Nazaroff, W. W. (1984) Optimizing the total-alpha three-count technique for
measuring concentrations of radon progeny in residences. Health Phys. 46:
395-405.
Nazaroff, W. W.; Cass, G. R. (1986) Mathematical modeling of chemically
reactive pollutants in indoor air. Environ. Sci. Technol. 20: 924-934.
Nazaroff, W. W.; Boegel, M. L.; Hollowell, C. D.; Roseme, G. D. (1981) The use
of mechanical ventilation with heat recovery for controlling radon and
radon-daughter concentrations in houses. Atmos. Environ. 15: 263-270.
Nelms, L. H. ; Mason, M. A.; Tichenor, B. A. (1986) The effects of ventilation
rates and product loading on organic emission rates from particleboard.
In: Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 469-485.
Nelms, L. H. ; Mason, M. A.; Tichenor, B. A. (1987) Determination of emission
rates and concentration levels of p-dichlorobenzene. Accepted for
presentation at: 80th annual meeting of the Air Pollution Control
Association; June; New York, NY. Pittsburgh, PA: Air Pollution Control
Association; paper no. 87-.
6-24
-------
Nero, A. V. (1985) Indoor concentrations of radon 222 and its daughters:
sources, range, and environmental influences. Berkeley, CA: Lawrence
Berkeley Laboratory; report no. LBL-19346.
Nesnow, S.; Argus, M.; Bergman, H. ; Chu, K. Frith, C.; Helmes, T.; McGaughy,
R.; Ray, V.; Slaga, T. J.; Tennant, R.; Weisberger, E. (1986) Chemical
carcinogens: a review and analysis of the literature of selected
chemicals and the establishment of the Gene-Tox Carcinogen Data Base.
Mutat. Res. 185: 1-195.
Neulicht, R. M.; Core, J. (1982) The impact of residential wood combustion
appliances on indoor air quality. In: Proceedings of the Air Pollution
Control Association specialty conference on residential wood and coal
combustion; March; Louisville, KY. Pittsburgh, PA: Air Pollution Control
Association; pp. 240-252.
New York State Energy Research and Development Authority. (1985) Indoor air
quality, infiltration and ventilation in residential buildings. New York,
NY: New York State Energy Research and Development Authority; NYSERDA
report 85-10.
Nilsby, I. (1949) Allergy to moulds in Sweden: a botanical and clinical study.
Acta Allergologica (Copenhagen) 2: 57-90.
Nipe, M. R. (1981) Atmospheric contaminant fading. Text. Chem. Color. 13:
136-146.
Nitta, H.; Maeda, K. (1982) Personal exposure monitoring to nitrogen dioxide.
Environ. Int. 8: 243-246.
Norsted, S. W. ; Kozinetz, C. A.; Annegers, J. F. (1985) Formaldehyde complaint
investigations in mobile homes by the Texas Department of Health.
Environ. Res. 37: 93-100.
O'Connor, J. J. (1913) The economic cost of the smoke nuisance to Pittsburgh.
Pittsburgh, PA: University of Pittsburgh, Mellon Institute of Industrial
Research and School of Specific Industries. (Smoke investigation bulletin
no. 4).
O'Donnell, R. D.; Mikulka, P.; Heinig, P.; Theodore, J. (1971) Low level
carbon monoxide exposure and human psychomotor performance. Toxicol.
Appl. Pharmacol. 18: 593-602.
Ogston, S. A.; Florey, C. du V.; Walker, C. H. M. (1985) The Tayside infant
morbidity and mortality study: effect on health of using gas for cooking.
Br. Med. J. 290: 957-960.
Ohnishi, Y.; Kinouchi, T.; Manabe, Y.; Tsutsui, H.; Otsuka, H.; Tokiwa, H.;
Otofuji, T. (1984) Nitro compounds in environmental mixtures and foods.
In: Waters, M. D.; Shandu, S. S. ; Lewtas, J.; Claxton, L.; Strauss, G.;
Nesnow, S. , eds. Short-term bioassays in the analysis of complex
environmental mixtures IV. New York, NY: Plenum Press; pp. 195-204.
6-25
-------
Oliver, K. D. ; Pleil, J. D.; McClenny, W. A. (1986) Sample integrity of trace
level volatile organic compounds in ambient air stored in Summa polished
cannisters. Atmos. Environ. 20: 1403-1411.
Olsen, J. H. ; Dossing, M. (1982) Formaldehyde induced symptoms in day care
centers. Am. Ind. Hyg. Assoc. J. 43: 366-370.
Olsen, J. H.; Jensen, S. P.; Hink, M.; Faurbo, K.; Breum, N. 0.; Jensen, 0. M.
(1984) Occupational formaldehyde exposure and increased nasal cancer risk
in man. Int. J. Cancer 34: 639-644.
Ong, T.-M. (1975) Aflatoxin mutagenesis. Mutat. Res. 32: 35-53.
Ostro, B. (1984) A search for a threshhold in the relationship of air pollution
to mortality: a reanalysis of data on London winters. EHP Environ. Health
Perspect. 58: 397-399.
Ott, W. R.; Rodes, C. E.; Drago, R. J. ; Williams, C.; Burmann, F. J. (1986a)
Automated data-logging personal exposure monitors for carbon monoxide. J.
Air Pollut. Control Assoc. 36: 883-887.
Ott, W.; Wallace, L. ; Mage, D. ; Akland, G. ; Lewis, R. ; Sauls, H. ; Rodes, C. ;
Kleffman, D.; Kuroda, D.; Morehouse, K. (1986b) The Environmental
Protection Agency's research program on total human exposure. In:
Berglund, B. ; Berglund, U. ; Lindvall, T.; Spengler, J.; Sundell, J., eds.
Indoor air quality: papers from the third international conference on
indoor air quality and climate; August 1984; Stockholm, Sweden. Environ.
Int. 12: 475-494.
Palmgran, U., et al. (1986) Levels of viable and nonviable airborne
microorganisms registered with the CAMNEA-method in association with
Farmer's lung and with environment in stable. In: 3rd international
conference on aerobiology; Basel, Switzerland.
Pandey, M. R. (1984a) Prevalence of chronic bronchitis in a rural community of
the hill region of Nepal. Thorax 39: 331-336.
Pandey, M. R. (1984b) Domestic smoke pollution and chronic bronchitis in a
rural community of the hill region of Nepal. Thorax 39: 337-339.
Pandey, M. R. ; Regmi, H. N.; Neupane, R. P.; Gautam, A.; Bhandari, D. P.
(1984c) Domestic smoke pollution and respiratory function in rural Nepal.
In: Berglund, B. ; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings
of the 3rd international conference on indoor air quality and climate, v.
4, chemical characterization and personal exposure; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
413-420. Available from: NTIS, Springfield, VA; PB85-104214.
Pandey, M. R. ; Regmi, H. N.; Neupane, R. P.; Gautam, A. (1985) Domestic smoke
pollution and respiratory function in rural Nepal. Tokai J. Exp. Clin.
Med. 10: 471-481.
6-26
-------
Paul, R. A.; Johnson, T. (1985) The NAAQS exposure model (NEM) applied to
carbon monoxide: addendum. Research Triangle Park, NC: U. S.
Environmental Protection Agency; EPA report no. EPA-450/5-85-004.
Pellizzari, E. D. , et al. (1981) Total exposure assessment methodology (TEAM)
study. V. I: northern New Jersey; v. II: Research Triangle Park; v. Ill:
quality assurance. Washington, DC: U. S. Environmental Protection Agency,
Office of Research and Development; EPA contract no. 68-01-3849.
Pellizzari, E. D. ; Hartwell, T. D.; Leininger, C.; Zelon, H. ; Williams, S.;
Breen, J.; Wallace, L. (1982) Human exposure to vapor-phase halogenated
hydrocarbons. Fixed-site vs. personal exposure. In: Proceedings from the
symposium on ambient, sources, and exposure monitoring for non-criteria
pollutants.
Pepys, J. (1982) Chemical dusts, vapours, and fumes causing asthma. Environ.
Int. 8: 321-325.
Perdue, P. T.; Dickson, H. W. ; Haywood, F. F. (1980) Radon monitoring
instrumentation. Health Phys. 39: 85-88.
Phillips, J. L.; Winters, W. D. ; Rutledge, L. (1986) In vitro exposure to
electromagnetic fields: changes in tumour cell properties. Int. J.
Radiat. Biol. 49: 463-469.
Pickrell, J. A.; Griffis, L. C.; Mokler, B. V.; Kanapilly, G. M.; Hobbs, C. H.
(1984) Formaldehyde release from selected consumer products: influence of
chamber loading, multiple products, relative humidity, and temperature.
Environ. Sci. Technol. 18: 682-686.
Pirnay, F. ; Dujardin, J. ; Deroanne, R.; Petit, J. M. (1971) Muscular exercise
during intoxication by carbon monoxide. J. Appl. Physiol. 31: 573-575.
Pullan, C. R. ; Hey, E. N. (1982) Wheezing, asthma, and pulmonary dysfunction
10 years after infection with respiratory syncytial virus in infancy. Br.
Med. J. 284: 1667-1669.
Putz, V. R. (1979) The effects of carbon monoxide on dual-task performance.
Human Factors 21: 13-24.
Putz, V. R. ; Johnson, B. L.; Setzer, J. V. (1976) Effects of CO vigilance
performance. Effects of low level carbon monoxide on divided attention,
pitch discrimination, and the auditory evoked potential. Cincinnati, OH:
U. S. Department of Health, Education and Welfare, National Institute for
Occupational Safety and Health; DHEW (NIOSH) publication no. 77-124.
