[Federal Register Volume 59, Number 65 (Tuesday, April 5, 1994)]
[Unknown Section]
[Page 0]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 94-7619]
[[Page Unknown]]
[Federal Register: April 5, 1994]
_______________________________________________________________________
Part II
Department of Labor
_______________________________________________________________________
Occupational Safety and Health Administration
_______________________________________________________________________
29 CFR Parts 1910, 1915, 1926, and 1928
Indoor Air Quality; Proposed
Rule
DEPARTMENT OF LABOR
Occupational Safety and Health Administration
29 CFR Parts 1910, 1915, 1926, 1928
[Docket No. H-122]
RIN 1218-AB37
Indoor Air Quality
AGENCY: Occupational Safety and Health Administration (OSHA), Labor.
ACTION: Notice of proposed rulemaking; notice of informal public
hearing.
-----------------------------------------------------------------------
SUMMARY: By this notice, the Occupational Safety and Health
Administration (OSHA) proposes to adopt standards addressing indoor air
quality in indoor work environments. The basis for this proposed action
is a preliminary determination that employees working in indoor work
environments face a significant risk of material impairment to their
health due to poor indoor air quality, and that compliance with the
provisions proposed in this notice will substantially reduce that risk.
The provisions of the standard are proposed to apply to all indoor
``nonindustrial work environments.'' In addition, all worksites, both
industrial and nonindustrial within OSHA's jurisdiction are covered
with respect to the proposed provisions addressing control of
environmental tobacco smoke. The proposal would require affected
employers to develop a written indoor air quality compliance plan and
implement that plan through actions such as inspection and maintenance
of building systems which influence indoor air quality.
Provisions under the standard also propose to require employers to
implement controls for specific contaminants and their sources such as
outdoor air contaminants, microbial contamination, maintenance and
cleaning chemicals, pesticides, and other hazardous chemicals within
indoor work environments. Designated smoking areas which are to be
separate, enclosed rooms exhausted directly to the outside are proposed
to be required in buildings where the smoking of tobacco products is
not prohibited. Specific provisions are also proposed to limit the
degradation of indoor air quality during the performance of renovation,
remodeling and similar activities. Provisions for information and
training of building system maintenance and operation workers and other
employees within the facility are also included in this notice.
Finally, proposed provisions in this notice address the
establishment, retention, availability, and transfer of records such as
inspection and maintenance records, records of written compliance
programs, and employee complaints of building-related illness.
The Agency invites the submission of written data, views and
comments on all regulatory provisions proposed in this notice, and on
all relevant issues pertinent to those provisions. OSHA is also
scheduling an informal public hearing where persons may orally submit
their views. It is noted here that subsequent Federal Register notices
may be published subsequent to this notice, if the public presents
views leading to a substantial change in focus or it is otherwise
determined to be appropriate.
DATES: Comments on the proposed standard must be postmarked by June 29,
1994. Notices of intention to appear must be postmarked by June 20,
1994. Testimony and evidence to be submitted at the hearing must be
postmarked by July 5, 1994. The hearing will commence at 9:30 a.m. on
July 12, 1994.
ADDRESSES: Comments are to be submitted in quadruplicate or 1 original
(hardcopy) and 1 disk (5\1/4\ or 3\1/2\) in WP 5.0, 5.1, 6.0 or Ascii
to: The Docket Office, Docket No. H-122, Room N-2625, U.S. Department
of Labor, 200 Constitution Avenue, NW., Washington, DC 20210, Telephone
No. (202) 219-7894. (Any information not contained on disk, e.g.,
studies, articles, etc., must be submitted in quadruplicate.)
Notices of intention to appear and testimony and evidence are to be
submitted in quadruplicate to: Mr. Tom Hall, Division of Consumer
Affairs, Occupational Safety and Health Administration, 200
Constitution Avenue, NW., room N3649, Washington, DC 20210; (202) 219-
8615.
The hearing will be held in the auditorium of the U.S. Department
of Labor, 200 Constitution Avenue, NW., Washington, DC.
FOR FURTHER INFORMATION CONTACT: Proposal: Mr. James F. Foster,
Director of Information and Consumer Affairs, Occupational Safety and
Health Administration, 200 Constitution Avenue, NW., room N3641,
Washington, DC 20210; (202) 219-8151.
Informal Hearing Information: Mr. Tom Hall, Division of Consumer
Affairs, Occupational Safety and Health Administration, 200
Constitution Avenue, NW., room N3649, Washington, DC 20210; (202) 219-
8615.
Table of Contents
I. Supplementary Information
A. Events Leading to This Action
II. Health Effects
A. Sick Building Syndrome
B. Building-Related Illness
1. Indoor Air Contaminants
2. Microbial Contaminants
C. Environmental Tobacco Smoke
1. Pharmacokinetics
(a) Absorption and Distribution
(b) Metabolism
2. Irritation
3. Pulmonary Effects
4. Cardiovascular Effects
(a) Thrombus Formation
(b) Vascular Wall Injury
(c) Possible Mechanisms of Effect
(d) Acute Heart Effects
(e) Chronic Heart Effects
5. Reproductive Effects
6. Cancer
(a) Evidence of Association
(b) Epidemiological and Experimental Studies
7. Genotoxicity
8. Conclusions
D. Case Reports
1. Sick Building Syndrome and Building-Related Illness
2. Environmental Tobacco Smoke
III. Exposure
A. Sources of Indoor Air Contaminants
B. Microbial Contamination
C. Exposure Studies
1. Low-level Contaminants
2. Bioaerosols
3. Environmental Tobacco Smoke
(a) Chemistry
(b) Human Activity Pattern Studies Used to Assess Workplace
Exposure
(c) Indoor Levels of Environmental Tobacco Smoke Constituents
(d) Levels of Respirable Suspended Particulates and Nicotine
Found in Field Studies
(e) Biomarkers of Environmental Tobacco Smoke Exposure
(f) Inadequacy of General Dilution Ventilation to Address
Environmental Tobacco Smoke Exposure Control
IV. Preliminary Quantitative Risk Assessment
A. Introduction
B. Review of Epidemiologic Studies and Published Risk Estimates
C. Data Sources
D. OSHA's Estimates of Risk-Environmental Tobacco Smoke Exposure
E. OSHA's Risk Estimates--Indoor Air Quality
F. Pharmacokinetic Modeling of Environmental Tobacco Smoke
Exposure
1. Considerations for Selection of a Biomarker for Environmental
Tobacco Smoke
2. Cardiovascular Effects
3. Carcinogenicity
4. Evaluation of Cotinine as a Biomarker for Environmental
Tobacco Smoke
5. Description of Pharmacokinetic Models for Nicotine and
Cotinine
6. Application of Pharmacokinetic Modeling for Environmental
Tobacco Smoke Exposure Estimation
7. Analysis of Uncertainty
(a) Physiological Parameters
(b) Distribution Parameters
(c) Kinetic Parameters
V. Significance of Risk
A. Environmental Tobacco Smoke
B. Indoor Air Quality
VI. Preliminary Regulatory Impact Analysis
A. Introduction
B. Industry Profile
1. Affected Industries
2. Indoor Contaminants-Sources
3. Controlling Indoor Air
4. Building Characteristics
5. Profile of Affected Buildings
6. Buildings with Indoor Air Problems
7. Number of Employees Affected
8. Environmental Tobacco Smoke
(a) Smoking Ordinances and Policies
(b) Number of Nonsmokers Working Indoors
C. Nonregulatory Alternatives
1. Introduction
2. Market Imperfections
3. Alternative Nonregulatory Options
(a) Tort Liability
(b) Workers' Compensation
4. Conclusion
D. Benefits
1. Indoor Air Quality
2. Environmental Tobacco Smoke
3. Cost Savings
(a) Worker Productivity
(b) Property Damage, Maintenance and Cleaning Costs
E. Technological Feasibility and Compliance Costs
1. Technological Feasibility
2. Compliance Costs
(a) Developing Indoor Air Quality Compliance Programs
(b) Indoor Air Quality Operation and Maintenance Program
(c) Training for HVAC Maintenance Workers and Informing
Employees About the Indoor Air Quality Standard
(d) Compliance with Related Standards
(e) Air Contaminant-Tobacco Smoke
(f) Air Quality During Renovation and Remodeling
F. Economic Impact and Regulatory Flexibility Analysis
1. Economic Feasibility
2. Regulatory Flexibility Analysis
3. Environmental Impact
VII. Summary and Explanation
A. Scope and Application: Paragraph (a)
B. Definitions: Paragraph (b)
C. Indoor Air Quality Compliance Program: Paragraph (c)
D. Compliance Program Implementation: Paragraph (d)
E. Controls for Specific Contaminant Sources: Paragraph (e)
F. Air Quality During Renovation and Remodeling: Paragraph (f)
G. Employee Information and Training: Paragraph (g)
H. Recordkeeping: Paragraph (h)
I. Dates: Paragraph (i)
J. Appendices: Paragraph (j)
K. Specific Issues
VIII. State-Plan Standards
IX. Federalism
X. Information Collection Requirements
XI. Public Participation
XII. List of Subjects in 29 CFR Parts 1910, 1915, 1926, and 1928
XIII. Authority and Signature
XIV. Part 1910, 1915, 1926, 1928--Proposed Occupational Safety and
Health Standards
Supplementary Information
A. Events Leading to This Action
Concern about the health hazards posed by occupational exposure to
environmental tobacco smoke (ETS) prompted three public interest groups
to petition the Agency in May 1987 for an Emergency Temporary Standard
under section 6(c) of the Occupational Safety and Health (OSH) Act, 29
U.S.C. 655(c). The American Public Health Association and Public
Citizen submitted a joint petition; Action on Smoking and Health (ASH)
also submitted a petition. The petitions requested the prohibition of
smoking in most indoor workplaces.
OSHA determined, that available data with respect to exposures were
insufficient to demonstrate the existence of a ``grave danger,'' within
the meaning of section 6(c) of the OSH Act, from workplace exposure to
ETS. OSHA denied the petitions in September 1989 but continued to
investigate regulatory options.
In October 1989 ASH filed suit in the U.S. Court of Appeals for the
District of Columbia Circuit for review of OSHA's denial of its
petition for an Emergency Temporary Standard. The court denied ASH's
petition for review in May 1991, finding that OSHA has reasonably
determined that it could not sufficiently quantify the workplace risk
associated with tobacco smoke to justify an Emergency Temporary
Standard.
OSHA issued on September 20, 1991, a Request for Information (RFI)
(56 FR 47892) on indoor air quality problems, in order to obtain
information necessary to determine whether it would be appropriate and
feasible to pursue regulatory action concerning Indoor Air Quality
(IAQ). Issues on which comments were requested in the RFI included
health effects attributable to poor IAQ, ventilation systems
performance, exposure assessment, and abatement methods. Information
concerning specific contaminants such as ETS and bioaerosols was also
requested.
In March 1992, the AFL-CIO petitioned OSHA to promulgate an overall
IAQ standard. OSHA responded in May 1992 that such a standard was under
consideration.
In response to the RFI, over 1,200 comments were submitted by
interested persons, groups, unions, and industries. Issues of
particular concern identified in the comments, in addition to health
effects considerations, include the lack of ventilation performance
standards; the lack of worker training on the operation and maintenance
of Heating Ventilation and Air Conditioning (HVAC) systems; the lack of
pollutant source control; and the lack of available technical guidance
on IAQ issues and control techniques.
Of the comments that specifically addressed the question of whether
OSHA should regulate IAQ, a majority (75%) indicate support for
regulation. Of those that commented on the need for regulation,
approximately 21% were explicitly in favor of a regulation on ETS, more
than 41% were in favor of an overall IAQ regulation, and approximately
13% were in favor of a combined IAQ regulation.
Numerous comments focused on the adverse health effects of tobacco
smoke and of general indoor air pollution. The health effects of
concern relevant to both tobacco smoke and indoor air pollutants ranged
from the acute irritant effects to cancer.
Comments submitted in response to the RFI indicated wide support
for a regulatory approach that would focus on the design, operation and
maintenance of building ventilation systems, source reduction
methodology, and worker information and training programs. Commenters
also recommended that provisions should require that employers receive
training about the regulation and the need for compliance, and that
their training regarding building HVAC maintenance and operation be
tailored to the level of complexity of the HVAC system and their
personal degree of involvement.
Many commenters particularly felt that regulation of IAQ was
necessary to eliminate exposures to ETS in the workplace. Commenters
urged the Agency to either ban smoking completely from the workplace or
allow smoking only in separately ventilated, designated smoking areas
that were separate from work areas.