Putz, V. R. ; Johnson, B. L. ; Setzer, J. V. (1979) A comparative study of the
effects of carbon monoxide and methylene chloride on human performance.
In: Proceedings of the 1st annual NIOSH Science Symposium; Chicago, IL:
Pathotox Publishing Co.
Rantakallio, P. (1978) Relationship of maternal smoking to morbidity and
mortality of the child up to the age of five. Acta Paediatr. Scand. 67:
621-631.
6-27
-------
Rao, M. K. (1986) Approaches to air pollution monitoring and control in
submarines. In: Proceedings of the 1986 EPA/APCA symposium on
measurements of toxic air pollutants. Research Triangle Park, NC: U. S.
Environmental Protection Agency; pp. 95-103; EPA report no.
EPA-600/9-86-013.
Reed, C. E. (1981) Allergic agents. Bull. N. Y. Acad. Med. 57: 897-906.
Reed, C. E.; Swanson, M. C.; Lopez, M.; Ford, A. M.; Major, J.; Witmer, W. B.;
Valdes, T. B. (1983) Measurement of IgG antibody and airborne antigen to
control and industrial outbreak of hypersensitivity pneumonitis. JOM J.
Occup. Med. 25: 207-210.
Remijn, B.; Fischer, P.; Brunekreef, B. ; Lebret, E. ; Boleij, J. S. M.; Noij,
D. (1985) Indoor air pollution and its effect on pulmonary function of
adult non-smoking women: I. Exposure estimates for nitrogen dioxide and
passive smoking. Int. J. Epidemic!. 14: 215-220.
Repace, J. L (1981) The problem of passive smoking. In: Symposium on health
aspects of indoor air pollution; May; New York, NY. Bull. N. Y. Acad.
Med. 57: 936-946.
Repace, J. L. (1982) Indoor air pollution. Environ. Int. 8: 21-36.
Repace, J. L. (1984) Modeling and measurement of indoor pollution from
sidestream smoke. In: IARC monograph on sidestream smoke.
Repace, J. L. ; Lowrey, A. H. (1980) Indoor air pollution, tobacco smoke, and
public health. Science (Washington, DC) 208: 464-472.
Repace, J. L.; Lowrey, A. H. (1983) Modeling exposure of nonsmokers to ambient
tobacco smoke. Presented at: 76th annual meeting of the Air Pollution
Control Association; June; Atlanta, GA. Pittsburgh, PA: Air Pollution
Control Association; paper no. 83-64.2.
Repace, J. L.; Lowrey, A. H. (1985) A quantitative estimate of nonsmokers1
lung cancer risk from passive smoking. Environ. Int. 11: 3-22.
Research Triangle Institute. (1986) Review of indoor air sorbents interim
report. Research Triangle Park, NC: U. S. Environmental Protection
Agency; cooperative agreement no. CR812522-01. [upublished].
Revza, K. L. (1984) Effectiveness of local ventilation in removing simulated
pollution from point sources. Lawrence Berkeley Laboratories; LBL-16701.
Riley, R. L. (1982) Indoor airborne infection. Environ. Int. 8: 317-320.
Rittfeldt, L. ; Sandberg, M. ; Ahlberg, M. S. (1984) Indoor air pollutants due
to vinyl floor tiles. In: Berglund, B. ; Lindvall, T. ; Sundell, J., eds.
Indoor air: proceedings of the 3rd international conference on indoor air
quality and climate, v. 3, sensory and hyperreactivity reactions to sick
buildings; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council
for Building Research; pp. 297-302.
6-28
-------
Roberts, J. W. ; Ruby, M. G. ; Warren, G. R. (1987) Mutagenic activity of house
dust. In: Sandhu, S. S.; Claxton, L. ; Mass, M.; Moore, M.; Mumford, J.;
DeMarini, D. , eds. Short-term bioassays in the analysis of complex
environmental mixtures V. New York, NY: Plenum Press.
Rockwell, T. J. ; Weir, F. W. (1975) The interactive effects of carbon monoxide
and alcohol on driving skills. Columbus, OH: Ohio State University
Research Foundation.
Rodricks, J. V.; Hesseltime, C. W. ; Mehlman, M. A. (1977) Mycotoxins in human
and animal health. Park Forest South, IL: Pathotox.
Roger, L. J. ; Kehrl, H. R. ; Hazucha, M. ; Horstman, D. H. (1985)
Bronchoconstriction in asthmatics exposed to sulfur dioxide during
repeated exercise. J. Appl. Physiol. 59: 784-791.
Ruh, C.; Gebefuegi, I.; Korte, F. (1984) The indoor biocide pollution:
occurrence of pentachlorophenol and lindane in homes. In: Berglund, B.;
Lindvall, T. ; Sundell, J. , eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 4,
chemical characterization and personal exposure; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
309-315. Available from: NTIS, Springfield, VA; PB85-104214.
Rummo, N. ; Sarlanis, K. (1974) The effect of carbon monoxide on several
measures of vigilance in a simulated driving task. J. Saf. Res. 6:
126-130.
Rylander, R.; Haglind. (1984) Airborne endotoxin and humidifier disease. Clin.
Allergy 14: 109-112.
Salvatore, S. (1974) Performance decrement caused by mild carbon monoxide
levels on two visual functions. J. Saf. Res. 6: 131-134.
Salvin, V. S. (1969) Ozone fading of dyes. Text. Chem. Color. 1: 245-251.
Samfield, M. (1985) Importance of cigarette sidestream smoke - further
aspects. Tabak J. Int. (6): 448.
Sanchez, D. C. ; Mason, M. ; Norris, C. (1987) Methods and results of
characterization of organic emissions from an indoor material. In:
Symposium on the characterization of contaminant emissions from indoor
sources; Chapel Hill, NC. Atmos. Environ. 21: 337-345.
Sardinas, A. V.; Giulietti, M. A.; Most, R. S.; Honcher, P. (1979) Health
effects associated with urea-formaldehyde foam insulation in Connecticut.
J. Environ. Health 41: 270-272.
Savitz et al. (1986) Presentation at the Department of Energy contractors
review. November; Denver, CO.
Schenker, M. B.; Weiss, S. T.; Murawski, B. J. (1982) Health effects of
residence in homes with urea formaldehyde foam insulation: a pilot study.
Environ. Int. 8: 359-363.
6-29
-------
Schenker, M. B. ; Samet, J. M. ; Speizer, F. E. (1983) Risk factors for
childhood respiratory disease: the effect of host factors and home
environmental exposures. Am. Rev. Respir. Dis. 128: 1038-1043.
Schiff, H. I.; MacKay, G. I.; Casteldine, C.; Harris, G. W. ; Iran, Q. (1986)
In: Proceedings of the 1986 EPA/APCA symposium on measurement of toxic
air pollutants; April; Raleigh, NC; pp. 834-844.
Schilling, R. S. F.; Letai, A. D.; Hiu, S.-L.; Beck, G. J.; Schoenberg, J. B.;
Bouhuys, A. (1977) Lung function, respiratory disease, and smoking in
families. Am. J. Epidemic!. 106: 274-283.
Schmitt, C. H. A. (1960) Lightfastness of dyestuffs on textiles. Am. Dyest.
Rep. 49: 974-980.
Schmitt, C. H. A. (1962) Daylight fastness testing by the Langley system. Am.
Dyest. Rep. 51: 664-675.
Schulte, J. H. (1973) Effects of mild carbon monoxide intoxiation. Arch.
Environ. Health 5: 524-530.
Seifert, B. ; Ullrich, D. ; Nagel, R. (1984) Analyse und Vorkommen von
gasfoermigen organischen Verbindungen in der Luft von Innenraeumen
[Analysis and presentation of gas-forming organic compounds in indoor
air]. In: Lahmann, E.; Jander, K., eds. Lufthygiene 1984 [Air hygiene
1984]. New York, NY: Gustav Fischer Verlag; pp. 1-13. (Schriftenreihe des
Vereins fuer Wasser-, Boden- und Lufthygiene no. 59).
Seifert, V. B.; Ullrich, D.; Mailahn, W. ; Nagel, R. (1986) Fluechtige
organische Verbindungen in der Innenraumluft [Volatile organic compounds
in indoor air]. Bundesgesundheitsblatt 29: 417-424.
Seppanen, 0. (1986) Ventilation efficiency in practice. In: Proceedings of the
ASHRAE conference IAQ '86: managing indoor air for health and energy
conservation; April; Atlanta, GA. Atlanta, GA: American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc.; pp. 559-567.
Sexton, K. ; Repetto, R. (1982) Indoor air pollution and public policy.
Environ. Int. 8: 5-10.
Sexton, K. ; Ryan, P. B. (1987) Assessment of human exposure to air pollution:
methods, measurements, and models. In: Watson, A.; Bates, R. R.; Kennedy,
D., eds. Air pollution, the automobile and public health: research
opportunities for quantifying risk. Washington, DC: National Academy of
Sciences Press; in press.
Sexton, K. ; Webber, L. M.; Hayward, S. B.; Sextro, R. G.; Offerman, F. J.
(1984) Characterization of particulate and organic emissions from major
indoor sources. In: Berglund, B.; Lindvall, T. ; Sundell, J., eds. Indoor
air: proceedings of the 3rd international conference on indoor air
quality and climate, v. 4, chemical characterization and personal
exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council
for Building Research; pp. 163-168. Available from: NTIS, Springfield,
VA; PB85-104214.
6-30
-------
Sextro, R. G.; Offerman, F. J. ; Nazaroff, W. W. (1984) Evaluation of indoor
aerosol control devices and their effects on radon progeny
concentrations. Berkeley, CA: Lawrence Berkeley Laboratory; report no.
LBL-17598.
Shapiro, J. (1956) Radiation dosage from breathing radon and its daughter
products. AMA Arch. Ind. Health 14: 169-177.
Shephard, R. J. ; Collins, R. ; Silverman, F. (1979) "Passive" exposure of
asthmatic subjects to cigarette smoke. Environ. Res. 20: 392-402.
Sheppard, A. R.; Eisenbud, M. (1977) Biological effects of electric and
magnetic fields of extremely low frequency. New York, NY: New York
University Press; p. 3-3.
Shumway, R. H.; Tai, R. Y.; Tai, L. P.; Rewitan, Y. (1983) Statistical
analysis of daily London mortality and associated weather and pollution
effects. Sacramento, CA: California Air Resources Board; contract no.
Al-154-33.
Sinclair, J. D.; Psota-Kelty, L. A. (1985) Indoor-outdoor concentrations and
indoor surface accumulations of ionic substances. Atmos. Environ. 19:
315-323.
Sittig, M. (1985) Asbestos. In: Handbook of toxic and hazardous chemicals and
carcinogens. 2nd ed. Park Ridge, NJ: Noyes Publications; pp. 92-93.
Skaaret, E. (1986) Contaminant removal performance in terms of ventilation
effectiveness. In: Berglund, B.; Berglund, U.; Lindvall, T.; Spengler,
J. ; Sundell, J., eds. Indoor air quality: papers from the third
international conference on indoor air quality and climate; August 1984;
Stockholm, Sweden. Environ. Int. 12: 419-427.
Smith, K. R. (1986) The importance of indoor air pollution to personal
exposures in developing countries. In: Berglund, B.; Berglund, U.;
Lindvall, T. ; Sundell, J. , eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 6,
evaluations and conclusions for health sciences and technology; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 43-47.
Smolen, M. (1984) The healthy home: radon remedies. Here's how to rid your
home of this deadly gas. Rodale's New Shelter 5: 28-32, 81.
Snihs, J. 0. (1985) Non-occupational exposure to ionizing radiation. In:
Castellani, A., ed. Epidemiology and quantisation of environmental risk
in humans from radiation and other agents. New York, NY: Plenum Press;
pp. 363-374.
Soczek, M. L. ; Ryan, P. B.; Treitman, R. D. ; Spengler, J. D. ; Porter, J. L.;
Billick, I. H. (1986) The Boston residential N02 characterization study:
an evaluation of the survey methodology. Presented at: 79th annual
meeting of the Air Pollution Control Association; June; Minneapolis, MN.
Pittsburgh, PA: Air Pollution Control Association; paper no. 86-5.9.
6-31
-------
Solli, H. M. ; Andersen, A.; Stranden, E.; Langard, S. (1985) Cancer incidence
among workers exposed to radon and thoron daughters at a niobium mine.
Scand. J. Work Environ. Health 11: 7-13.
Solomon, W. R. (1975) Assessing fungus prevalence in domestic interiors. J.
Allergy Clin. Immunol. 56: 235-242.
Solomon, W. R.; Burge, H. A. (1984) Allergens and pathogens. In: Walsh, P.;
Dudney, C.; Copenhaver, E., eds. Indoor air quality. Boca Raton, FL: CRC
Press, Inc.; pp. 173-191.
Speizer, F. E.; Ferris, B., Jr.; Bishop, Y. M. M.; Spengler, J. (1980)
Respiratory disease rates and pulmonary function in children associated
with N02 exposure. Am. Rev. Respir. Dis. 121: 3-10.
Spence, J. W. ; Haynie, F. ; Upham, J. B. (1975) Effects of gaseous pollutants
on paints: a chamber study. J. Paint Technol. 47: 57-63.
Spengler, J. p.; Ferris, B. G., Jr.; Dockery, D. W.; Speizer, F. E. (1979)
Sulfur dioxide and nitrogen dioxide levels inside and outside homes and
the implications on health effects research. Environ. Sci. Technol. 13:
1276-1280.
Spengler, J. D. ; Dockery, D. W.; Turner, W. A.; Wolfson, J. M.; Ferris, B. G.,
Jr. (1981) Long-term measurements of respirable sulfates and particles
inside and outside homes. Atmos. Environ. 15: 23-30.
Spengler, J. D. ; Treitman, R. D.; Tosteson, T. D.; Mage, D. T.; Soczek, M. L.
(1985) Personal exposures to respirable particulates and implications for
air pollution epidemiology. Environ. Sci. Technol. 19: 700-707.
Spicer, C. W. ; Slivon, L. E.; Coutant, R. W.; Graves, M. E.; Shadwick, D. S.;
McClenny, W. A.; Mulik, J. D. ; Fitz-Simons, T. R. (1986) Intercomparison
of sampling techniques for toxic organic compounds in indoor air. In:
Proceedings of the 1986 EPA/APCA symposium on measurement of toxic air
pollutants; April; Raleigh, NC; pp. 45-60.
Staats, E. (1980) Need for a formal risk/benefit review. The pesticide
chlordane. U. S. Environmental Protection Agency, Comptroller General,
General Accounting Office letter B-199618 to administrator of EPA; August
5.
Stark, A. A.; Essigmann, J. M.; Demain, A. L.; Skopek, T. R.; Wogan, G. N.
(1979) Aflatoxin Bl mutagenesis, DNA binding, and adduct formation in
Salmonella typhimurium. Proc. Natl. Acad. Sci. U. S. A. 76: 1343-1347.
Starr, H. G., Jr.; Aldrich, F. D. ; McDougal, W. D., III; Mounce, L. M. (1974)
Contribution of household dust to the human exposure to pesticides.
Pestic. Monit. J. 8: 209-212.
State of California. (1984) Building closeout procedure. Sacramento, CA:
Office of the State Architect.
6-32
-------
Stayner, L. T. ; Elliott, L. ; Blade, L.; Keenlyside, R.; Halperin, W. (1986) A
retrospective cohort mortality study of workers exposed to formaldehyde
in the garment industry.
Sterling, D. A. (1985) Volatile organic compounds in indoor air: an overview
of sources, concentrations, and health effects. In: Gammage, R. B.; Kaye,
S. B.; Jacobs, V. A., eds. Indoor air and human health. Chelsea, MI:
Lewis Publishers, Inc.; pp. 387-402.
Sterling, T. D. ; Kobayashi, D. (1981) Use of gas ranges for cooking and
heating in urban dwellings. J. Air Pollut. Control Assoc. 31: 162-165.
Stewart, R. D.; Newton, P. E. ; Kaufman, J. ; Forster, H. V.; Klein, J. P.
(1978) Effect of a rapid 4 percent carboxyhemoglobin saturation increase
on maximal treadmill exercise. CRC-APRAC-CAPM-22-75.
Sugimura, T. (1985) Carcinogenicity of mutagenic heterocyclic amines formed
during the cooking process. Mutat. Res. 150: 33-41.
Surgeon General of the United States. (1986) The health consequences of
involuntary smoking: a report of the Surgeon General. Rockville, MD: U.
S. Department of Health and Human Services, Public Health Service.
Swanson, M. C.; Agarwal, M. K.; Reed, C. E. (1985) An immunochemical approach
to indoor aeroallergen quantitation with a new volumetric air sampler:
studies with mite, roach, cat, mouse, and guinea pig antigens. J. Allergy
Clin. Immunol. 76: 724-729.
Sykora, J.; Karol, M.; Keleti, G. ; Novak, D. (1982) Amoebae as sources of
hypersensitivity pneumonitis. Environ. Int. 8: 343-347.
Tager, I. B.; Weiss, S. T. ; Rossner, B. ; Speizer, F. E. (1979) Effect of
parental cigarette smoking on the pulmonary function of children. Am. J.
Epidemiol. 110: 15-26.
Tager, I. B. ; Weiss, S. T.; Munoz, A.; Rosner, B. ; Speizer, F. E. (1983)
Longitudinal study of the effects of maternal smoking on pulmonary
function in children. New. Engl. J. Med. 309: 699-703.
Tashkin, D. P.; Clark, V. A.; Simmons, M.; Reems, C.; Coulson, A. H.; Bourque,
L. B.; Sayre, J. W. ; Detels, R. ; Rokaw, S. (1984) The UCLA population
studies of chronic obstructive respiratory disease: relationship between
parental smoking and children's lung function. Am. Rev. Respir. Dis. 129:
891-897.
Taylor, N. A.; Pickering, C. A. C. ; Turner-Warwick, M. ; Pepys, J. (1978)
Respiratory allergy to a factory humidifier contaminant presenting as
pyrexia of undetermined origin. Br. Med. J. 2(6129): 94-95.
Tejada, S. (1986) J. Environ. Anal. Chem. 26: 167-185.
Tennessee Valley Authority. (1985) Indoor air quality study, phase II.
Chattanooga, TN: Tennessee Valley Authority, Division of Conservation and
Energy Management.
6-33
-------
Tennessee Valley Authority. (1986) TVA/BPA indoor air quality study, v. 1.
Chattanooga, TN: Tennessee Valley Authority, Division of Conservation and
Energy Management.
Thomas, J. R.; Schrot, J. ; Liboff, A. R. (1986) Low-intensity magnetic fields
alter operant behavior in rats. Bioelectromagnetics 7: 349-357.
Thun, M. J. ; Lakat, M. F.; Altman, R. (1982) Symptoms survey of residents of
homes insulated with urea-formaldehyde foam. Environ. Res. 29: 320-334.
Tichenor, B. A.; Mason, M. A. (1986) Characterization of organic emissions
from selected materials in indoor use. Presented at: 79th annual meeting
of the Air Pollution Control Association; June; Minneapolis, MN.
Pittsburgh, PA: Air Pollution Control Association; paper no. 86-16.5.