OSHA believes that data submitted to the record, and other
evidence, support the conclusion that air contaminants and other air
quality factors can act to present a significant risk of material
impairment to employees working in indoor environments. Adverse health
effects associated with poor IAQ may include sensory irritation,
respiratory allergies, asthma, nosocomial infections, humidifier fever,
hypersensitivity pneumonitis, Legionnaires' disease, and the signs and
symptoms characteristic of exposure to chemical or biologic substances
such as carbon monoxide, formaldehyde, pesticides, endotoxins, or
mycotoxins.
The Agency believes that available data support proposing
regulation of IAQ, including exposure to ETS. Further stimulus for this
determination was provided by conclusions reached in a report published
in December, 1992 by the Environmental Protection Agency, addressing
hazards associated with exposure to ETS. In that study, Respiratory
Health Effects of Passive Smoking: Lung Cancer and Other Disorders [Ex.
4-311], EPA concluded that exposure to ETS presents an excess risk of
induction of cancer in humans. OSHA has submitted this proposed
standard to the U.S. Environmental Protection Agency which is reviewing
it in detail for purposes of submitting detailed comments to the
docket.
For the reasons noted above, and discussed in the following
sections, OSHA is proposing to address indoor air quality problems,
including exposure to ETS, as set forth in this notice.
II. Health Effects
Indoor air quality problems can occur in all types and ages of
buildings; in newly constructed buildings, in renovated or remodeled
buildings, and in old buildings. Problems in new, clean buildings are
rarely, if ever, related to microbial growth, since the physical
structures are new [Ex. 3-61]. Older buildings that have not been
adequately maintained and operated may have problems with bioaerosols
if parts of the building have been allowed to become reservoirs for
microbial growth. Also, if inadequate outside air is provided,
regardless of the age of the building, chemical and biological
contaminants will build up to levels that can cause health effects in
some workers. In addition, other physical factors such as lack of
windows, noise, and inadequate lighting, and ergonomic factors
involving uncomfortable furniture and intensive use of video display
units, etc., will cause discomfort in occupants that may be
inaccurately attributed to air quality.
Some information contained in the docket indicates that these
chronic health complaints are psychological, however, OSHA believes
that chronic health complaints related to poor indoor air quality are
unlikely to be due to mass psychogenic illness, even though a
psychological overlay is common. It is true that poor management,
boring work, poor lighting conditions, temperature variations, poor
ergonomic design, and noise may all lower the threshold for complaint.
Nevertheless, air quality complaints usually have some basis, although
they are often difficult to assess with specificity [Exs. 3-61C, 4-
144].
Indoor air quality problems are generally classified as Sick
Building Syndrome (SBS) or Building-related Illness (BRI). However, a
very important constituent of poor indoor air quality is ETS because of
the serious health effects that result from exposure. The following
discussion will first identify the health effects associated with SBS
and BRI. A discussion of the health effects associated with exposure to
ETS will follow.
It is important to note that OSHA considers these health effects to
be material impairments of health when the worker is clinically
diagnosed with a condition that is either caused or aggravated by poor
indoor air quality in the workplace. For example, in the formaldehyde
standard (29 CFR 1910.1048) [Ex. 4-107] OSHA determined that a
physician's diagnosis of irritation met the requirement of material
impairment of health. In addition, OSHA considers all the other health
effects discussed, which are more clinically severe than irritation, to
be material impairments of health as well.
A. Sick Building Syndrome
Typically, health effects caused by poor indoor air quality have
been categorized as SBS or BRI. In 1983, the World Health Organization
published a list of eight non-inclusive symptoms that characterize Sick
Building Syndrome [Ex. 4-325]. These include irritation of the eyes,
nose and throat; dry mucous membranes and skin; erythema; mental
fatigue and headache; respiratory infections and cough; hoarseness of
voice and wheezing; hypersensitivity reactions; and nausea and
dizziness. Generally, these conditions are not easily traced to a
specific substance, but are perceived as resulting from some
unidentified contaminant or combination of contaminants. Symptoms are
relieved when the employee leaves the building and may be reduced or
eliminated by modifying the ventilation system. Comments to the docket
indicate that such symptoms have been observed in and reported by
workers [Exs. 3-446, 4-87].
In some instances, outbreaks of SBS are identified with specific
pollutant exposures, but in general only general etiologic factors
related to building design, operation and maintenance can be identified
[Ex. 4-274]. In 1987, Woods et al. [Ex. 3-745] conducted a stratified
random telephone survey of 600 U.S. office workers across the national.
Twenty four percent reported that they were dissatisfied with the air
quality at the office; while 20% perceived their performance to be
hampered by poor indoor air quality. Women were nearly twice as likely
to report a productivity effect of poor indoor air quality than men
(28% versus 15%). Based on this, Woods et al. [Ex. 3-745] hypothesized
that 20% of U.S. office workers are exposed to indoor conditions which
manifest as SBS. In fact, complaints about SBS have become so numerous
that 37 out of 53 states and territories have designated a building
complaints investigation contact person [Ex. 4-310].
Breysse [Ex. 4-32] reported on symptoms associated with new
carpeting in a state office building, in order of prevalence: headache,
eye and throat irritation, nausea, dizziness, eye tearing, chest
tightness, diarrhea, cough, muscle aches, burning nose, fatigue, dark
urine, and rashes. Twenty out of 35 persons were affected. Air sampling
was conducted before and after carpet removal; a similar range of
aliphatic hydrocarbons was found after removal, but in much lower
concentrations. Many individuals who believe the building they work in
is implicated in SBS, have described similar effects. Symptoms usually
include one or more of the following: mucous membrane (eye, nose, or
throat) irritation, dry skin, headache, nausea, fatigue, and lethargy
[Ex. 4-293]. These symptoms are generally believed to result from
indoor air pollution. There is no secondary spread of symptoms to
others outside the building who are exposed to the occupants (unlike
the situation faced by many chemical and asbestos workers). Anderson
[Ex. 4-10] suggested the possible causes for SBS as related to
psychosocial, chemical, physical, or biological factors.
Anderson [Ex. 4-10] distinguished SBS symptoms as different from
mass psychogenic illness; although in general the causes of SBS are
unknown, he suggested that most SBS symptoms could be explained by
stimulation of sensory nerve fibers in the upper airways and the face
(referred to as common chemical sense). Because these fibers can
respond in only one way, SBS cases largely have the same symptoms
irrespective of the cause [Ex. 4-10].
It is now known that there is a variety of important health effects
from indoor air pollution. In addition to the indoor environmental
disease caused by infectious agents, carcinogens or toxins; the indoor
environment may create conditions that can produce skin and mucosal
allergy and hyperactivity reactions, sensory effects (odors and
irritations), airways effects (from both acute and chronic exposures),
neuropsychological effects, and psychosocial effects, especially due to
the lack of social support [Ex. 4-200].
Indoor air pollution may be caused by physical, chemical, or
microbiological agents, and is aggravated by poor ventilation. The
causation of SBS by indoor air pollution was first objectively
demonstrated in 1984 in a study of 62 Danish subjects suffering from
``indoor climate symptoms'' [Ex. 4-20]. These subjects reported
primarily eye and upper respiratory irritation, but were otherwise
healthy individuals, and did not suffer from asthma, allergy, or
bronchitis. The subjects were exposed to a mixture of 22 volatile
organic chemicals commonly found in the indoor environment at
concentrations of 0, 5, and 25 mg/m3. These concentrations
corresponded respectively to ``clean'' air, average polluted air in
Danish houses, and maximum polluted air in Danish houses. After
exposure, the Digit Span test was administered. The Digit Span test
consists of the subject being allowed to view a series of random digits
for a short period of time; the numbers are then covered up and the
subject asked to repeat the sequence backwards. This test is reported
to be sensitive to situational anxiety and alertness, and therefore a
measure of stress and ability to concentrate. Bach et al. found
significant declines in performance on the digit span test following
exposure to these low levels of volatile organic chemicals,
demonstrating objectively the existence of SBS [Ex. 4-20].
Molhave et al. [Ex. 4-228], in reporting on the same 62 subjects,
found that subjects exposed for 2\3/4\ hrs did not adapt, and that the
subjects reacted to irritation of the mucous membranes and not to odor
intensity. The exposure was doubled-blind, and neither the subjects nor
the testers knew the exposure.
Although these problems have been demonstrated to be real, they may
affect only a small percentage of building occupants. Also, there are
various degrees of problems which may occur. Some individuals who
experience relatively mild and treatable symptoms such as headache, may
be able to cope with the sick building environment for extended
periods, although suffering from increased stress. Other individuals,
more seriously affected, may find symptoms so severe that they may be
unable to be in the building for extended periods, or at all. Still
others may become temporarily or permanently disabled.
It has been suggested that SBS may not be one syndrome but a number
of sub-syndromes [Ex. 4-170]. This hypothesis suggests that the
symptoms particularly associated with chemical exposure include
fatigue; headache; dry and irritated eyes, nose, and throat; and
sometimes include nausea and dizziness. Those symptoms most related to
microbial exposures would result in itchy, congested, or runny nose;
itchy watery eyes; and sometimes include wheezing, tight chest, or flu-
like symptoms. The overlapping symptoms in each case are eye, nose, and
throat irritation, perhaps making the two sub-syndromes, chemical and
microbial, difficult to distinguish. Jones concludes that there is a
need for a treatment protocol as well as a diagnostic protocol, which,
in addition to describing corrective actions available in response to
different diagnostic findings, would also provide guidelines for the
design and implementation of follow-up studies of buildings and
individuals in order to assess treatment effectiveness [Ex. 3-170].
Randolph and Moss [Ex. 4-258] have written about a number of
problems ascribed to indoor air pollution in the chemically sensitive
patient. These problems include irritability from natural gas fumes,
allergy to dust from forced air ventilation systems, intoxication and
even hallucination from paint fumes. Randolph describes chemical
sensitivity to dry cleaning chemicals, and rug shampoo, and implicates
moldy carpets in producing allergenic substances. He also describes
joint pain, malaise, and fatigue due to pesticide exposure; and skin
rashes from exposure to plasticizers. Randolph further describes
intolerance to highly scented products such as deodorant soaps, toilet
deodorants, and disinfectants, especially pine-scented ones. Other
patients have reported reacting to strong perfumes and other cosmetics.
So-called air fresheners often prove to be particularly troublesome. He
also describes that some patients are sensitive to the odors from hot
plastic-coated wires in electronic equipment.
There is little data on the perceptions of victims of SBS. Shapiro
[Ex. 4-282] has complied a summary of 16 case-histories of SBS in the
victims' own words. It is useful to review these for insight into the
problems from the victims' point of view.
One episode that Shapiro [Ex. 4-282] reported on was in a building
occupied by a government agency. As a result of problems related to
carpeting and other suspected causes, five workers were reported to
have left the agency, 11 were relocated to alternative workspace or
worked at home, and 100 reported to the agency's medical officer that
they had SBS related problems. The range of self-reported symptoms
included a variety of moderate and acute respiratory problems;
headache; sore throat; burning of the eyes, lungs, and skin; rashes;
fatigue; laryngitis; clumsiness; disorientation; loss of balance;
nausea; numbness in extremities and face; and difficulty with mental
tasks.
The patient's reported that the diagnoses of the occupational
health physicians they visited included upper and lower respiratory
irritation, intoxication-type syndrome, occupational asthma, and
chronic hypersensitivity pneumonitis.
The central nervous system effects reported by many do not lend
themselves to ready diagnosis [Ex. 4-282]. Some of the lesser affected
individuals either saw no physician at all or saw a family doctor or
allergist who was not familiar with occupational or environmental
health [Ex. 4-282].
The Air Force Procedural Guide [Ex. 4-199] on dealing with SBS
takes a practical view: ``* * * in most cases the sick building
syndrome does not have a clearly understood etiology and many of the
SBS studies and investigations were inconclusive. The significance of
exposure that [what chemical or physical agent concentrations cause
symptoms] can be pathogenic remains unanswered, but the realities of
worker complaints and discomfort are valid reasons to seriously address
this problem.''
In summary, SBS is not a well-defined disease with well-defined
causes. It appears to be a reaction, at least in part due to
stimulation of the common chemical sense, to a variety of chemical,
physical or biological stimuli. Its victims display all or some of a
pattern of irritation of the mucous membranes, and the worst affected
individuals have neurological symptoms as well.
B. Building-Related Illness
Building-related illness (BRI) describes specific medical
conditions of known etiology which can often be documented by physical
signs and laboratory findings. Such illnesses include sensory
irritation when caused by known agents, respiratory allergies,
nosocomial infections, humidifier fever, hypersensitivity pneumonitis,
Legionnaires' disease, and the symptoms and signs characteristic of
exposure to chemical or biologic substances such as carbon monoxide,
formaldehyde, pesticides, endotoxins, or mycotoxins [Exs. 3-61, 4-144].