Tichenor, B. A.; Jackson, M. D. ; Merrill, R. G. (1986) Measurement of organic
emissions from indoor materials - small chamber studies. Presented at:
EPA/APCA symposium on measurement of toxic air pollutants; April;
Raleigh, NC. EPA report no. EPA-600/9-86-013.
Tokiwa, H.; Nakagawa, R. ; Horikawa, K. (1985) Mutagenic/carcinogenic agents in
indoor pollutants: the dinitropyrenes generated by kerosene heaters and
fuel gas and liquefied petroleum gas burners. Mutat. Res. 157: 39-47.
Tomenius, L. (1986) 50-Hz electromagnetic environment and the incidence of
childhood tumors in Stockholm County. Bioelectromagnetics 7: 191-207.
Transmission/distribution health and safety. (1986) In:
Transmission/distribution health and safety report. 3(10): 1.
Traynor, G. W.; Allen, J. R.; Apte, M. G.; Dillworth, J. F. ; Girman, J. R.;
Hollowell, C. D.; Koonce, J. F., Jr. (1982a) Indoor air pollution from
portable kerosene-fired space heaters, wood-burning stoves, and
wood-burning furnaces. In: Proceedings of the Air Pollution Control
Association specialty conference on residential wood and coal combustion;
March; Louisville, KY. Pittsburgh, PA: Air Pollution Control Association;
pp. 253-263.
Traynor, G. W. ; Anthon, D. W.; Hollowell, C. D. (1982b) Technique for
determining pollutant emissions from a gas-fired range. Atmos. Environ.
16: 2979-2987.
Traynor, G. W.; Apte, M. G. ; Dillworth, J. F. ; Hollowell, C. D. ; Sterling, E.
M. (1982c) The effects of ventilation on residential air pollution due to
emissions from a gas-fired range. Environ. Int. 8: 447-452.
Traynor, G. W.; Allen, J. R.; Apte, M. G.; Girman, J. R.; Hollowell, C. D.
(1983) Pollutant emissions from portable kerosene-fired space heaters.
Environ. Sci. Technol. 17: 369-371.
Traynor, G. W., et al. (1984) Indoor air pollution and interroom transport due
to unvented space heaters. LBL-17600. Available from: NTIS, Springfield,
VA; CONF-840803-2.
6-34
-------
Traynor, G. W.; Apte, M. G. ; Sokol, H. A.; Chuang, J. C.; Mumford, J. L.
(1986) Selected organic pollutant emissions from unvented kerosene
heaters. Presented at: 79th annual meeting of the Air Pollution Control
Association; June; Minneapolis, MN. Pittsburgh, PA: Air Pollution Control
Association; paper no. 86-52.5.
Tribukait et al. (1986) May; Sweden; conference.
Trichopoulos, D.; Kalandidi, A.; Saparros, L.; MacMahon, B. (1981) Lung cancer
and passive smoking. Int. J. Cancer 27: 1-4.
Trzos, R. J.; Petzold, G. L.; Brunden, M. N. ; Swenberg, J. A. (1978) The
evaluation of sixteen carcinogens in the rat using the micronucleus test.
Mutat. Res. 58: 79-86.
Tucker, W. G. (1986a) EPA research on indoor air quality. Presented at: 79th
annual meeting of the Air Pollution Control Association; June;
Minneapolis, MN. Pittsburgh, PA: Air Pollution Control Association; paper
no. 86-11.5.
Tucker, W. G. (1986b) Research overview: sources of indoor air pollutants. In:
Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 395-404.
Tuthill, R. W. (1984) Woodstoves, formaldehyde, and respiratory disease. Am.
J. Epidemic!. 120: 952-955.
U. S. Department of Energy. (1985) Indoor air quality environmental
information handbook. U. S. Department of Energy; DOE/EV/10450-1.
U. S. Department of Health and Human Services. (1984) The health consequences
of smoking: chronic obstructive lung disease: a report of the Surgeon
General. Rockville, MA: Public Health Service, Office on Smoking and
Health; PHS report no. PHS/84-50205.
U. S. Environmental Protection Agency. (1978) Sprayed asbestos-containing
materials in buildings: a guidance document. Research Triangle Park, NC:
Office of Air Quality Planning and Standards; EPA report no.
EPA-450/2-78-014. Available from: NTIS, Springfield, VA; PB-280376.
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon
monoxide. Research Triangle Park, NC: Office of Health and Environmental
Assessment, Environmental Criteria and Assessment Office; EPA report no.
EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PB81-244840.
U. S. Environmental Protection Agency. (1980) National household pesticide
usage study, 1976-77. Washington, DC: Office of Pesticide Programs; EPA
report no. EPA-540/9-80-002. Available from: NTIS, Springfield, VA;
PB81-181570.
6-35
-------
U. S. Environmental Protection Agency. (1982a) Air quality criteria for
participate matter and sulfur oxides: volume III. Research Triangle Park,
NC: Office of Health and Environmental Assessment, Environmental Critera
and Assessment Office; EPA report no. EPA-600/8-82-029c.
U. S. Environmental Protection Agency. (1982b) Air quality criteria for oxides
of nitrogen: final report. Research Triangle Park, NC: Office of Health
and Environmental Assessment, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-82-026F. Available from: NTIS,
Springfield, VA; PB83-163337.
U. S. Environmental Protection Agency. (1984) Revised evaluation of health
effects associated with carbon monoxide exposure: an addendum to the 1979
air quality criteria document for carbon monoxide. Washington, DC: Office
of Health and Environmental Assessment, Environmental Criteria and
Assessment Office; EPA report no. EPA-600/8-83-033F. Available from:
NTIS, Springfield, VA; PB85-103471.
U. S. Environmental Protection Agency. (1985a) Health risks due to radon in
structures: a strategy management plan for assessment and mitigation.
September; unpublished.
U. S. Environmental Protection Agency. (1985b) Technologies and costs for the
removal of radon from potable water supplies, second draft. Science and
Technology Branch, Criteria and Standards Divison, Office of Drinking
Water. September [unpublished].
U. S. Environmental Protection Agency. (1985c) Asbestos waste management
guidance: generation transport and disposal. Washington, DC: Office of
Solid Waste; May.
U. S. Environmental Protection Agency. (1986a) Second addendum to air quality
criteria for particulate matter and sulfur oxides (1982): assessment of
newly available health effects information. Research Triangle Park, NC:
Office of Health and Environmental Assessment,Environmental Criteria and
Assessment Office; EPA report no. EPA-600/8-86-020A.
U. S. Environmental Protection Agency. (1986b) Air quality criteria for lead:
volume II of four. Research Triangle Park, NC: Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office;
EPA report no. EPA-600/8-83/028bF.
U. S. Environmental Protection Agency. (1986c) Radon reduction techniques for
detached houses: technical guidance. Washington, DC: Office of Research
and Development; EPA report no. EPA-625/5-86-019.
U. S. Environmental Protection Agency. (1986d) Radon reduction methods: a
homeowner's guide. OPA-86-005.
U. S. Environmental Protection Agency. (1986e) Interim indoor radon and radon
decay products measurement protocols. Office of Radiation Programs; EPA
report no. EPA-520/1-86-04.
Upham, J. B.; Haynie, F. H. ; Spence, J. W. (1976) Fading of selected drapery
fabrics by air pollutants. J. Air Pollut. Control Assoc. 26: 790-792.
6-36
-------
van Houdt, J. J.; Boleij, J. S. M. (1984) Mutagenic activity of indoor
airborne particles compared to outdoors. In: Berglund, B.; Lindvall, T.;
Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 2, radon, passive
smoking, particulates and housing epidemiology; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp.
169-176.
van Houdt, J. J.; Jongen, W. M. F.; Alink, G. M. ; Boleij, J. S. M. (1984)
Mutagenic activity of airborne particles inside and outside homes.
Environ. Mutagen. 6: 861-869.
Vedal, S.; Schenker, M. B.; Samet, J. M.; Speizer, F. E. (1984) Risk factors
for childhood respiratory disease: analysis of pulmonary function. Am.
Rev. Respir. Dis. 130: 187-192.
Vu Due, T. ; Favez, C. M. P. (1981) Characteristics of motor exhausts in an
underground car park: mass size distribution and concentration levels of
particles. J. Environ. Sci. Health A16: 647-660.
Wade, W. A., Ill; Cote, W. A.; Yocom, J. E. (1975) A study of indoor air
quality. J. Air Pollut. Control Assoc. 25: 933-939.
Walker, M. V.; Weschler, C. J. (1980) Water-soluble components of
size-fractionated aerosols collected after hours in a modern office
building. Environ. Sci. Techno!. 14: 594-597.
Wallace, L. A. (1983) Carbon monoxide in air and breath of employees in an
underground office. J. Air Pollut. Control Assoc. 33: 678-682.
Wallace, L. A. (1986a) An overview of the Total Exposure Assessment Methodology
(TEAM) Study, v. I. Washington, DC: U. S. Environmental Protection
Agency; February; (in press).
Wallace, L. A. (1986b) Personal exposures, indoor and outdoor air
concentrations, and exhaled breath concentrations of selected volatile
organic compounds measured for 600 residents of New Jersey, North Dakota,
North Carolina, and California. Toxicol. Environ. Chem. 12: 215-236.
Wallace, L. A.; Pellizzari, E. D.; Gordon, S. M. (n.d.) Organic chemicals in
indoor air: a review of human exposure studies and indoor air quality
studies. In:
Wallace, L. A.; Pellizzari, E. ; Hartwell, T. ; Rosenzweig, M. ; Erickson, M.;
Sparacino, C.; Zelon, H. (1984a) Personal exposure to volatile organic
compounds: I. Direct measurements in breathing-zone air, drinking water,
food, and exhaled breath. Environ. Res. 35: 293-319.
Wallace, L. A.; Pellizzari, E. D.; Hartwell, T. D. ; Zelon, H. S.; Sparacino,
C. M.; Whitmore, R. (1984b) Analyses of exhaled breath of 355 urban
residents for volatile organic compounds. In: Berglund, B. ; Lindvall, T. ;
Sundell, J. , eds. Indoor air: proceedings of the 3rd international
conference on indoor air quality and climate, v. 4, chemical
characterization and personal exposure; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 15-20.
Available from: NTIS, Springfield, VA; PB85-104214.
6-37
-------
Wallace, L. A.; Pellizzari, E. D.; Hartwell, T. D.; Sparacino, C. M. ; Sheldon,
L. S.; Zelon, H. S. (1984c) Personal exposures, outdoor concentrations,
and breath levels of toxic air pollutants measured for 425 persons in
urban, suburban, and rural areas. Presented at: 77th annual meeting of
the Air Pollution Control Association; June; San Francisco, CA.
Pittsburgh, PA: Air Pollution Control Association; paper no. 84-1.8.
Wallace, L. A.; Pellizzari, E. D.; Hartwell, T. D.; Whitmore, R.; Sparacino,
C. M.; Zelon, H. S. (1986) Total exposure assessment methodology (TEAM)
study: personal exposures, indoor-outdoor relationships, and breath
levels of volatile organic compounds in New Jersey. In: Berglund, B.;
Berglund, U. ; Lindvall, T. ; Spengler, J.; Sundell, J., eds. Indoor air
quality: papers from the third international conference on indoor air
quality and climate; August 1984; Stockholm, Sweden. Environ. Int. 12:
369-387.
Walling, J. F. (1984) The utility of distributed air volume sets when sampling
ambient air using solid adsorbents. Atmos. Environ. 18: 855-859.
Walling, J. F.; Bumgarner, J. E.; Driscoll, D. J.; Morris, C. M.; Riley, A.
E. ; Wright, L. H. (1986) Apparent reaction products desorbed from tenax
used to sample ambient air. Atmos. Environ. 20: 51-57.
Wallingford, K. M.; Carpenter, J. (1986) Field experience overview:
investigating sources of indoor air quality problems in office buildings.
In: Proceedings of the ASHRAE conference IAQ '86: managing indoor air for
health and energy conservation; April; Atlanta, GA. Atlanta, GA: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.;
pp. 448-453.
Walrath, J. (1983) Mortality among embalmers. Am. J. Epidemiol. 118: 432.
Walrath, J. ; Fraumeni, J. F. , Jr. (1983) Mortality patterns among embalmers.
Int. J. Cancer 31: 407-411.
Walsh, P. J. ; Dudney, C. S.; Copenhaver, E. D. , eds. (1984) Indoor air
quality. Boca Raton, FL: CRC Press, Inc.
Ware, J. H.; Dockery, D. W. ; Spiro, A., Ill; Speizer, F. E.; Ferris, B. G.,
Jr. (1984) Passive smoking, gas cooking, and respiratory health of
children living in six cities. Am. Rev. Respir. Dis. 129: 366-374.
Ware, J. H.; Ferris, B. G., Jr.; Dockery, D. W.; Spengler, J. D.; Stram, D.
0.; Speizer, F. E. (1986) Effects of ambient sulfur oxides and suspended
particles on respiratory health of preadolescent children. Am. Rev.
Respir. Dir. 133: 834-842.
Watson, W. D. ; Jaksch, J. A. (1982) Air pollution: household soiling and
consumer welfare losses. J. Environ. Econ. Manage. 9: 248-262.
Weber, A. (1984) Annoyance and irritation by passive smoking. Prev. Med. 13:
618-625.
6-38
-------
Weber-Tschopp, A.; Fischer, T.; Grandjean, E. (1976) Objektive und subjektive
physiologische Wirkungen des Passivrauchens [Physiological and
psychological effects of passive smoking]. Int. Arch. Occup. Environ.
Health 37: 277-288.
Weiser, P. C.; Merrill, C. G.; Dickey, D. W.; Kurt, T. L.; Cropp, G. J. A.
(1978) Effects of low-level carbon monoxide exposure on the adaptation of
healthy young men to aerobic work at an altitude of 1,610 meters. In:
Folinsbee, L. J.; Wagner, J. A.; Borgia, J. F. ; Drinkwater, B. L.;
Gliner, J. A.; Bedi, J. F. , eds. Environmental stress. Individual human
adaptations. New York, NY: Academic Press; pp. 101-110.
Weiss, S. T.; Tager, I. B.; Speizer, F. E.; Rosner, B. (1980) Persistent
wheeze: its relation to respiratory illness, cigarette smoking, and level
of pulmonary function in a population sample of children. Am. Rev.
Respir. Dis. 122: 697-707.
Wendel, G. J.; Stedman, D. H.; Cantrell, C. A.; Damrauer, L. (1983)
Luminol-based nitrogen dioxide detector. Anal. Chem. 55: 937-940.
Wertheimer, N.; Leeper, E. (1979) Electrical wiring configuration and
childhood cancer. Am. J. Epidemiol. 109: 273-284.
Wertheimer, N. ; Leeper, E. (1986) Possible effects of electric blankets and
heated waterbeds on fetal development. Bioelectromagnetics 7: 13-22.
White, J. R. ; Froeb, H. F. (1980) Small-airways dysfunction in non-smokers
chronically exposed to tobacco smoke. New Engl. J. Med. 302: 720-723.
Wiedemann, H. P.; Mahler, D. A.; Loke, J.; Virgulto, J. A.; Snyder, P.;
Matthay, R. A. (1986) Acute effects of passive smoking on lung function
and airway reactivity in asthmatic subjects. Chest 89: 180-185.
Wilson, N. K.; Lewis, R. G.; Chuang, C. C.; Petersen, B. A.; Mack, G. A.
(1985) Analytical and sampling methodology for characterization of
polynuclear aromatic compounds in indoor air. Presented at: 78th annual
meeting of the Air Pollution Control Association; June; Detroit, MI.
Pittsburgh, PA: Air Pollution Control Association; paper no. 85-30A.2.
Winchester, J. W. ; Nelson, J. W. (1979) Sources and transport of trace metals
in urban aerosols. Research Triangle Park, NC: U. S. Environmental
Protection Agency, Office of Research and Development; EPA report no.
EPA-600/3-79-019.
Winneke, G. (1973) Behavioral effects of methylene chloride and carbon
monoxide as assessed by sensory and psychomotor performance. In:
Xintaras, C. ; Johnson, B. L. ; de Groot, I., eds. Bahavioral toxicology.
Early detection of occupational hazards. Proceedings of a workshop; June;
Cincinnati, OH. U. S. Department of Health, Education, and Welfare
publication no. (NIOSH) 75-126.
Wolff, G. T. ; Stroup, C. M. ; Stroup, D. P. (1983) The coefficient of haze as a
measure of particulate elemental carbon. J. Air Pollut. Control Assoc.
33: 746-750.
6-39
-------
Woods, J. E. (1983) Sources of indoor air contaminants. American Society of
Heating, Refrigerating, and Air-conditioning Engineers, Inc.; AO83-10.
World Health Organization. (1985a) Air quality guidelines: indoor air
pollutants. Geneva Switzerland: World Health Organization, Regional
Office for Europe.
World Health Organization. (1985b) Organic solvents and the central nervous
system. Copenhagen, Denmark: World Health Organization; doc. no. 5.
World Health Organization. (1987) Acrylonitrile. In: Air quality guidelines,
volume I. World Health Organization; pp. 1-1 - 1-23.
Wright, C. G.; Jackson, M. D. (1975) Insecticide residues in non-target areas
of rooms after two months of crack and crevice application. Bull.
Environ. Contam. Toxicol. 13: 123-128.
Wright, C. G. ; Leidy, R. B. (1982) Chlordane and heptachlor in the ambient air
of houses treated for termites. Bull. Environ. Contam. Toxicol. 28:
617-623.
Wright, G.; Randell, P.; Shephard, R. J. (1973) Carbon monoxide and driving
skills. Arch. Environ. Health 27: 349-354.
Wright, C. G.; Leidy, R. B.; Dupree, H. E., Jr. (1981) Insecticides in the
ambient air of rooms following their application for control of pests.
Bull. Environ. Contam. Toxicol. 26: 548-553.
Wu, A. H.; Henderson, B. E.; Pike, M. C. ; Yu, M. C. (1985) Smoking and other
risk factors for lung cancer in women. JNCI J. Natl. Cancer Inst. 74:
747-751.
Wynder, E. L.; Hoffmann, D. (1967) Tobacco and tobacco smoke: studies in
experimental carcinogenesis. New York, NY: Academic Press.
Yamanaka, S.; Maruoka, S. (1984) Short communication: mutagenicity of the
extract recovered from airborne particles outside and inside a home with
an unvented kerosene heater. Atmos. Environ. 18: 1485-1487.
Yocom, J. E. (1982) Indoor-outdoor air quality relationships. A critical
review. J. Air Pollut. Control Assoc. 32: 500-520.
Yocom, J. E. ; Stankunas, A. R.; Bradow, F. V. P. (1982) A review of air
pollutant damage to materials. In: Impact of air pollutants on materials
- a report of panel 3: "environmental impact," of the NATO/CCMS pilot
study on air pollution control strategies and impact modelling. Brussels,
Belgium: North Atlantic Treaty Organization Information Service; no. 139.
Yocom, J. E. ; Baer, N. S.; Robinson, E. (1986) Air pollution effects on
physical and economic systems. In: Stern, ed. Air pollution, v. VI.