Some of these conditions are caused by exposure to bioaerosols
containing whole or parts of viruses, fungi, bacteria, or protozoans.
These illnesses are often potentially severe and, in contrast to SBS
complaints, are often traceable to a specific contaminant source, such
as mold infestation and/or microbial growth in cooling towers, air
handling systems, and water-damaged furnishings. Symptoms may or may
not disappear when the employee leaves the building. Susceptibility is
influenced by host factors, such as age and immune system status.
Mitigation of building-related illnesses requires identification and
removal of the source, especially in cases involving hypersensitivity
responses.
1. Indoor Air Contaminants
Comments submitted to the docket in response to the RFI and
contained in the literature indicate that specific substances or
classes of substances have been implicated as contributing to poor
indoor air quality problems. These substances, either alone or in
synergy, have produced health effects that OSHA believes can be
considered material impairment [Ex. 4-124]. In most cases, people
likely to be at risk have specific susceptibility.
But such susceptibility is common and adverse effects can arise
suddenly following exposure. The relevant effects can be categorized
into six categories: irritation, pulmonary, cardiovascular, nervous
system, reproductive, and cancer.
Common chemical sense or irritation perception is mediated through
receptors found not only throughout the nasal, pharyngeal, and
laryngeal areas of the respiratory system but also on the surface of
the eyes, specifically the conjunctiva and cornea [Ex. 4-239]. It is
partially through the stimulation of these receptors that exposed
persons perceive irritation. Many comments to the docket, from
citizens, researchers, and indoor air consultants, raised the issue
about the irritating effects related to known indoor air contaminants.
The air contaminants of concern include formaldehyde [Exs. 3-14, 3-32,
3-38, 3-188, 3-440a, 3-446, 3-575, 4-125, 4-144, 4-214], volatile
organic compounds (VOCs) [Exs. 3-32, 3-446, 3-500, 4-145, 4-243, 4-
320], ozone [Exs. 3-14, 4-42, 4-134, 4-236, 4-237], carpet-associated
chemicals [Exs. 3-25, 3-444D, 3-576, 4-144, 4-214], vehicle exhausts
[Exs. 3-6, 3-63, 3-206, 3-238, 3-360, 3-437, 3-444D, 3-631, 3-659],
combustion gases [Ex. 3-32], particulates [Exs. 3-32, 3-446, 3-500],
man-made mineral fibers (fiberglass, glasswool and rockwool) [Ex. 4-
33], and pesticides [Ex. 3-446]. The irritation effects present as
sensory irritation of the skin and upper airways, irritation of eye,
nose and throat, dry mucous membranes, erythema, headache, and abnormal
taste [Ex. 3-14, 4-33]. The pulmonary effects include upper and lower
respiratory tract effects such as rapid breathing, fatigue, increased
infection rate, broncho-constriction, pulmonary edema, asthma,
allergies and flu-like symptoms. Acute exposure to low level of air
contaminants results in primarily reversible effects, while chronic
exposure may result in pulmonary fibrosis that can result in
irreversible damage [Exs. 3-14, 4-33].
These health effects were associated, as reported in many comments
to the docket, with specific contaminants, including asbestos [Exs. 3-
38, 3-440A, 3-500], combustion gases [Exs. 3-14, 3-34, 3-440A, 3-446,
3-500], formaldehyde [Exs. 3-32, 3-38, 3-188, 3-440A, 4-124], ozone
[Exs. 4-42, 4-237], VOCs [Ex. 3-32], vehicular exhaust [Ex. 3-63], and
particulates [Exs. 3-32, 3-38, 3-440A, 3-500].
Individuals with underlying pulmonary disease, such as asthma, are
more susceptible than others to acute exposure to these indoor air
contaminants and experience coughing and wheezing at low levels of
exposure. Synergism may occur between chemical contaminants, such as
ozone and VOCs, in aggravating asthma [Ex. 4-33]. These affected
individuals may also be at increased risk of pulmonary infections due
to the synergistic effect between chemical and microbial contaminants
[Ex. 4-33].
Cardiovascular effects have also been associated with poor indoor
air quality. These effects are presented as headache, fatigue,
dizziness, aggravation of existing cardiovascular disease, and damage
to the heart. These effects are associated with exposure to combustion
gases such as carbon monoxide [Exs. 3-38, 3-440A], VOCs [Ex. 3-500],
and particulates [Ex. 3-500].
Nervous system effects have also been produced due to exposure to
poor indoor air quality. These effects include headache, blurred
vision, fatigue, malaise with nausea, ringing in the ears, impaired
judgement, and polyneuritis. These effects are associated with exposure
to carbon dioxide [Ex. 3-14], carbon monoxide [Exs. 3-32, 3-38, 3-446,
3-500], formaldehyde [Exs. 3-32, 3-38, 3-446, 3-500], and VOCs [Exs. 3-
32, 3-446, 3-500].
Relevant reproductive effects include menstrual irregularities and
birth defects and are associated with exposure to formaldehyde [Exs. 3-
446, 3-500] and VOCs [Exs. 3-446, 3-500].
The occurrence of cancer has also been attributed to exposures
associated with poor indoor air quality. In particular, cancer of the
lung, including mesothelioma, esophagus, stomach, and colon have been
associated with exposure to asbestos [Exs. 3-6, 3-14, 3-38, 3-188, 3-
440A, 3-500], radon [Exs. 3-35, 3-38, 3-188, 3-440A, 3-500], vehicular
exhausts [Exs. 3-84, 3-206, 3-360H], combustion gases [Ex. 3-500], VOCs
[Exs. 3-446, 3-500, 4-294], and particulates [Ex. 3-500].
2. Microbial Contamination
Building-related illnesses can result in serious illness and death.
Indoor transmission of disease caused by obligate pathogens (microbes
that require a living host) is common in indoor environments,
especially those that are overcrowded and inadequately ventilated [Ex.
4-33]. Diseases in this category include influenza, rhinovirus or
colds, and measles. Indoor transmission of disease caused by
opportunistic microorganisms usually affects compromised individuals,
those with existing conditions that make them more susceptible to
infection, such as pulmonary disease or immunodeficiency. Legionnaires'
disease, pulmonary tract infections, and humidifier fever are diseases
that fall into this category. Diseases that affect the immune system
include allergic reactions, as seen in antibody-mediated responses
(asthma and rhinitis) and interstitial lung disease, as seen in cell-
mediated reactions (hypersensitivity pneumonitis) [Ex. 4-33]. All of
these diseases produce substantial amounts of illness each year [Exs.
4-33, 4-41, 4-214].
In the U.S., Legionnaires' disease is considered to be a fairly
common, serious form of pneumonia. The Legionella bacterium is one of
the top three bacterial agents in the U.S. which causes sporadic
community-acquired pneumonia. Because of the difficulty in clinically
distinguishing this disease from other forms of pneumonia, many cases
go unreported. Although approximately 1,000 cases are reported to the
Centers for Disease Control and Prevention annually, it has been
estimated that over 25,000 cases of the illness actually occur. This
disease burden is estimated to result in over 5,000 to 7,000 deaths per
year [Ex. 4-41]. Brooks et al. [Ex. 4-33] reported that as many as
116,000 cases occur each year. Of these cases, it is estimated that
between 35,000 and 40,000 die. The attack rate for L. pneumophila
ranges from 0.1 to 5%. The case fatality rate ranges from 15 to 20%
[Ex. 4-214].
Two serious allergic or hypersensitivity diseases are asthma and
hypersensitivity pneumonitis (extrinsic allergic alveolitis). An
estimated 3% of the U.S. population suffers from asthma (approximately
9,000,000 people) [Ex. 4-41]. These individuals may be more susceptible
to bioaerosol contamination or chemical contamination of the indoor
environment.
Hypersensitivity pneumonitis is triggered by recurrent exposure to
microbials, fumes, vapors, and dusts [Ex. 4-33]. The lung interstitium,
terminal bronchioles, and alveoli react in an inflammatory process that
can organize into granulomas and progress to fibrosis. The symptoms of
acute episodes of this disease are malaise, fever, chills, cough and
dyspnea. The symptoms of chronic episodes are serious respiratory
symptoms such as progressive dyspnea. Chronic disease can lead to
irreversible pulmonary structural and functional changes [Ex. 4-33].
Approximately 15% (20,250) of 135,000 hospital admissions per year
that last an average of more than eight days are due to allergic
disease [Ex. 4-41]. Burge and Hodgson estimate that these
hospitalizations cost five million work days per year. The prevalence
of symptoms consistent with hypersensitivity pneumonitis, an
interstitial lung disease caused by organic dusts or by aerosols has
been examined in subpopulations at well-defined, increased risk, such
as farmers (0.1-32%) or pigeon breeders (0.1-21%) [Exs. 4-41, 4-214].
The only unbiased source of complaint rates in unselected office
workers are control buildings used in the study of hypersensitivity
pneumonitis in the U.S. Arnow et al. [Ex. 4-15] reported complaints
consistent with hypersensitivity pneumonitis in 1.2 percent and Gamble
et al. [Ex. 4-116] in 4 percent of these populations. Since no clinical
data are available, it is not known how these complaints are related to
actual disease, and it is unknown whether these complaints are
associated with lost work time, doctor visits or hospital admissions
[Ex. 4-41].
Humidifier fever, a less serious variant of hypersensitivity
pneumonitis, also is caused by exposure to microorganisms contained in
an aerosol. Attack rates in building epidemics have been as high as
75%, whereas complaint rates are usually 2-3% in nonepidemic situations
[Ex. 4-41]. Because of the similarity of the individual symptoms to
other diseases (fever, headache, polyuria, weight loss and joint pain),
it is often difficult to separate actual disease from complaints
related to the common cold in nonepidemic situations [Exs. 4-33, 4-41].
While rare, a workplace epidemic of humidifier fever can virtually shut
down an entire building, and only removal of the contamination will end
the epidemic [Exs. 4-41, 4-144, 4-214].
Microbial contamination of building structures, furnishings, and
HVAC system components contribute to poor indoor air quality problems,
especially those related to building-related illnesses. OSHA believes
that consequent health effects constitute material impairment of health
[Exs. 3-61, 4-41]. These can be categorized as irritation, pulmonary,
cardiovascular, nervous system, reproductive, and cancer effects.
Irritation effects, either from the physical presence of
bioaerosols or from exposure to VOCs released by biologicals, have been
demonstrated in susceptible workers [Ex. 3-32]. In addition, water
leakage on furnishings or within building components can result in the
proliferation of microorganisms that can release acutely irritating
substances into the air. Typically, where microorganisms are allowed to
grow, a moldy smell develops. This moldy smell is often associated with
microbial contamination and is a result of VOCs released during
microbial growth on environmental substrates [Ex. 4-41].
Pulmonary effects which have been associated with exposure to
bioaerosols include rhinitis, asthma, allergies, hypersensitivity
diseases, humidifier fever, spread of infections including colds,
viruses, and tuberculosis, and the occurrence of Legionnaire's disease
[Exs. 3-17, 3-32, 3-38, 3-61B, 3-188, 3-440A, 3-446, 3-500, 4-41, 4-
144, 4-214].
Building-related asthma has also recently been documented in office
workers [Exs. 3-61, 4-43] and some case reports show it to be
associated specifically with humidifier use. Biocides used in
humidification systems are suspected causes of office-associated asthma
[Ex. 4-103].
Cardiovascular effects manifested as chest pain, and nervous system
effects manifested as headache, blurred vision, and impaired judgment,
have occurred in susceptible people following exposure to bioaerosols
[Exs. 3-32, 3-446]. It has been suggested that these effects may be
caused by VOCs released by the microbiologicals, or they may be a
complication of related pulmonary effects.
The development of cancer in susceptible people is possible
following exposure to certain types of toxigenic fungi and mycotoxins.
However, the probability of such exposures occurring in workplaces
covered by this standard is probably limited. Mycotoxins (toxins
produced as secondary metabolites by many fungi) are among the most
carcinogenic of known substances, and are also acutely toxic. The
American Conference of Governmental and Industrial Hygienists wrote
``[t]he toxigenic fungi are common contaminants of stored grain and
other food products and have caused well-described outbreaks of acute
systemic toxicosis as well as specific organ carcinogenesis when such
food is consumed * * * It appears clear that massive contamination with
a highly toxigenic fungus strain of a site in which aerial dispersion
of metabolic products occurred would be necessary to induce acute
symptoms. However, considering the carcinogenicity of many fungal
toxins, an examination of the risks of chronic inhalation exposure
appears justified'' [Ex. 3-61].