Academic Press; pp. 145-246.
6-40
-------
Zawacki, T. S., et al. (1984) Efficiency and emission improvements for
gas-fired space heaters. Task 2. Unvented space heaters emissions
reduction. Available from: NTIS, Springfield, VA; PB-237734.
Zawacki, T. S., et al. (1985) Measurements of emissions rates from gas-fired
space heaters. Institute of Gas Technology; interim report, contract no.
5082-241-0772.
Zawacki, T. S., et al. (1986) Measurements of emissions rates from gas-fired
space heaters. Institute of Gas Technology; final report, contract no.
5082-241-0772.
6-41
-------
ATTACHMENT 1. RESPONSE TO ISSUES FROM HARVARD WORKSHOP, JANUARY 1987
At the request of the U.S. EPA's Office of Research and Development,
Harvard University organized a workshop to review ORD's Preliminary Assessment
for the Indoor Air Program. A group of 20 of the world's leading researchers
on indoor air pollution, Harvard staff and EPA personnel participated in this
review. The format of the workshop was that the attendees were divided into
small groups which focused discussion on the information compiled on specific
pollutants. Each group leader then summarized the work of the group in a series
of viewgraphs, which were then presented for discussion by the assembled parti-
cipants at two plenary sessions.
A number of general recommendations were made concerning the document as a
whole, as well as suggestions concerning specific pollutants.
There was consensus that the information in the preliminary assessment
would be more effectively presented in a pollutant category format, similar to
that of WHO publications. Since building design, maintenance, and ventilation
cut across many pollutant categories, yet contribute unique concerns to indoor
air pollution problems, it was also suggested that a separate chapter address-
ing these issues should be included. This would both emphasize the importance
of building-type related concerns, and avoid the redundancy of repeating such
concerns with each relevant pollutant. Still another general suggestion was
that a crude exposure estimate for the various pollutants should be attempted,
based on source prevalence and user patterns, but that the initial approach of
attempting to quantify risk should be avoided. It was also recommended that
summaries which clearly state the major knowns and unknowns should be given for
each pollutant category.
ORD's response to these points has been to reformat the document into
pollutant category form, addressing pollutant characterization, sources and
occurrence, monitoring, exposure, health effects, and control and mitigation
for each pollutant category. Welfare effects were organized into a separate
chapter. Heating, ventilation and air-conditioning (HVAC) and building design
A-l
-------
and maintenance were also considered in a separate chapter. Finally, summaries
were included for each pollutant category. These delineated the major knowns
and unknowns for the category. In addition, a chapter was added which summa-
rized information on a source-category basis. From a research and control
viewpoint, it is clear that pollutants are emitted from sources, and that these
sources must be mitigated in order to reduce pollutants which cause effects on
health and welfare. Since the Research Needs Assessment, which discusses
research options, is organized by source for practical and research purposes,
the final chapter of the Information Assessment serves not only as summary, but
leads into the source-category based Research needs portion.
More specific recommendations were to define clearly and to establish the
distinction between monitoring and exposure, and to address the issue of total
integrated exposure. Indoor air pollution is then to be put in the perspective
of total exposure. The new version of the Information Assessment addresses
these issues in the introduction.
Comments which were addressed to specific pollutants primarily addressed
gaps in information, errors of fact, additional areas of concern, or differing
views on approach to pollutants. Such concerns have been evaluated and the
necessary changes have been made. Finally, an editing for consistency of style
and to remove redundancies was suggested. Both a technical and stylistic
editing have been done.
The primary issue in considering indoor air pollution is reducing risk to
the public by reducing exposure. There have been a number of studies which
calculated exposure to specific pollutants in a variety of indoor environments.
Measurements were made under differing conditions of sampling methodology and
averaging times, so that comparison of results across these studies is problem-
atic. In further iterations of this document attempts to calculate meaningful
exposures across the range of indoor pollutants will be made. In the present,
a sketch of exposures derived from source prevalence information and work
patterns, is provided, which does not attempt to incorporate the multiple
assumptions and complex variables required for exposure calculation or model-
ing. However, since this is a vital issue in estimating risk, a special
workshop addressing the various aspects of indoor air pollutant exposures is
planned. Subsequent iterations of this document can then address this and the
total exposure times. The report from Harvard detailing the findings from the
workshop is attached.
A-2
-------
SUMMARY OF PRINCIPAL FINDINGS FROM HARVARD WORKSHOP ON THE
REVIEW OF THE PRELIMINARY ASSESSMENT DOCUMENT
FOR THE U.S. EPA INDOOR AIR PROGRAMS
John D. Spengler and Haluk Ozkaynak
Energy and Environmental Policy Center
Harvard University
65 Winthrop Street
Cambridge, MA 02138
Prepared for:
Dr. Michael Berry
Environmental Criteria Assessment Office
United States Environmental Protection Agency
Research Triangle Park, NC 27711
March 10, 1987
A-3
-------
CONTENTS
Page
Executi ve Summary 1
I. Background 3
II. Overview Comments 5
III. Comments by Specific Pollutants 8
Radon 8
ETS 9
Nitrogen Dioxide 10
Carbon Monoxide 11
Carbon Dioxide 11
Particles and Fibers 11
Biologicals 12
VOCs 13
Formaldehyde 14
Pesticides 14
IV. Recommendations 15
Tables 18
A-4
-------
EXECUTIVE SUMMARY
This report summarizes the principal findings from a workshop held at
Harvard University on January 27-29, 1987 to review a preliminary assessment
document for the EPA indoor air program. The purpose of the workshop was to
evaluate the document for completeness and to recommend how EPA could best
modify it bothe technically and structurally. A group of more than twenty
scientists plus EPA and Harvard University staff participated in individual
and general discussions to critique sections of the document and its overall
format. Comments made both prior to and at the meeting have been provided to
EPA separate from this report.
The purpose of this report is to summarize the essential comments and
recommendations made during the meeting, it is not an independent assessment
of the preliminary assessment document. Both general and pollutant-specific
comments are presented. We have noted where there was not general agreement
among the participants on certain issues. For example, the entire group of
participants did not agree on how much or in what form risk assessment calcula-
tions should be handled in this document. However, most of the technical
recommendations made by group leaders were acceptable to all the scientists
attending the meeting.
The fore most recommendation, universally agreed upon, was to reformat the
document, and recent WHO publications were suggested as a model. The reviewers
also suggested that the document could be edited to eliminate repetitions, un-
necessary details, and citations of secondary references. For certain pollut-
ants, however, more information needs to be added to the document, especially
on the distribution of population exposures and related technical issues
(e.g., a clearer definition of terms and averaging times, health effects of
peak and chronic exposures, and complex mixture/synergism problems).
Overall, the reviewers complimented the Agency on an excellent job of pro-
ducing a complex and important document within a short period of time. Again,
the major criticisms were directed towards reorganizing the existing material
and presenting the technical information more coherently on the distribution
of population exposures and health effects so that a priority ranking of indoor
environment-related problems and mitigation alternatives could be implemented.
A-5
-------
I. BACKGROUND
On January 27-29, 1987 Harvard University hosted a workshop to review a
draft indoor air assessment document from EPA's Environmental Criteria Assess-
ment Office (ECAO). Twenty scientists were invited to attend the workshop and
a small group of Harvard staff served as facilitators and reporters. A panel
of EPA staff also attended to observe the discussions. All of the workshop
participants are listed in Table 1.
Before the workshop the non-EPA participants were asked to provide short
written comments based on then review of the document and respond to questions
regarding the priority ranking of indoor pollutants. Copies of these materials
were made available to EPA/ECAO.
Michael Berry and Harriet Ammann of EPA/ECAO opened the workshop by de-
scribing the purpose of the document, the reason it was generated, and the
areas on which they would like the reviewers to concentrate.
The first part of the workshop, consisting of an afternoon and morning
session, was devoted to small group discussions. Participants were divided
into groups and given assigned topics. The leaders of the groups were asked
to produce viewgraphs on what transpired in their group's discussions. The
groups were arranged so that participants met with different colleagues in the
two sessions. Table 2 shows the participation and topics covered in the small
groups.
The second part of the workshop, also consisting of an afternoon and
morning session, was devoted to plenary sessions where the leaders of the
small groups presented viewgraphs on the topics their groups had discussed.
The other participants could then comment on topics their groups had not
covered. Copies of the viewgraphs presented in the plenary sessions were
given to EPA/ECAO.
The comments provided below are meant as a summary of the major points
that were made at the workshop. The first section deals with general comments
on the assessment document. The second discusses points made about individual
indoor pollutants.
A-6
-------
II. OVERVIEW COMMENTS
This section draws upon the general discussions conducted in the plenary
sessions as well as the reports from the two subgroups assigned to overview
the document as a whole.
1. There is a need for an introductory section for the EPA Indoor Air
Quality Assessment Document. This section would express the motivation
behind the document and the need to examine indoor pollutants. This
section should discuss the type of health effects that have been
associated with indoor air pollutants. It should mention the number of
people that are potentially exposed to indoor pollutants by
concentrations or by sources. The section should outline the optional
strategies that should be considered in reducing exposures, properly
noting specific problems that different approaches might have. The last
World Health Organization (WHO) indoor air quality report has a section
on the advantages and disadvantages of various regulatory approaches.
2. There should be a section that discusses the physical and chemical proper-
ties of indoor air pollutants. Units of emission and concentrations could
be explained in this section. Furthermore, the issues of averaging times
and concentrations should be explained here as well. For some pollutants
the long-term integrated concentrations are important, while for others
the concern might be for short-term exposurew. Some indoor air pollutants
have dynamic emission rates (e.g., formaldehyde, radon), for these pol-
lutants the relationships between emissions and sampling times should be
pointed out. This section is needed for two reasons: (1) to unify various
units and terms, and (2) to establish the proper perspective for the reader
to interpret data.
3. If the current structure is retained, the following changes are required:
a. Define purpose of each chapter;
b. Include a better concluding section for each chapter;
c. Remove substantial redundancy;
d. Revise mitigation section which is currently inadequate;
A-7
-------
e. Include a section (not a chapter) on modeling. A comprehensive review
of models, however, is not necessary. This section should state the
usefulness and application of models to test theory, estimate para-
meters, simulate concentrations for study design or exposure estima-
tion, and test mitigation strategies.
4. It s recommended that the report be shortened. A format similar to the
WHO reports is suggested. Using this format, the references do not have
to be comprehensive. Rather than using tables of data, the range of con-
centrations, emissions, and typical values would be reported.
5. The report should address integrated exposure assessment. However, Chapter
9 is not the proper approach. It was generally felt that Chapter 9 should
be eliminated. Instead, the Agency should start the process of exposure
assessment. While it may not be possible at this point to develop a full
risk assessment with exposure and potency information on all indoor con-
taminants, the report should state what should be done and what is needed.
The published risk numbers from other reports could be referred to with
commentary about uncertainty.
The Agency should be heading towards an integrated exposure assessment
that differentiates source contributions. It is a worthwhile goal to work
towards the development of attributable risk concepts. For example, of all
environmental factors affecting excess childhood respiratory illness, what
percentage can e attributed to indoor versus outdoor sources? Similarly, from
the population exposure to carcinogenic organic compounds, what is the
attributable risk due to indoor exposures versus hazardous waste site
exposures?
Risk assessment is an ongoing process that should be integral to EPA's
Indoor Air Quality program and should be updated on a regular basis. Risk
assessment could benefit EPA's indoor air quality program by preevaluating the
usefulness of a particular research area and by testing mitigation strategies.
A-8
-------
III. COMMENTS BY SPECIFIC POLLUTANTS
In this section we highlight the essential comments and recommendations
made by workshop participants under each pollutant category. A significant
fraction of these were discussed during the oral presentations of the group
leaders at the workshop. Copies of transparencies listing most of these
comments have already been provided to EPA/ECAO. In the following we
summarize what we believe to be the most essential of these suggestions. It
should be noted, however, that some of the comments listed below are more in
the form of research recommendations rather than specific criticisms of the
draft EPA document. Therefore, such topics may need to be acknowledged in the
report instead of expending a great deal of effort at this time in assessing
the present state of knowledge regarding such issues.
The following lists comments, criticisms, and recommendations by
pollutant type directed at the document reviewed.
Radon
1. This section needs to be rewritten by an EPA expert on radon.
2. Welfare effects should be identified (i.e., impact on property values).
It is expected that the needs of the real estate community (old and new
homes) will become an important public issue. Furthermore, we need to
address the new construction of houses quickly.
3. Health data need to be developed. This should include delineation of
effects due to attached and unattached fraction.
4. Epidemiological studies are needed to evaluate the effects of chronic,
low-level exposures to radon to assess mitigation strategies.
5. Inevitably, air cleaners will be used as a low-cost alternative to radon
mitigation for secondary mitigation, and temporary mitigation. Also, air
cleaners are becoming much more prevalent in homes today and their effect
on health effects of radon must be known.
6. Research needs that should be emphasized also include:
Interpretation of seasonal and structural factors affecting radon
concentrations;
Dose/response relationships;
Complex mixture interactions;
A-9
-------
Interpretation of recent epidemiology and new animal studies that
are being published or underway; and
Available mitigation options.
ETS
1. The report needs discussion of:
The specific compounds of interest;
Space/time considerations;
Identification of surrogates;
Bio-markers;
Questionnaires; and
Measurement approaches.
2. Temporal and spatial variability of exposures is an information gap that
needs to be addressed.
3. Breakdown of exposure data by location and time.
4. Better characterization of:
Differences in MS/SS/ETS;
Interactions;
Decay and fate of contaminants;
Health effects on children; and
Adequacy of air cleaners and restrictions in public access buildings.
Nitrogen Dioxide
1. Multiple sources need to be mentioned.
2. Appliance misuse and improper venting of furnace, heaters, water heaters,
and dryers may be substantial contributor to higher indoor NCL concentra-
tions.
3. Measurement and better definition of short-term and peak NO^ exposures
should be undertaken.
4. Figure 5.1 should be deleted (dated material).
5. All combustion appliances should not be lumped together.
6. Because of increased use of kerosene heaters, the health effects of SOp
and other byproducts of kerosene heaters need to be addressed.
7. Use of LA data should be cleared with Phil Baker of SoCal Gas.
8. Boston data are not available to public and should not be cited. Instead,
the document could use Portage, WI data or Watertown, MA data as enclosed.
A-10
-------
9. Topics for further research and discussion include:
N02 adsorption effects (fate and reaction products not well known);
Averaging time for health effects of NO^ and dose-response data;
Better characterization of tails of the distribution of population
exposures to NC^;
Collecting field emission data on NO,, source strengths for space and
kerosene heaters; and
Evaluation of various mitigation options.
Carbon Monoxide
1. The document should mention that although CO is the only indoor pollutant
that is known to be fatal, it is not considered to be a critical indoor
health problem.
2. The document should mention the issue of concerns with the backdrafting
of exhaust flue gases into homes resulting in elevated CO levels in homes
but also in high concentrations of other combustion products as well.
3. The document should recognize the fact that a number of indoor chronic CO
poisoning can go undetected due to a specificity of the symptoms.
However, it is not likely that the small number of homes having CO
problems can easily be detected even in large-scale monitoring studies.
Carbon Dioxide
I. C0? should be added as a separate section or under the HVAC section.
2. The document should include a discussion of the physical and chemical
factors affecting C02 concentrations (ventilation, infiltration,
transportation, and sinks).
3. It also should note the importance of building factors: design,
operation, and maintenance monitoring.
Particles and Fibers
1. There is no need to redo what is in criteria documents.
2. The section on particles has technical errors and omissions. It is
suggested that it should be structured along the WHO air quality
guidelines document or the document on asbestos.
3. EPA should also utilize other Agency sources (e.g., WHO, NCRP, ASTM, DOE)
in summarizing important data on exposures and health effects in
comparable format.
A-ll
-------
4. Guidelines on air cleaners or a program to define the value of these
devices should be included.
5. Man-made-mineral fibers need to be included in the document.
6. The EPA asbestos program needs careful evaluation. Problems will occur
when it is recognized that home exposures may be very large following
asbestos removal.
Biologicals
1. The document needs to indicate that aeropathogens and aeroallergens are
major contributing cause of acute morbidity.
2. Terms and sampling systems should be defined.
3. The major needs in this field are for:
Baseline data on concentrations and health effects of these
pollutants;
Standardization of measurement methods;
Clear definition of humidifier fever and other allergic illnesses;
Mycotoxin and endotoxin health effects and their carcinogenic
potential (use of antimicrobials may have unwanted side effects);
and
Education of public, building designers, engineers, and supervisors
regarding reservoirs, amplifiers, and disseminators of these
pollutants.
VOCs
1. Information contained in Table 3-6 should be synthesized better (give
concentration values, not just references).
2. The difference between monitoring and exposure should be clarified.
3. The difficulties with "unknown" VOCs and the heterogeneous chemical
composition of VOCs should be mentioned.
4. Information on averaging times for the data presented should be provided.
5. Review of sources of VOCs should include outside sources as well as
commuting exposures. Automobile emissions should be included as
important sources of outdoor as well as indoor organic pollutant
exposures.
6. A table should be provided to show an overview of emitted organics by
source type.
A-12
-------
7. The possibility of biological markers and breath samples as indicators of
exposure needs to be mentioned.
8. The qualitative character of Table 6-16 should be stressed. Remove "use"
column from this table.
9. A new section should be written to discuss modeling.
10. Areas where scientific knowledge is insufficient should be identified,
such as:
Source emission characterization;
Acute and chronic health effects;
Odor and sick building syndrome problems;
Toxicology of mixes;
Better instrumentation and/or surveys; and
Effects of temperature, humidity, and ventilation.
11. Mitigation or control considerations should be emphasized: source
removal, consumer education, and product use, substitution
Formaldehyde
1. The document should refer to other source documents since scientific
knowledge on formaldehyde sources and exposures even effects is quite
good.
2. A reference to the Canadian work should be added in Table 6-13 and
footnotes to Table 6-12.
Pesticides
1. Section is disjointed as now written. The knowledge of pesticides is in
general adequate, but the chapter is not.
2. The document should summarize and reference available documents.
3. The document should give some guidance of where exposures may be the most
important (offices, restaurants, etc.).
4. Information on biological markers should be included in the exposure
section.
5. Section 6.12 is inadequate and should be rewritten.
6. Occupational data on toxicity should be included so that extrapolations
for the general population can be made.
A-13
-------
7. A summary of observed health effects should be included. If different
effects in various groups, such as infants, are expected, they should be
noted.
8. A summary of available health effects data at different pesticide concen-
trations should be presented.
A-14
-------
IV. RECOMMENDATIONS
The reviewers of the document all recognized the challenge faced by EPA
in putting together an ambitious document within a short period of time.
Although the document is quite extensive in its coverage, it suffers from
"multiple-author syndrome." The most important suggestion made by the
reviewers was the need to restructure the document to avoid repetition and
the superfluity of details that could be found elsewhere. The reviewers
thought the document needed an introductory section discussing the purpose and
key issues of concern in terms of population exposures, potential health
effects, and available control strategies. Also suggested was the possibility
of considering an alternative format similar to those used in recent WHO
reports.