In summary, most of the health effects associated with SBS and BRI
occur in indoor environments were concentrations of pollutants are much
less than the OSHA Permissible Exposure Levels (PELs) (29 CFR
1910.1000) [Ex. 4-3]. It is important to point out that the PELs are
chemical-specific standards that are not only based on health effects
but also on technological feasibility, cost restraints and a
``healthy'' worker exposed for a 40-hour work week. In the industrial
workplace, hazards are minimized by the use of administrative and
engineering controls and the use of personal protective equipment. The
nonindustrial environment, however, does not have these controls.
Ventilation systems are designed only to remove occupant-generated
contaminants, such as carbon dioxide and odors. These types of systems
were not designed to dilute multiple point sources of contaminants that
are typically found in nonindustrial workplaces (see section III).
Unless adequate ventilation and source controls are utilized and
adequately maintained, many of the chemical contaminants can
concentrate to levels that induce symptoms. The possibility exists that
synergistic effects occur. These effects occur not only between
substances to enhance their toxicity but also by lowering the
resistance to lung infection in susceptible persons.
C. Environmental Tobacco Smoke
ETS is composed of exhaled mainstream and sidestream smoke. The
chemical composition and exposure sources of ETS are described in the
Exposure section of this preamble (see Section III). The
pharmacokinetics of ETS have been widely studied and are described in
the following section.
A wide spectrum of health effects have been associated with
exposure to ETS. These effects include mucous membrane irritation,
decrease in respiratory system performance, adverse effects on the
cardiovascular system, reproductive effects, and cancer. The following
section also presents more detailed information on these health
effects.
1. Pharmacokinetics
Whether a chemical elicits toxicity or not depends not only on its
inherent potency and site specificity but also on how the human system
can metabolize and excrete that particular chemical. To produce health
effects, the constituents of ETS must be absorbed and must be present
in appropriate concentration at the sites of action. After absorption,
some of these contaminants are metabolized to less toxic metabolites
while some carcinogens are activated by metabolism in the body.
Available biomarkers of ETS, such as nicotine, clearly show that
nonsmoker exposure is of sufficient magnitude to be absorbed and to
result in measurable levels of these biomarkers. There is sufficient
evidence in the literature to indicate that several components of
sidestream smoke are rapidly absorbed and widely distributed within the
body. However, the extent of absorption, distribution, retention and
metabolism of these contaminants in the body depends upon various
physiological and pharmacokinetic parameters that are influenced by
gender, race, age and smoking habits of the exposed individuals. These
parameters and others may result in differences in susceptibility among
exposed subpopulations. Nicotine is one of the most widely studied
constituents of tobacco smoke. There have been numerous studies on the
pharmacokinetics of nicotine in both animals and man.
(a) Absorption and distribution. Absorption and distribution of
tobacco smoke constituents are usually measured by using surrogate
markers. A correlation between nicotine absorption and exposure to
tobacco smoke has between demonstrated, thus making nicotine an
appropriate marker for tobacco smoke in pharmacokinetic studies. The
steady state volume of distribution for nicotine is large indicating
that it is widely distributed within the body [Ex. 4-185]. Nicotine has
been shown to bind with plasma proteins which may interfere with
elimination and thereby prolong retention in the body. The studies in
the docket clearly indicate that nicotine and other constituents of
tobacco smoke are readily absorbed and distributed throughout the body
thereby increasing the potential of producing adverse effects at more
then one target site.
(b) Metabolism. Nicotine is rapidly eliminated, primarily via
metabolism and urinary excretion. The investigation of metabolism in
vivo and in vitro, has resulted in the identification of more than 20
metabolic products in the plasma and urine of humans and animals. The
principle metabolic pathways of nicotine appear to involve oxidation of
the pyrrolidine ring to yield nicotine-1'-N-oxide and cotinine, the
latter being the major metabolite and the precursor of many of the
metabolic products of nicotine. Some of the metabolites detected in the
urine of rats after intravenous administration in a study by Kyerematen
et al. [Ex. 4-185] are listed in Table II-1. In humans, cotinine is the
major degradation product of nicotine metabolism and has a serum half-
life of about 17 hours compared to two hours for the parent compound,
nicotine [Exs. 4-27, 4-253]. Trans-3'-hydroxycotinine in the free form
constitutes the largest single metabolite in smokers' urine accounting
for 35-40% of the urinary nicotine metabolite [Exs. 4-48, 4-241].
Smokers and nonsmokers differ in their metabolism of nicotine and
cotinine [Exs. 4-133, 4-184, 4-279]. The half-life values for urinary
elimination of nicotine and cotinine were found to be significantly
shorter in smokers than nonsmokers [Ex. 4-186]. Plasma nicotine
clearance was faster in smokers than in nonsmokers in this study. More
rapid elimination of nicotine and cotinine has been attributed to the
inductive effects of chronic cigarette smoking on the hepatic
metabolism of many xenobiotic agents. However, Benowitz et al. [Ex. 4-
29] were unable to confirm published research suggesting that smokers
metabolize nicotine and cotinine more rapidly than nonsmokers.
Variations in nicotine metabolism occur among individuals.
Variations also occur due to differences in gender and race [Exs. 4-26,
4-186, 4-314]. It has also been suggested that the metabolism of
nicotine between smokers and nonsmokers may differ. Male smokers have
been shown to metabolize nicotine faster than do female smokers after
intravenous infusion of nicotine and active smoking. However, this
difference was not observed by Benowitz and Jacob [Ex. 4-23] during a
study of daily intake of nicotine in smokers versus nonsmokers. The
metabolism of nicotine has also been studied in animals. Male rats (4
strains) were shown to metabolize nicotine faster than did females [Ex.
4-185].
In summary, the potential effect of nicotine, and other ETS
constituents in the body, is governed by interactions between several
physiological and pharmacokinetics parameters. These interactions may
lead to longer retention of toxic constituents, thus prolonging the
effects on the target organs resulting in tissue injury.
2. Irritation
Exposure to ETS is capable of inducing eye and upper respiratory
tract irritation. Common chemical sense or irritation perception is
mediated through receptors in the fifth, ninth, and tenth cranial
nerves. These receptors are found throughout the nasal, pharyngeal, and
laryngeal areas of the respiratory system and also on the surface of
the eyes [Ex. 4-239]. It is partially through the stimulation of these
receptors that exposed persons perceive irritation.
Table II-1.--Urinary Excretion of Nicotine and Metabolites in Male and Female Rats After Intravenous
Administration of [\14\C]Nicotine (0.5 mg/kg)
----------------------------------------------------------------------------------------------------------------
Male Female
---------------------------------------------------------------
Recovery of Recovery of
Metabolite administered t1/2 administered t1/2
radioactivity (Hr) radioactivity (Hr)
(percentage) (percentage)
----------------------------------------------------------------------------------------------------------------
Nicotine........................................ 10.8 plus-
minuse> 1.5 2.5 plus-
minuse> 0.4 \1\24.0 plus-
minuse> 4.6 \2\5.6 plus-
minuse> 0.5
Cotinine........................................ 9.3 plus-
minuse> 0.8 6.0 plus-
minuse> 0.6 \1\5.7 plus-
minuse> 0.7 \2\6.8 plus-
minuse> 0.8
Nicotine-N-oxide................................ 10.8 plus-
minuse> 0.9 1.6 plus-
minuse> 1.4 7.8 plus-
minuse> 1.4 2.6 plus-
minuse> 0.3
Cotinine-N-oxide................................ 8.5 plus-
minuse> 1.6 7.5 plus-
minuse> 0.8 \1\3.7 plus-
minuse> 1.0 6.8 plus-
minuse> 0.6
3-Pyridylacetic acid............................ 1.8 plus-
minuse> 0.3 5.8 plus-
minuse> 0.3 1.2 plus-
minuse> 0.2 \3\ND
3-(3-Pyridyl)--oxobutyric acid 2.7 plus-
minuse> 0.6 5.3 plus-
minuse> 0.9 2.4 plus-
minuse> 0.7 6.0 plus-
minuse> 0.6
3-Hydroxycotinine............................... 5.7 plus-
minuse> 0.5 6.7 plus-
minuse> 0.8 5.6 plus-
minuse> 1.5 9.9 plus-
minuse>1.5
-(3-Pyridyl)--
methylaminobutyric acid........................ 4.2 plus-
minuse> 0.6 5.9 plus-
minuse> 0.8 \1\1.4 plus-
minuse> 0.4 ND
Nornicotine..................................... 8.1 plus-
minuse> 0.9 4.1 plus-
minuse> 0.6 8.1 plus-
minuse> 1.8 \1\8.3 plus-
minuse>1.3
Demethylcotinine................................ 0.8 plus-
minuse> 0.1 ND <0.3 ND
-(3-Pyridyl)--oxo-N-
Methylbutramide................................ 1.8 plus-
minuse> 0.3 3.5 plus-
minuse> 0.6 \1\0.6 plus-
minuse> 0.3 ND
Isomethylnicotinium ion......................... 2.1 plus-
minuse> 4.5 plus-
minuse> 0.7 <0.3 ND
Allohydroxydemethylcotinine..................... 2.8 plus-
minuse> 0.4 9.8 plus-
minuse> 1.4 1.9 plus-
minuse> 0.6 10.0 plus-
minuse>1.6
---------------------------------------------------------------
Total....................................... 69.4 plus-
minuse> 3.0 .............. 65.0 plus-
minuse> 3.6 ..............
----------------------------------------------------------------------------------------------------------------
\1\0.01 0.05.
\2\p 0.01.
\3\ND, not determined; concentration too low to estimate t1/2 accurately.
The ability of tobacco smoke to elicit irritation may be enhanced
by low relative humidity and varies according to concentration [Ex. 4-
239]. Irritating components of ETS are contained in both the vapor
phase and the particulate phase (see Tables III-6 and III-7). These
effects have been studied in both experimental (e.g., animals studies;
clinical and chamber studies on humans) and field (e.g., surveys and
epidemiological studies) studies. The NRC report [Ex. 4-239] summarized
these studies and concluded that even though the specific components of
ETS that cause irritation were not identified, the overall effects were
eye and throat irritation and immunological responses. Weber [Ex. 4-
317] reported the results of a field study that included 44 workrooms
where smoking was taking place. Eye irritation was reported by 52 out
of 167 workers. Nonsmokers reacted more than smokers to the ETS; 36 of
the 52 workers who reported eye irritation at work were nonsmokers [Ex.
4-317]. Asano et al. [Ex. 4-18] reported significant eye irritation, as
measured by blinking rates, in both healthy smoking and nonsmoking
adults following exposure to ETS. Nonsmokers reported more eye
irritation than smokers did. Effects such as eye irritation and nasal
stuffiness were reported to OSHA in comments to the docket [Exs. 3-38,
3-58, 3-59, 3-188, 3-438D, 3-440A].
3. Pulmonary Effects
Much of the literature relevant to the association between non-
cancerous health effects and ETS has focused on children. Because
children are undergoing development and maturation, they are not
physiologically equivalent to adults exposed to the same conditions.
Therefore, findings in studies conducted with respect to ETS and
children may not be directly applicable to adults. However, a number of
studies have investigated the relationship between ETS and pulmonary
health effects in adults.
Studies which are restricted to adults vary by numerous factors,
such as the population studied, the measures used to estimate exposure
to ETS, and the physiologic and health outcomes examined. The studies
also varied in the consideration of potential confounders. A number of
studies have found relationships between ETS exposure and pulmonary
health effects. These studies have: (1) used pulmonary function tests,
which may be more sensitive than methods used in other studies, to
detect physiological changes occurring in the small airways of the
lungs (e.g., forced mid-expiratory flow rate (FEF25-75), and
forced end-expiratory flow rate (FEF75-85)); (2) studied older
populations with a longer history of exposure to ETS; (3) stratified
the level of ETS exposure with significant findings more likely to
occur in persons with higher exposures; and (4) more frequently found
significant changes in lung function in men, although adverse pulmonary
effects to ETS have also been shown in women. The following discussion
summarizes the results of these studies [Exs. 4-18, 4-37, 4-62, 4-148,
4-173, 4-176, 4-178, 4-180, 4-209, 4-210, 4-278, 4-295, 4-321].
Asano et al. [Ex. 4-18] demonstrated the acute physiologic changes
which occur as a result of exposure to ETS. Nonsmokers had more
pronounced changes in eye blinking rates (a measure of eye irritation),
expired carbon monoxide, increased heart rate and systolic blood
pressure.
Studies of ETS and chronic health effects in adults differ by how
they define ``never smokers'', ``exsmokers'', and how other various
levels of ETS exposure are defined, either in nominal, ordinal or
interval scales; and whether or not they take into account exposure
both in the workplace and at home. The potential for misclassification
bias occurs when ``nonsmokers'' are loosely defined and used as the
comparative group to passive smokers. Several studies considered the
confounding impact of environmental air pollution [Ex. 4-278], indoor
cooking fuels [Exs. 4-37, 4-62] or occupational exposures to dusts and
fumes [Exs. 4-176, 4-178, 4-209, 4-210, 4-321].
There have been fewer longitudinal studies [Exs. 4-148, 4-278, 4-
295] as compared to the majority which have been cross-sectional
studies. The duration of exposure, which is critical to producing a
measurable health effect, was quantified by number of years directly in
several studies [Exs. 4-37, 4-148, 4-173, 4-295, 4-321], or indirectly
by the age of the population under study [Exs. 4-176, 4-209, 4-210]. In
those studies which had carefully assessed for level of exposure and
had specified a duration of at least 10 years, significant pulmonary
function decrements were noted in both men and women [Exs. 4-37, 4-148,
4 176, 4-321]. Overall, changes in pulmonary indices are more likely to
occur in men than in women, however, several studies have documented
statistically significant physiological changes in pulmonary function
occurring in women [Exs. 4-37, 4-176, 4-178, 4-321].
Understanding the significance of findings is complicated because
studies used a variety of measures from spirometry. Although most
studies evaluated FVC (forced vital capacity) and FEV1 (forced
expiratory volume in one second), fewer studies have measured
FEF25-75 or FEF75-85 [Exs. 4-176, 4-180, 4-209, 4-210, 4-
321]. These later measures have been suggested as being more sensitive
to detecting changes in the small airways where effects of ETS are most
likely to occur [Exs. 4-46, 4-216, 4-230, 4-231]. However, there is no
clear consensus in the medical literature as to the routine clinical
use of FEF25-75 or FEF75-85, or their diagnostic value in
independently detecting small airway disease [Ex. 4-8].
Estimates of the decrement in FEV1 due to ETS exposure in
passive smokers as compared to never smokers, ranges from 80
milliliters (ml) [Ex. 4-148] to 190 ml [Ex. 4-37]. When this decrement
is expressed as a percent of FEV1, it has been estimated to be
5.7% in males, or 7.3% when these same subjects were matched for age
[Ex. 4-210]. As a means of comparison, the average loss in lung volume
per year due to aging alone is estimated to be 25 to 30 ml [Ex. 4-329].
The American Thoracic Society [Ex. 4-8] specifies that spirometry
equipment have a level of accuracy within 50 ml. Since pulmonary
function maneuvers are very effort dependent, intra-individual
variation between the three best efforts should be within 5% to be
acceptable. The importance of these spirometry criteria is emphasized
by the fact that the FEV1 may result in being 100 to 200 ml lower
than when a maximal effort is given by the subject. Furthermore, a
decrease of 15% must be achieved before certain pulmonary indices are
considered outside of normal limits. Given this perspective, although
changes in pulmonary function tests may truly occur as a result of
exposure to ETS over a number of years, the actual clinical impact may
not be apparent in the healthy, young individual. Older individuals and
those with preexisting pulmonary disease are more susceptible to the
pulmonary effects of exposure to ETS.
Outside of respiratory changes being documented through pulmonary
function testing, other symptoms have been found to be significantly
associated with ETS exposure. Hole et al. [Ex. 4-148] found a
significant increase in the prevalence of infected sputum, persistent
sputum, dyspnea and hypersecretion in passive smokers as compared to
controls. Furthermore, rates increased as those exposed were stratified
by level of exposure to passive smoke from low to high. Kauffmann et
al. [Ex. 4-178] noted a significant increased risk for dyspnea in
American (Odds Ratio (OR)=1.42) and French women (OR=1.43), and an
increased risk for wheeze in American women (OR=1.36). Schwartz and
Zeger [Ex. 4-278] found an increased risk for phlegm or sputum in a 3-
year longitudinal study (OR=1.41). This risk was raised to 1.76 when
asthmatics, who may be medicated, were excluded from the analysis.
As small airway disease progresses to chronic obstructive pulmonary
disease (COPD) (also referred to as chronic obstructive lung disease
(COLD)), the impact of ETS becomes more detectable. Kalandidi et al.
[Ex. 4-173] reported an adjusted odds ratio of 2.5 (90% Confidence
Interval (CI), 1.3 to 5.0) for Greek women never smokers exposed to
their husbands' tobacco smoke.
While there is a clear trend, and in several studies a
statistically significant finding of a demonstrated decrease in
pulmonary function indices, or an increase in respiratory symptoms in
passive smokers, the impairment nonsmokers suffer by the exposure may
not be immediately obvious. It is important to note that these findings
have been demonstrated in otherwise healthy individuals. Based upon the
finding of White and Froeb [Ex. 4-321], Fielding and Phenow [Ex. 4-102]
have described such changes as being equivalent to those found in light
smokers, who smoke from 1 to 10 cigarettes per day. Where a decrease of
100 to 200 ml of FVC or FEV1 may be clinically insignificant in
healthy persons, such a change may be significant for workers with
already impaired pulmonary function [Exs. 3-438D, 3-440A, 4-76, 4-182].
These changes may be the pivotal point at which a worker becomes unable
to continue to work.
Cellular effects on the pulmonary tissue have also been observed in
animals exposed to ETS during experimental studies. Several studies
reviewed by OSHA have demonstrated that chronic cigarette smoke
exposure produces an accumulation of alveolar macrophages (AM) (the
presence of AM indicates a body's response to environmental insults),
within the respiratory bronchioles of many animals species. This effect
is similar to that seen in human smokers [Exs. 4-31, 4-58, 4-109, 4-
110, 4-140, 4-147, 4-150, 4-179, 4-212, 4-249]. Increased elastase
secretion by alveolar macrophages from mice chronically exposed to
cigarette smoke has also been observed [Ex. 4-322].
Accumulation of polymorphonuclear leucocytes (PMNs) is also an
indication of the body's response to environmental insults. PMNs were
found in the alveolar septum of cigarette smoke-exposed hamsters,
similar to the PMNs observed in the lungs of human smokers [Ex. 4-204].
In contrast to the focal nature of the alveolar macrophages
accumulation, the accumulation of PMN is diffuse. Studies of PMN
leukocyte function have not been systematically evaluated in smoke-
exposed animals.
Other studies also show effects of ETS exposure at the cellular
level. For example, young lambs exposed to ETS for one month did not
develop detectable pulmonary system effects or alteration in lung
mechanics or airway responsiveness. However, the lambs did develop
inflammation of pulmonary cells [Ex. 4-290]. A cytotoxic effect of
tobacco smoke was also demonstrated by decreased intracellular
adenosine triphosphate (ATP) content in guinea pig alveolar macrophages
and lowered cell bacteriocidal activity in a study by Firlik [Ex. 4-
104]).
Exposure to tobacco smoke has been shown to increase the
permeability of the respiratory epithelial membrane to macromolecules.
Burns et al. [Ex. 4-45] have shown that exposure of guinea pigs to
tobacco smoke followed by fluorescein isothiocyanate-dextran (FITC-D,
molecular weight 10,000) increased the amount of intact FITC-D that
crossed the respiratory epithelium into the vascular space.
Transmission electron-microscopic studies showed that the FITC-D
diffused across damaged type I pneumocyte membranes and cytoplasm to
reach the basal lamina and entered the alveolar capillaries through the
endothelial junction. Damage to alveolar epithelium was more frequent
for the smoke-exposed animals than the room air-exposed animals.
Aryl hydrocarbon hydroxylase (AHH) participates in the activation
of various carcinogens, such as benzo(a)pyrene. This is one of the many
carcinogens found in ETS. Both mainstream and sidestream smoke are
capable of inducing pulmonary AHH activity. Gairola [Ex. 114] has
demonstrated the induction of pulmonary AHH activity in Sprague-Dawley
rats and male C57BL mice after exposure to either mainstream or
sidestream smoke from University of Kentucky Reference cigarettes (2R1)
for seven days per week for 16 weeks. However, no such induction was
noted in Hartley guinea-pigs under similar conditions, indicating a
species difference. The mainstream and the sidestream smoke were
equally effective in inducing the AHH activity.
There is consistent evidence that decrements in pulmonary function
and increases in respiratory symptoms occur in current smokers and in
exsmokers. However, in passive smokers these health effects are not as
easily demonstrated. The Environmental Protection Agency's December
1992 report, Respiratory Health Effects of Passive Smoking: Lung Cancer
and Other Disorders [Ex. 4-311], reviewed an abundance of evidence
showing persistent physiologic changes in children's respiratory
function and related health effects as a result of exposure to ETS.
Studies evaluating these same effects are not as plentiful in adults.
However, the EPA concluded, ``recent evidence suggests that passive
smoking has subtle but statistically significant effects on the
respiratory health of adults'' [Ex. 4-311].
The weight of the evidence shows that exposure to ETS results in
decreases in pulmonary function indices and increases in respiratory
symptoms in otherwise healthy men and women who are exposed to ETS for
periods of 10 or more years. The risk of developing COPD appears to be
increased in passive smokers with lifelong exposures to ETS. Whether
these changes impact upon respiratory function to a degree that
impairment occurs may be dependent upon the individual's pulmonary
status and overall health condition.
4. Cardiovascular Effects
A developing body of research indicates that the cardiovascular
effects of ETS exposure on the health of nonsmokers include acute
effects, such as exacerbation of angina, as well as chronic effects,
such as atherosclerosis [Exs. 4-123, 4-291, 4-330].
Cardiovascular diseases [Exs. 4-91, 4-136] such as myocardial
infarction [Ex. 4-12], sudden death, and arterial thrombosis occur more
frequently in cigarette smokers as opposed to nonsmokers [Exs. 4-86, 4-
233]. The same chemicals which produce these effects in active smokers
are present in ETS. These include nicotine, carbon monoxide, polycyclic
aromatic hydrocarbons (PAHs) and tobacco glycoproteins.
The following discussion on cardiovascular effects covers thrombus
formation, vascular wall injury and the possible mechanisms of these
effects in nonsmokers. Discussion of the acute and chronic health
effects follows.
(a) Thrombus Formation. Blood clots in the coronary arteries are an
important component of an acute myocardial infarction (MI). An
additional component of the acute MI is the presence of atherosclerotic
plaques in the walls of the coronary arteries. Platelets are involved
in both the acute formation of blood clots and the chronic formation of
atherosclerotic plaques.
There is evidence that ETS exposure can cause platelets to become
more easily activated thus predisposing the platelets to become
involved in forming clots and atherosclerotic plaques. For example,
evidence exists that demonstrates that the platelets of nonsmokers
exposed to ETS are more easily activated [Exs. 4-40, 4-80]. The study
by Burghuber [Exs. 4-40] demonstrates that the platelet activating
capabilities of ETS are more prominent in nonsmokers than in smokers.
The results of this study suggest that nonsmokers are at a greater risk
of blood clot formation secondary to ETS exposure than smokers.
Acute ETS exposure also results in an increased platelet
aggregation, which is an initial stage of the development of coronary
thrombosis or vasoconstriction. This vasoconstriction can lead to the
development of coronary atherosclerosis after chronic exposure [Exs. 4-
111, 4-123, 4-272]. Environmental smoke exposure also can increase
platelet-activating factor (PAF), platelet factor 4, beta-
thromboglobulin, and fibrinogen concentration which provides a marker
of its effect on coronary heart disease [Exs. 4-85, 4-157, 4-224].
(b) Vascular Wall Injury. Atherosclerotic plaque formation is a
complicated chronic process that can lead to constriction of the lumen
of the blood vessels, resulting in reduced blood supply to the
myocardial tissues. It is thought that an essential step in plaque
formation is injury to the endothelial lining of the arterial wall. ETS
has been implicated in causing injury to the endothelial cells which
line the arterial walls. This was demonstrated in the study by Davis et
al. [Ex. 4-80] which identified an increase in the number of
endothelial cell carcasses in the circulation of healthy people after
being exposed to ETS.
ETS has also been implicated in stimulating smooth muscle cell
proliferation and in altering blood lipids. Each of these can
contribute to plaque formation which leads to an increased
susceptibility to heart attacks.
(c) Possible Mechanisms of Effect. At least three mechanisms are
described in the literature by which ETS may place stress on the heart
by increasing myocardial oxygen demand, decreasing myocardial oxygen
supply or interfering with the cell's ability to utilize oxygen for
energy production.