Regarding the pollutant-specific sections, there were a number of
recommendations for displaying exposure and effect information more clearly.
Issues of consistency in units, terminology, synergisms, and measurement
methods were also raised. The need for better characterizing time issues and
peak exposure effects was also pointed out (VOC, radon, N0?). For certain
pollutants (particles and formaldehyde) briefer presentations supported with
primary references or more complete information sources were suggested.
However, for the pesticide and radon sections additional information on
exposures and health effects by population groups were recommended.
Another suggestion was to include a new or a separate section on modeling
or building systems that includes modeling, ventilation, welfare, comfort,
energy, and other related issues. Adding a discussion on indoor C02 and its
relationship to HVAC systems was also considered by many reviewers to be
worthwhile.
The participants agreed that a more complete presentation of the
distribution of population exposures to various pollutants would be useful.
Recommended in particular was a graphical display of concentrations or expo-
sures plotted against the number of people exposed in order to aid the
evaluation of the extent of the indoor air problem. However, there was no
consensus among the workshop participants as to whether the Agency should
carry out the next logical step, namely, to perform a population-based risk
assessment using the exposure distributions and the health-risk potencies.
A-15
-------
One group was reluctant to consider a (risk-based) ranking methodology for
pollutants. They thought that no present methodology would allow an assess-
ment that would yield meaningful results, that is, within reasonable confi-
dence/error limits. Another group, however, suggested putting bounds on the
estimates of the number of populations at risk and identifying control
alternatives by examining the tails of the distributions (for exposures as well
as for percentage of population affected) and comparing these to the total
exposures or risks. Consideration of error and uncertainty of estimates
(though formally complex to implement) was also recommended.
It is clear that display of exposures by population groups or by region
would be helpful in identifying the magnitude of a certain indoor air
pollution problem. Although a proper risk assessment application should be
among one of the key goals of an indoor air research program, it could lead to
serious controversy if hastily or incorrectly done. As we mentioned above, we
suggest a progressive effort in this area, rather than calculating numerical
mortality or morbidity risk estimates without first developing a basic frame-
work for risk assessment. Qualitative ranking of information, exposures, poten-
tial risks, and priorities for various mitigation options should perhaps be
the starting point. However, while the information bases on exposures and
effects are being developed, the Agency should also investigate how best to
quantify the nature and magnitude of health risks resulting from exposure to
pollutants of both indoor and outdoor origin.
A-16
-------
TABLE 1. PARTICIPANTS IN WORKSHOP
TO REVIEW EPA INDOOR AIR ASSESSMENT DOCUMENT
Dr. Harriet Ammann
MD-52
Environmental Criteria Assessment Office
U.S. EPA
Research Triangle Park, NC 27711
919-541-4930
Dr. Neil Benowitz
Building 30, 5th Floor
San Franscisco General Hospital
1001 Potrero Avenue
San Francisco, CA 94110
415-821-8324
Dr. Michael Berry
MD-52
Environmental Criteria Assessment Office
U.S. EPA
Research Triangle Park, NC 27711
919-541-4172
Dr. Irwin Billick
Gas Research Institute
8600 West Bryn Mawr
Chicago, IL 60631
312-399-8304
Dr. Irwin Broder
The Gas Research Institute
223 College Street
Toronto M5T 1R4
CANADA
416-979-2744
Dr. Bert Brunekreef
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02215
617-732-1244
Dr. Harriet Burge
Box 0529
University of Michigan Medical Center
Ann Arbor, MI 48109-0529
313-764-0227
Dr. Daniel Costa
MD-82
Health Effects Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
A-17
-------
Dr. Robert Elias
MD-52
Environmental Criteria Assessment Office
U.S. EPA
Research Triangle Park, NC 27711
Dr. Judy Graham
MD-51
Health Effects Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Dr. David Grimsrud
221 San Carlos Avenue
Piedmont, CA 94611
415-486-6593
Dr. Alan Hawthorne
Bldg. 4500 S. MS-101
Oak Ridge National Laboratory
P.O. Box X
Oak Ridge, TN 37830
615-574-6246
Mr. Jim Kawecki
TRC
2121 Wisconsin Avenue, NW
Suite 220
Washington, DC 20007
Dr. Brian Leaderer
Pierce Foundation Laboratory
Yale University School of Medicine
290 Congress Avenue
New Haven, CT 06519
203-362-9901
Dr. Michael Lebowitz
University of Arizona
Health Science Center
Tucson, AZ 85724
602-626-6379
Dr. Richard Letz
Division of Environmental and Occupational Medicine
Mt. Sinai School of Medicine
10 East 102 Street
New York, NY 10029
212-348-1006
212-650-6173
A-18
-------
Dr. JoEllen Lewtas
MD-68
Health Effects Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Dr. Morton Lippman
Institute for Environmental Medicine
New York University Medical Center
Long Meadow Road
Tuxedo, NY 10987
914-351-5277
Dr. Joyce McCann
1235 Glen Avenue
Berkeley, CA 94708
415-642-4760
Mr. Jack McCarthy
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02215
617-732-0827
Dr. Lars Molhave
Cand. Scient.
Hygiejnisk Institut
Bygning 181
Uni vers i tetsparken
Arhus Universitetsparken
DK-8000 Arhus C
DENMARK
011-45-6-128288
Dr. Philip Morey
MN10-1451
Honeywell Indoor Air Quality Diagnostics
1985 Douglas Drive North
Golden Valley, MN 55422-3992
612-542-7069
Dr. David Otto
MD-58
Health Effects Research Laboratory
U.S. EPA
Research Triangle Park, NC 27711
Dr. Haluk Ozkaynak
EEPC
Harvard University
65 Winthrop Street
Cambridge, MA 02138
617-495-1313
A-19
-------
Dr. Peter W. Rand
Research Department
Maine Medical Center
Portland, ME 04102
207-871-2163
Dr. James Repace
ANR-443
U.S. EPA
Washington, DC 20460
Dr. Charles Rodes
MD-56
EMSL
U.S. EPA
Research Triangle Park, NC 27711
Dr. Steven Rudnick
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
617-732-1162
Dr. P. Barry Ryan
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
617-732-1167
Dr. Bernd Seifert
Institute for Water, Soil, and Air Hygiene
Corrensplatz 1
D-1000 Berlin 33
WEST GERMANY
011-49-30-8308-2734
Dr. Kirk Smith
Resource Systems Institute
East-West Center
1777 East-West Road
Honolulu, HI 96848
808-944-7519
Dr. Leslie Sparks
MD-54
AEERL
U.S. EPA
Research Triangle Park, NC 27711
Dr. John D. Spengler
Department of Environmental Sciences and Physiology
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
617-732-1255
A-20
-------
Dr. Jan Stolwijk
Yale University School of Medicine
Pierce Foundation
Laboratory
290 Congress Avenue
New Haven, CT 06519
203-785-2867
Dr. Gene Tucker
MD-54
AEERL
U.S. EPA
Research Triangle Park, NC 27711
Mr. William Turner
c/o Harriman Associates
292 Court Street
Auburn, ME 04210
207-784-5728
Mr. Matt van Hook
1133 N. Harrison Street
Arlington, VA 22205
Dr. Douglas Walkinshaw
Environmental Health Center
Room 206
Tunney's Pasture
Ottawa, Ontario
CANADA K1A OL2
613-957-1502
Dr. Lance Wallace
RD-680
U.S. EPA
401 M Street, SW
Washington, DC 20460
202-382-5792
Dr. Yukio Yanagisawa
Dept. of Environmental Sciences and Physiology
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
617-732-1165
Mr. John Yocom
TRC Environmental Consultants
800 Connecticut Blvd.
East Hartford, CT 06108
203-289-8631
A-21
-------
TABLE 2. GROUP ASSIGNMENTS AND RESPONSIBILITIES ON JANUARY 27, 1987
Group Leaders and Members
Billick Leaderer Lebowitz
McCann
Grimsrud
Benowitz
Lippmann
Rand
Turner
Yocum
Burge
Broder
Harvard University Participants
Brunekreef
EPA Observers
Rudnick
Yanagisawa
Focus of Review
Combustion
Products-
CO, N02
Pesticides
Particles
Particles
Combustion
Products-
CO, N02
Formaldehyde
Formaldehyde
Particles
Pesticides
Seifert
Stolwijk
Letz
Morey
McCarthy
Graham
Tucker
Elias
Lewtas
Rodes
Costa
Sparks
Otto
Repace
van Hook
Wallace
Tichenor
Kawecki
Pesticides
Combustion
Products-
CD, N02
Formaldehyde
Hawthorne
Molhave
Smith
Wai kinshaw
Ryan
No EPA
Document
Overview
A-22
-------
TABLE 2. GROUP ASSIGNMENTS AND RESPONSIBILITIES ON JANUARY 28, 1987
Group Leaders
Billick
Yocum
Lippmann
Stolwi jk
and Members
Leaderer
Benowitz
Smith
Walkinshaw
Lebowitz
Morey
Burge
Broder
Seifert
Molhave
McCann
Letz
Hawthorne
Rand
Turner
Grimsrud
Harvard University Participants
Brunekreef
EPA Observers
No EPA
McCarthy
Lewtas
Repace
Rodes
Kawecki
Ryan
Costa
Graham
Sparks
Yanagisawa
Tucker
Wallace
Otto
Rudnick
van Hook
Tichenor
Eli as
Focus of Review
Document
Overview
Summary of
Questions
Responses
ETS
Organics
Radon
Biological
ETS
Organics
Organics
Radon
Biological
Radon
Biological
ETS
A-23
U.S. GOVERNMENT PRINTING OFFICE: 1987- 7 "» 8 - 1 2 1 / 67013
------- |