One mechanism by which ETS may reduce oxygen supply is through the
formation of carboxyhemoglobin. Carboxyhemoglobin is formed when a
person is exposed to carbon monoxide, a component of ETS. The carbon
monoxide effectively competes with oxygen for the heme group of the
hemoglobin molecule in the red blood cell (RBC). In fact, carbon
monoxide has a much greater affinity for hemoglobin than does oxygen
and binds very strongly with hemoglobin making it unavailable for the
transport of oxygen. The heart muscle (myocardium) can experience
injury at the cellular level when the oxygen demanded by the heart
muscle exceeds the oxygen supplied by the blood. Therefore, the
formation of carboxyhemoglobin can decrease the ability of the blood to
deliver oxygen to the myocardium and can cause injury to the heart if
myocardial oxygen demand exceeds supply.
A number of studies have suggested that ETS exposure adversely
affects the myocardial oxygen supply-demand relationship; this would
predispose the heart to develop ischemia or exacerbate preexisting
ischemia. Direct or indirect exposure to tobacco smoke has been shown
to increase the hemodynamic determinants of myocardial oxygen demand
[Exs. 4-13, 4-242] at the same time that it potentially reduces both
myocardial oxygen supply and delivery by enhancing the development of
coronary atherosclerosis [Exs. 4-242, 4-323], causing coronary
vasoconstriction [Exs. 4-323, 4-324] and reducing the oxygen carrying
capacity of blood through increased carboxyhemoglobin levels [Ex. 4-
13]. As a result, fewer red blood cells are available to transport
oxygen to the body, and to the heart muscle itself. To compensate for
this reduced oxygen carrying capacity of the blood, the heart must work
harder, for example, by increasing the heart rate. This is an example
of one mechanism by which ETS may place even further stress on the
heart by increasing myocardial oxygen demand, precisely at a time when
the oxygen delivery capabilities of the blood are reduced.
A second mechanism by which ETS may increase myocardial oxygen
demand is via the direct effect of nicotine. The nicotine in ETS may
cause an increased resting heart rate and blood pressure in exposed
individuals.
One study examined the effects of ETS on healthy individuals during
exercise, and found that healthy individuals experienced fatigue at
lower work levels when exercising in the presence of ETS [Ex. 4-123].
The authors concluded that ETS exposure interfered with the heart
muscle cells' ability to utilize oxygen for energy production.
Consequently, ETS exposure may have an adverse impact on myocardial
metabolism and expose the heart muscle to an increased susceptibility
to injury. These mechanisms of cardiac stress and potential injury to
the heart are in agreement with accepted theories of cardiac injury.
(d) Acute Heart Effects. An acute effect of exposure to ETS is the
aggravation of existing heart conditions, such as angina. The National
Research Council (1986) reported, based on the effects of studies by
Anderson et al. [Ex. 4-9] and Aronow et al. [Exs. 4-14, 4-16, 4-17],
that angina patients are especially sensitive at carboxyhemoglobin
levels between 2 and 4%. Guerin et al. [Ex. 4-129] report that
physiologically adverse effects occur in humans at 2.5%
carboxyhemoglobin blood content. Cumulative carbon monoxide levels, due
to ETS that result in such an effect are not uncommon in work
environments [Ex. 4-129]. Acute exposure to ETS has been reported to
increase heart rate, elevate blood pressure, and increase
carboxyhemoglobin levels in both angina patients [Exs. 3-38, 4-222] and
in healthy subjects [Exs. 4-18, 4-217]. Acute exposure has also been
associated with slight changes in blood components thought to be
involved in the pathogenesis of atherosclerosis, such as endothelial
cell count, platelet aggregate ratio, and platelet sensitivity to
prostacyclin [Exs. 4-40, 4-80]. Many effects of ETS exposure, such as
ischemia, may be additionally aggravated by simultaneous exposure to
other compounds, such as solvents [Exs. 3-446, 4-99].
(e) Chronic Heart Effects. The occurrence of coronary heart disease
in ETS-exposed nonsmokers has been studied by various epidemiological
researchers [Exs. 4-85, 4-120, 4-122, 4-138, 4-139, 4-142, 4-148, 4-
154, 4-191, 4-277, 4-295]. Small, but statistically significant (at p
0.05), increases in coronary heart disease mortality [Exs.
4-85, 4-138, 4-139, 4-142, 4-277] indicate a modest impact of long-term
ETS tobacco smoke exposure on the cardiovascular health of nonsmokers.
The relative risks calculated in these studies ranged from 1.3 to 2.7.
The ability of ETS exposure to induce coronary heart disease has
also been studied in animals. Zhu et al. [Ex. 4-330] exposed rats to
ETS and showed a dose-related increase in myocardial infarct size and a
decrease in bleeding time. But there were no significant differences in
serum triglycerides, high density lipoprotein and cholesterol. This
study showed that air nicotine, carbon monoxide, and total particulate
concentrations increased with ETS exposure, and this increased exposure
led to a continuous increase in plasma carboxyhemoglobin, nicotine, and
cotinine levels in ETS-exposed rats. There was a positive relationship
between the infarct size and air nicotine, carbon monoxide, total
particulate concentrations and plasma carboxyhemoglobin, nicotine, and
cotinine levels. The average concentrations of air nicotine, carbon
monoxide and particulates, according to the authors, were 30-fold, 3-
fold and 10- fold higher, respectively, than in a heavy smoking
environment. The duration of exposure, however, was short compared to
even a rat's lifetime. Infarct size nearly doubled following only 180
hours of ETS exposure distributed over a six week period.
In the same study, the effect of ETS exposure on platelet function
and aortic and pulmonary artery atherosclerosis in New Zealand male
rabbits was demonstrated. The increase of atherosclerosis after
exposure to ETS was shown to be independent of changes in serum lipids
and exhibited a dose-response relationship in this study. Average air
nicotine, carbon monoxide and total particulate concentrations were
1,040 g/m\3\, 60.2 ppm and 32.8 mg/m\3\ for high dose group
and 30 g/m\3\, 18.8 ppm and 4.0 mg/m\3\ for low dose group and
<1 g/m\3\, 3.1 ppm and 0.13 mg/m\3\ for the control group.
Atherosclerosis in this study was significantly increased in the high
dose group.
Olsen [Ex. 245] exposed rats daily to smoke from University of
Kentucky 2R1 Reference cigarettes for 10 minutes, 7 times a week for 4,
8 or 20 weeks. Sidestream (SS) smoke was collected by a moving column
of air spiked every minute with a puff of fresh mainstream (MS) smoke.
Rats were exposed to this SS smoke collected in a 2 L/min air flow
using a glass container placed over a burning cigarette. A fraction of
this air flow containing SS smoke was diluted with fresh room air and
continuously diverted to the rats as follows: 50%, 25% and 10% SS
smoke. Carboxyhemoglobin content for each treatment group was
determined immediately after the last smoke exposure and percent
carboxyhemoglobin for each group was found to be: 4 week exposure-
mainstream=7.2plus-minuss>1.2 and 25%
sidestream=11.8plus-minuss>0.7; 8 week exposure
mainstream=6.1plus-minuss>1.2 and 25%
sidestream=11.9plus-minuss>0.9; 20 week exposure
mainstream=8.3plus-minuss>0.9, 10%
sidestream=6.30plus-minuss>0.5, 25%
sidestream=10.8plus-minuss>0.8 and 50%
sidestream=18.3plus-minuss>1.2. This indicates a tobacco smoke-
related detrimental effect on blood components, thus increasing the
probability that coronary disease would develop over a longer exposure
period.
Research has shown that passive exposure to tobacco smoke damages
endothelial cells and increases the number of circulating anuclear
carcasses of endothelial cells [Ex. 4-80]. ETS appears to alter cardiac
cellular metabolism in such a way that renders the myocyte less capable
of producing adenosine triphosphate (ATP). Reduced oxidative
phosphorylation in cardiac mitochondrial fractions taken from rabbits
exposed to ETS has been demonstrated [Ex. 4-130]. Studies have
indicated that the reduction in mitochondrial respiration secondary to
ETS exposure is likely due to decreased cytochrome oxidase activity
[Exs. 4-130, 4-131].
Nicotine, a component of tobacco smoke, has been shown in in vitro
studies, to inhibit the release of prostacyclin, through inhibition of
cyclooxygenase, from the rings of rabbit or rat aorta. Nicotine could
also affect platelets by releasing catecholamines which lead to
increased thromboxane A2 [Ex. 4-25]. Passive smoke also increases blood
viscosity and hematocrit due to relative hypoxia induced by chronic
carbon monoxide exposure [Ex. 4-25]. Nicotine, contained in cigarette
smoke can lead to catecholamine release, which enhances platelet
adhesiveness and decreases the ventricular fibrillation threshold. This
threshold is also affected by carbon monoxide levels [Exs. 4-25, 4-
196]. Cigarette smoke also increases the lipolysis that increases
levels of plasma free fatty acids, which result in enhanced synthesis
of LDL [Ex. 4-234].
In conclusion, there are multiple pathways by which ETS may damage
the heart. ETS exposure has been demonstrated to both increase
myocardial oxygen demand and decrease myocardial oxygen supply. If
oxygen demand exceeds supply for a long enough period of time, then
myocardial cell injury or even cell death can occur. In addition, ETS
exposure may cause platelets to become less sensitive to the anti-
clotting regulatory substances in the blood and therefore increase the
tendency of the blood to clot. An increased tendency for the blood to
clot may lead to an increased susceptibility to heart attacks.
ETS exposure may also contribute to the chronic formation of
arterial wall plaques which are implicated in the event of an acute
myocardial infarction. The two mechanisms described by which ETS
exposure may stimulate plaque formation are endothelial cell injury and
increased platelet activation.
Different people will have different abilities to deal with the
increased stress on the heart and the increased tendency of the blood
to clot as a result of ETS exposure. For example, a young, otherwise
healthy individual may be able to tolerate short-term ETS exposure
without apparent difficulty, although asymptomatic arterial wall injury
may occur which can contribute to cardiac injury in the future.
However, an older person with pre-existing coronary artery disease and
therefore minimum cardiac reserve may not be able to tolerate short-
term ETS exposure, due to the increased stress on the heart.
5. Reproductive Effects
Data on the reproductive effects due to the exposure of nonsmoking
pregnant women to ETS has been presented in many studies [Exs. 3-438,
4-92, 4-132, 4-174, 4-208, 4-273, 4-285, 4-287, 4-299]. This is
important since many nonsmoking women continue to work throughout their
pregnancies. Pregnant women working in indoor environments without
tobacco smoking restrictions, as in restaurants, comprise one of the
most heavily ETS-exposed groups [Exs. 4-151, 4-287].
Low birthweight has also been shown to be associated with paternal
smoking, implying passive exposure to tobacco smoke by the nonsmoking
mother [Exs. 4-92, 4-273]. Passive exposure to tobacco smoke is
estimated to double the risk of low birthweight in a full-term baby
[Ex. 4-208]. Nonsmoking pregnant women who are exposed to ETS have been
reported to deliver neonates that range 24 to 120 grams lighter in
weight than those babies delivered by nonexposed pregnant women [Exs.
4-132, 4-174, 4-208, 4-273]. This relationship between passive smoking
and low birthweight remains statistically significant even after
accounting for mother's age, parity, social class, sex of baby, and
alcohol consumption. This effect is more apparent in neonates born to
actively smoking women who deliver babies that weigh, on average, 200
grams less than those of nonsmoking women [Ex. 4-101]. The reduction in
birthweight is clinically significant at the low end of the birthweight
distribution. These infants have higher perinatal mortality [Ex. 4-
239].
Other reproductive effects that have been ascribed to maternal ETS
exposure include miscarriage, an increase in congenital abnormalities
[Exs. 4-239, 4-299], and numerous other physiological effects [Ex. 4-
297]. It was reported that these effects may be part of a general
immunosuppressive condition associated with the occurrence of low
birthweight [Ex. 4-299]. This effect may predispose the baby to
respiratory tract infections.
The effects of environmental smoke exposure on the fetus may have
long-term sequelae into childhood and adulthood [Exs. 4-53, 4-181, 4-
213, 4-225, 4-239, 4-51, 4-297]. There is limited evidence which
suggests that growth retardation observed in the fetus is reflected in
the growing child as reductions in lung development [3-438]. This is
especially relevant if that child continues to be exposed to ETS
throughout childhood and into adulthood [Exs. 4-177, 4-297]. Prenatal
exposure to ETS and exposure to ETS as a child may also increase an
individual's cancer risk, perhaps by a factor of two (2) [Exs. 4-65, 4-
164, 4-252].
Experimental research on the adverse reproductive effects
associated with ETS exposure in animals is limited. However, one study
[Ex. 4-6] demonstrated such effects. Sciatic nerve tissue taken from
the offspring of ETS-exposed female mice revealed definite toxic
effects on the neonatal tissue [Ex. 4-6]. Pregnant female mice (C57BL/
KsJ) were exposed to low-tar cigarette smoke in a special smoking
chamber. Cigarette smoke was blown into the chamber for 4 minutes, 5
times daily, except on weekends when this was done 3 times daily. At 18
days of gestation, blood samples were taken and carbon monoxide levels
were measured. Ultrastructural abnormalities of fetal tissue revealed
swollen mitochondria with distorted cristae, some indication of
deformed mitochondria, darkened nuclei with condensations of nuclear
material, lamellar bodies, granules and myelin bodies similar to those
found in human toxicity studies. The blood samples from pregnant mice
revealed a mean carbon monoxide saturation in the hemoglobin of 9%
which is equivalent to that found in humans who actively smoke 10-20
cigarettes per day.
6. Cancer
Concern over the carcinogenic effects of ETS was expressed in many
comments submitted to the docket, such as Exs. 3-32, 3-35, 3-38, 3-207,
3-438, 3-440A, and 3-449. The results of epidemiological and
experimental studies indicate that exposure to ETS is causally
associated with cancer of the lung in chronically-exposed nonsmokers. A
discussion of this evidence follows.
(a) Evidence of Association.--The results of epidemiological
studies taken in the aggregate suggest that nonsmoker exposure to ETS
is causally-related to the development of lung cancer.
Evidence of specificity of effect is provided by active smoking
studies that report a causal association with lung cancer [Ex. 4-311].
It was therefore logical to examine nonsmokers with passive exposure to
tobacco smoke, since the chemicals found in passive smoke are
qualitatively similar to those in mainstream smoke. Active smoking
induces all four major histological types of human lung cancer--
squamous-cell carcinomas, small-cell carcinomas, large-cell carcinomas,
and adenocarcinomas [Ex. 4-311]. The results of lung cancer studies
that examined the variation in tumor cell type induced by ETS exposure
indicate that mostly adenocarcinomas and squamous cell carcinomas are
produced by ETS exposure. Some studies have reported an excess of
adenocarcinomas, while others have reported excesses in squamous cell
and small-cell carcinomas. From this information, it is apparent that
similar tumor cell types are induced by ETS exposure as are induced by
active smoking.
The unequivocal causal association between active tobacco smoking
and lung cancer in humans, as well as the corroborative evidence of the
carcinogenicity of tobacco smoke provided by animal bioassays and in
vitro studies and the chemical similarity between mainstream smoke and
ETS, clearly establish the plausibility that ETS is also a human lung
carcinogen (Table II-2). In addition, biomarker studies verify that ETS
exposure results in detectable uptake of tobacco constituents by
nonsmokers [Exs. 4-50, 4-311].
Table II-2.--43 Chemical Compounds Identified in Tobacco Smoke for
Which There is ``Sufficient Evidence'' of Carcinogenicity in Humans or
Animals [Ex. 4-160]
Acetaldehyde
Acylonitrile
Arsenic
Benz (a)anthracene
Benzene
Benzo (a)pyrene
Benzo(b)fluoranthene
Benzo (k)fluoranthene
Cadmium
Chromium VI
DDT
Dibenz(a,h)acridine
Dibenz(a,j)acridine
Dibenz(a,h)anthracene
Dibenzo (a,i)pyrene
Dibenzo (a,e)pyrene
Dibenzo (a,l)pyrene
Dibenzo (a,h)pyrene
Formaldehyde
Hydrazine
Lead
Nickel
N-nitrosodiethanolamine
N-nitrosodiethylamine
N'-nitrosodimethylamine
N'-nitrosonornicotine
N-nitrosopiperidine
N-nitrosodi-n-propylamine
N-nitrosopyrrolidine
N-nitrosodi-n-butylamine
ortho-toluidine
Styrene
Urethane
Vinyl chloride
1,1-dimethylhydrazine
2-nitropropane
2-napthylamine
4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone
4-aminobiphenyl
5-methylchrysene
7H-dibenzo(c,g)carbazole
Indeno (1,2,3,-cd)pryene
(b) Epidemiological and Experimental Studies. There are at least 32
epidemiological studies that have attempted to evaluate the
carcinogenic potential of ETS. OSHA analyzed these studies and
determined that 14 were positive for an association [Exs. 4-36, 4-65,
4-106, 4-119, 4-121, 4-142, 4-143, 4-153, 4-158, 4-187, 4-252, 4-275,
4-276, 4-292, 4-300], 5 were equivocal with a positive trend [Exs. 4-4,
4-47, 4-117, 4-122, 4-171], and 13 were equivocal [Exs. 4-35, 4-38, 4-
52, 4-118, 4-148, 4-164, 4-175, 4-183, 4-192, 4-283, 4-286, 4-296, 4-
326]. [See the Risk Assessment section for further discussion.]
OSHA considered the consistency of the association to determine if
the finding of the same exposure effect occurred in different
populations and different types of studies. The great number of
epidemiological studies available on ETS were conducted by different
researchers, on different populations, in various countries with
diverse study designs. This extensive amount of data increases
confidence that the associations seen between ETS exposure and the
development of lung cancer is externally consistent and is not due to
artifacts or a product of some unidentified, indirect factors unlikely
to be common to all of the studies. The fact that exposure to ETS is
common dilutes the risk estimates derived from these studies because
the comparison group has some exposure to ETS. A recent Centers for
Disease Control and Prevention (CDC) report [Ex. 4-50] found that 100%
of a subset of the National Health and Nutrition Evaluation Survey
(NHANES) III conducted by the National Center for Health Statistics had
detectable levels of cotinine in their bodies indicating that everyone
in the sample had detectable exposure to tobacco smoke [Ex. 4-50].
Cotinine is a metabolite of nicotine and is used as a surrogate of
exposure to tobacco smoke. This indicates that the cancer risk may
indeed be greater since the relationship in these studies has been more
exposed versus less exposed instead of exposed versus nonexposed.
Many potential sources of bias, such as publication bias (the
tendency of scientific journals to publish studies with positive
results), misclassification bias (smokers or former smokers claiming to
be nonsmokers), and recall bias (the reliance on self-reporting of both
personal smoking habits and exposure to others' tobacco smoke) can not
account for the elevation in risks seen in these various studies. Also,
the relative risks that were estimated from prospective study data are
similar to those estimated from case/control study data. Biases that
may be problematic to case-control studies are not a problem in
prospective studies. Since the results from both types of studies are
similar it is apparent that these biases are not important in the case-
control studies (e.g., misclassification bias and recall bias). This
information strengthens the confidence of a causal connection.
Animal studies have shown the carcinogenicity of cigarette smoke.
Limited existing data suggest that sidestream smoke may contain more
carcinogenic activity per milligram of cigarette smoke concentrate than
does mainstream smoke [Ex. 3-689D]. Currently, OSHA is aware of only a
few experimental inhalation studies with sidestream smoke or ETS
reported in the literature. A discussion of these studies follows.
Otto and Elmenhorst [Ex. 4-247] have shown that there are
carcinogenic constituents in the vapor phase of tobacco smoke. They
exposed C57B1 and BLH mice to the gas phase of cigarette mainstream
smoke of 12 cigarettes for 90 minutes daily over 27 months. The
particulate matter was removed by passing the smoke through a Cambridge
filter. The percentages of mice with lung adenomas were 5.5% and 32% in
the smoke-exposed C57B1 and BLH mice, as compared to 3.4% and 22% for
their respective controls. Leuchtenberger and Leuchtenberger [Ex. 4-
197] have also shown that the rate of tumors among mice exposed to the
gas phase was greater than animals exposed to the whole smoke.
Pulmonary adenomas and adenocarcinomas were induced in Snell's mice by
the gas phase but not by the whole smoke in this study. These studies
demonstrate that the carcinogenicity of tobacco smoke is not limited to
the particulate phase.
Studies have also reported hyperplasia and metaplasia in the
trachea and bronchi of mice exposed to cigarette smoke by the
inhalation route [Exs. 4-226, 4-327]. Four lung tumors and emphysema
were detected in 100 male and female C57B1 mice exposed, nose only, to
fresh mainstream smoke [Ex. 4-135].
Pulmonary squamous neoplasms were detected in female Wistar rats
exposed to a 1:5 smoke-to-air mixture for 15 seconds of every minute
during an 11 minute exposure twice a day, 5 days per week, for the
lifespan of the animals [Ex. 4-79]. Respiratory tumors were also
observed in Fischer-344 rats exposed, nose only, to a 1:10 smoke to air
mixture for approximately 30 seconds every minute, 7 hours per day, 5
days per week for 128 weeks [Ex. 4-77]. The incidence of laryngeal
leukoplakias in Syrian golden hamsters ranged from 11.3% for the
animals that received the low dose to 30.6% of those animals that
received the highest dose. These animals were exposed to a 1:7 smoke-
to-air mixture for 10 to 30 minutes, 5 days a week, nose only, for a
period of up to 52 weeks [Ex. 4-88]. Exposing hamsters twice a day, 5
days a week for up to 100 weeks resulted in almost 90% of the exposed
hamsters having hyperplastic or neoplastic changes in the larynx in a
study by Bernfeld et al. [Ex. 4-30]. Lung tumors have been reported in
beagle dogs exposed to the smoke from nonfilter cigarettes [Ex. 4-19].
However, no tumors were seen in rabbits exposed to cigarette smoke for
up to 5\1/2\ years [Ex. 4-149].
Sidestream condensates have also been shown to cause
carcinogenicity when implanted into female Osborne-Mendel rat lungs
[Ex. 4-127]. Cigarette smoke condensate fraction from sidestream smoke
was implanted at a dose level of one cigarette per animal in this
study.
Coggins et al. [Ex. 4-59] reported epithelial hyperplasia in the
nasal cavity of high-dosed rats exposed to environmental tobacco smoke.
They exposed Sprague-Dawley rats of both sexes, nose only, to ``aged
and diluted sidestream smoke'' (ADSS) at 0.1, 1 or 10 mg of
particulates per meter for 14 days and found ``slight to mild''
epithelial hyperplasia and inflammation in the most rostral part of the
nasal cavity in the 10 mg group only. They also found that these
changes were reversible if the animals were kept without further
exposure for an additional 14 days. No effects in the lung were
reported. Similar results of mild hyperplasia were also obtained when
male rats were exposed to the same concentrations for up to 13 weeks
[Ex. 4-60]. In this study the authors reported hypercellularity and the
thickening of the respiratory epithelium of the dorsal nasal conchae
and adjacent wall of the middle meatus.
Rats are obligatory nose-breathers, and the anatomy and physiology
of the respiratory tract and the biochemistry of the lung differ
between rodents and humans. Because of these distinctions, laboratory
animals and humans are likely to have different deposition and exposure
patterns for the various cigarette smoke components in the respiratory
system. For example, rodents have extensive and complex nasal
turbinates where significant particle deposition could occur,
decreasing exposure to the lung. These anatomical and physiological
differences, aside from the subchronic exposure, may partially account
for absence of any lung tumors in the study by Coggins et al.
The application of cigarette smoke condensate (CSC) to mouse skin
is a widely employed assay for the evaluation of carcinogenic
potential. CSC assays may not, however, reveal all of the carcinogenic
activity of actual cigarette smoke, because these condensates lack most
of the volatile and semi-volatile components of whole smoke. Benign
skin tumors and carcinomas were seen in Swiss-ICR mice exposed to
cigarette tar from the sidestream smoke of nonfilter cigarettes
suspended in acetone and applied to skin for 15 months [Ex. 4-327]. In
lifetime rat studies, intrapulmonary implants of mainstream smoke
condensate in a lipid vehicle caused a dose-dependent increase in the
incidence of lung carcinomas [Exs. 4-75, 4-289].
The polyamines contained in tobacco smoke, spermidine, spermine,
and their diamine precursor, putrescine, are believed to have an
essential role in cellular proliferation and differentiation. Formation
of putrescine from ornithine is catalyzed by ornithine decarboxylase
(ODC), the rate-limiting enzyme in polyamine biosynthesis. A
significant increase in lung and trachea ornithine decarboxylase
activity was observed by Olsen [Ex. 4-245] after an eight week exposure
of male Sprague-Dawley rats to MS smoke. All dilutions of SS smoke
exposure caused significant increase in trachea ODC activity but did
not influence the lung ODC activity.
Environmental tobacco smoke induced carcinogenicity is also
supported by a case-control study of lung cancer in pet dogs [Ex. 4-
259]. The study compared the incidence of lung cancer in pet dogs
exposed to their owners' smoking versus dogs whose owners did not
smoke. Dogs have a very low natural incidence of lung cancer. There was
an elevated risk of lung cancer (Relative Risk = 1.6) observed in pets
with smoking owners. However, the analysis was statistically
insignificant, perhaps in part due to small sample size.
7. Genotoxicity
Short-term mutagenicity tests have gained widespread acceptance as
an initial step in the identification of potential carcinogens.
Extensive use of these tests has come about because they are easy to
perform and are inexpensive and also because of the reported high
positive correlations between short-term mutagenicity tests and
carcinogenicity. It has been reported that 90 percent of the
carcinogens tested are mutagens and 90 percent of the noncarcinogens
are nonmutagens.
Several short-term bioassays have been performed to evaluate the
genotoxicity of cigarette smoke. While most of them have evaluated the
effect of cigarette smoke condensate, some have attempted to evaluate
either the gas phase or the whole smoke.
The most commonly employed assay for mutagenic activity employs
various strains of Salmonella typhimurium. Whole smoke as well as
cigarette smoke condensate of tobacco have been shown to be mutagenic
in Salmonella typhimurium strain TA 1538 [Ex. 4-21]. Sidestream smoke
was also found to be mutagenic in a system where the smoke was tested
directly on the bacterial plates [Ex. 4-246]. Sidestream smoke and
extracts of ETS collected from indoor air [Exs. 4-202, 4-5, 4-198, 4-
201, 4-203] also exhibited mutagenic activity in this bacterial strain.
Claxton et al. [Ex. 4-55] found that sidestream smoke accounted for
approximately 60% of the total S. typhimurium mutagenicity per
cigarette, 40% from the sidestream smoke particulates and 20% from the
semi-volatiles. The highly volatile fraction, from either mainstream or
sidestream smoke was not mutagenic.
Condensates from both mainstream [Exs. 4-89, 4-193] and sidestream
smoke [Ex. 4-90] have also been reported to have mutagenic activity.
Doolittle et al. [Ex. 4-89] demonstrated the genotoxicity of the
sidestream smoke from the Kentucky Reference cigarette (1R4F) by
employing several different assays. In their study, sidestream smoke
produced positive results in Salmonella typhimurium strains TA98,
TA100, TA1537, and TA1538 in the presence of S9 mix from aroclor-
induced rat liver but produced negative results in strain TA1535. They
also showed that sidestream smoke produced positive results in the
Chinese hamster ovary cells chromosomal aberration assay and in the
Chinese hamster ovary cell sister-chromatid exchange assay both with
and without metabolic activation. They demonstrated that the sidestream
smoke was weakly positive in inducing DNA repair in cultured rat
hepatocytes. However, sidestream smoke was nonmutagenic in the Chinese
hamster ovary cell-HGPRT assay both with and without metabolic
activation but it was found to be cytotoxic in this system.
In their further studies, Doolittle et al. [Ex. 4-90] observed
similar responses when they measured the genotoxic activity of
mainstream cigarette smoke condensate (CSC) from Kentucky reference
research cigarette (1R4F). As seen with sidestream smoke, CSC in this
study was mutagenic in Salmonella typhimurium strain TA98, TA100,
TA1537, and TA1538 in the presence of S9 mix but was negative in strain
TA1535. CSC was also positive in the Chinese hamster ovary (CHO) cells-
chromosomal aberration assay and in the CHO-sister-chromatid exchange
assay both with and without metabolic activation. CSC was weakly
positive in inducing DNA repair in cultured rat hepatocytes. However,
again as seen with sidestream smoke, CSC was nonmutagenic in the CHO-
HGPRT assay, with or without metabolic activation but was found to be
cytotoxic in this system. The results from these two studies appear to
indicate that sidestream smoke behaves very much like mainstream smoke
in these assays.
Mohtashamipur et al. [Ex. 4-227] demonstrated significant mutagenic
activity in the urine of rats exposed to sidestream smoke. In this
study, cigarettes were machine smoked under standardized laboratory
conditions and the sidestream smoke of two cigarettes was directed
through metabolism cages containing rats. The urine of these rats was
collected 24 hours prior to the SS exposure and 24 hours after the
onset of the exposure. The individual urine samples of all (10) rats
after exposure showed significantly higher activity for direct-acting
mutagens (in strain TA1538) than the urine samples of the same rats
before the exposure.
The formation of DNA adducts is widely accepted as an initial step
in the carcinogenesis process. The measurement of DNA adducts by the
\32\P-postlabeling assay has been used as a way to assess DNA damage
following exposure to cigarette smoke. Lee et al. [Ex. 4-194] exposed
Sprague-Dawley rats to 0.1, 1.0 and 10 mg total particulate matter/m\3\
of aged and diluted sidestream smoke (ADSS) for 6 hours per day for 14
consecutive days. They examined the DNA from lung, heart, larynx and
liver after 7 and 14 days of exposure and after 14 days of recovery.
They also examined alveolar macrophages for chromosomal aberrations.
Exposure related DNA adducts were found in the highest dose test.
However, no elevation in chromosomal aberrations was observed in
alveolar macrophages in this study. Similar results were also obtained
when animals were exposed to the same three concentrations for up to 90
days. DNA adducts were seen in lung, heart and larynx DNA of the
animals exposed to the highest concentration of ADSS [Ex. 4-195]. The
adduct levels were highest after 90 days of exposure and were
significantly reduced in all target tissues 90 days after cessation of
exposure. Again, chromosomal aberrations in alveolar macrophages were
not elevated in any group after 90 days of exposure. The authors
concluded that the concentration of DNA adducts formed in the lung
tissue did not increase linearly as the ADSS concentration was
increased from 1 to 10 mg.
Several short-term tests have been performed in eukaryotic systems.
A solution of the gas phase of mainstream cigarette smoke has been
shown to induce reciprocal mitotic recombination in Saccharomyces
cerevisiae D3 and petite mutants in an isolate of strain D3 [Ex. 4-
163]. Whole mainstream cigarette smoke induced mitotic gene conversion,
reverse mutation, and reciprocal mitotic recombination in strain D7 of
Saccharomyces cerevisiae [Ex. 4-113]. Transformation of mammalian cells
was induced in several cell systems using the cigarette smoke
condensate from mainstream cigarette smoke [Exs. 4-22, 4-161, 4-188, 4-
267, 4-268, 4-298].
Another in vitro assay that measures the number of sister-chromatid
exchanges (SCEs) induced has been employed widely to determine the
mutagenic activity of cigarette smoke. Valadand-Berrieu and Izard [Ex.
4-313] used a solution of the gas phase from cigarette mainstream smoke
and showed that this solution induced a significant dose-related
increase in sister-chromatid exchanges. Putman et al. [Ex. 4-257] have
also demonstrated dose-dependent increases in sister chromatid exchange
frequencies in bone-marrow cells of mice exposed to cigarette smoke for
2 weeks.
Review of the literature clearly demonstrates that MS smoke and ETS
exposure causes cancer in humans. These results are supported not only
by animal studies but also by studies that show SS smoke to be both
genotoxic and clastogenic.
8. Conclusions
The epidemiological and clinical studies, taken in aggregate,
indicate that exposure to environmental tobacco smoke may produce
mucous membrane irritation, pulmonary, cardiovascular, reproductive,
and carcinogenic effects in nonsmokers. Exposure to ETS may aggravate
existing pulmonary or cardiovascular disease in nonsmokers. In
addition, animal studies show that both mainstream and sidestream
tobacco smoke produce similar adverse effects.
D. Case Reports
1. Sick Building Syndrome and Building-Related Illness
Many case reports of material impairment of health due to
occupational exposure to poor IAQ have been reported to OSHA through
submission to the indoor air quality docket [H-122]. These adverse
health effects range from irritation effects to more severe, life-
threatening building-related illnesses, such as Legionnaire's disease,
and cancer.
Ford Motor Company responded in docket comment 3-447, that
``[p]resently, at Ford, we investigate an average of two IAQ complaints
per month which are predominantly classified as Sick Building Syndrome.
We have seen Building-Related Illness, but these incidents have been
rare and associated with specific contaminant episodes. The IAQ
complaints we generally investigate are characterized by general
malaise, headache, and flu-like symptoms that are said to disappear
when the occupants leave the building * * * Of the IAQ problems
investigated, about 20 percent can be attributed to PTS [passive
tobacco smoke]/ETS. Upper respiratory irritation or eye irritation
typically are associated with these complaints.'' Similar types of
health effects were reported to the agency in docket comments 3-1, 3-
22, 3-58, 3-142C, 3-367, 3-413, 3-529, 3-632, 3-634, 3-642, 3-659, and
3-698.
One comment [Ex. 3-433 reported that ``based upon approximately 30
IAQ investigations in a member company over the past two and one-half
years, the following adverse health effects have been reported in
office environments: eye, nose, and throat irritations; headaches,
nausea, dizziness, fatigue; cough, shortness of breath, chest
tightness. These so-called ``sick building syndrome (SBS)'' symptoms
often disappear when the person leaves the building environment. These
symptoms are usually subjective and non-specific, lacking a physician's
diagnosis of a definite illness.'' Others have reported [Ex. 3-377]
that ``as air flow and ventilation are cut back, our workers are
becoming sick. Many are exposed to contaminants or other harmful
substances; and, without ventilation, these sources linger and cause
nausea, skin irritations and other unhealthy symptoms of illness. In
severe cases, these contaminants and bacteria have been known to
contribute to upper respiratory infections.'' Comment 3-570 reported
similar health effects due to poor indoor air quality.
More serious health conditions have been reported ranging from
severe asthma to central nervous systems disorders. For example,
Comment 3-158 responded that ``I have developed a serious asthma
condition due to indoor air quality problems. Besides, three of the
remaining five employees at the branch office have been diagnosed with
chronic fatigue syndrome. In conversations with various health care
professionals, I have come to the conclusion that the diagnoses of
chronic fatigue syndrome were actually sick building syndrome. Of the
six employees at the branch office, four of the six are moderate to
heavy smokers. This does not take into consideration the other factors
that could be causing poor indoor air quality problems in the office.''
Comment 3-631 was a collection of reports from the workers in one
building that illustrate the poor conditions of a building that can
lead to serious health effects in workers. Health problems experienced
by workers in this building included chronic sinus infections;
headaches; fatigue; eye, nose and throat irritations; difficulty
breathing and congestion; allergies; and asthma. These health problems
seem to clear up when the workers were out of the building over a
weekend or a vacation.
The physical condition of this building was obviously in disrepair
since the commenters reported pails of stagnant water, collected from
leaks in the roof, were left in hallways. Water in ``[t]hese pails
ha[d] overflowed and run down the stairs. What [wa]s left in the pails
evaporate[d] leaving a gross residue of who knows what.'' The water
leaks from the roof caused mold infestation and water damage. Water
logged insulation hung in the ceiling out in a hallway. There was an
obvious lack of routine, sufficient cleaning. Dust and particulate
matter were visible in the air. The bathrooms were dirty. Smells of
sewer gas, mold, and diesel and other vehicular fumes permeated the
office space. Ventilation problems were evident since paint or varnish
fumes lingered whenever part of the inside physical structure of the
building was painted. Tar fumes were evident from constant patching of
the leaky roof. Insect infestation of the building was evident.
Pesticide fumes lingered whenever the building was spray[ed] for
roaches and steam bugs. Workers sighted cockroaches, silverfish, and
steam bugs near the coffee shop and on back stairs. The comment
continued that ``a sink faucet in the lunch room has been leaking for
years and water runs on the counter under the toaster and microwave.
The water heater had leaked for about 2 months before it was fixed. At
that time the carpet was soaked and water was running under the wall
into a supervisor's office. There is a moldy odor from this carpet and
the floor below.''
Cancer has also been reported to be associated with poor indoor air
quality. A courthouse in San Diego, California [Ex. 3-55], ``is
notorious for poor air quality and employee respiratory illness and
cancer.'' It was reported to OSHA that many long-term employees have
cancer (stomach and lung cancer), terminal lung disease, chronic ear
and throat infections, and bronchial problems'' [Exs. 3-585, 3-635, 3-
637, 3-68].
Comment 3-630 from a union reported that ``[a]fter surveying
thousands of workers across the country, SEIU compiled actual survey
responses that list adverse health effects caused by indoor air
pollution. These include headaches, nose congestion or irritation,
throat irritation, dry cough, dry or itchy skin, dizziness, nausea,
lethargy or fatigue, colds, asthma/wheezing, chest tightness, runny
nose/post nasal drip, eye or contact lens irritation, respiratory
difficulties. In addition, EPA estimates that pollutants found in
indoor air are responsible for 2,500 to 6,500 cancer deaths each year''
[refer to Ex. 3-630L].
These concerns are not just relevant to office workers but also to
maintenance and other nonindustrial workers that work in indoor
environments. For example, comment 3-347 responded that ``[i]n our
closed, indoor work environments, air quality is a very real health and
safety concern to professional painters. I have seen firsthand
otherwise healthy men and women pass out or get violently ill as a
result of being exposed to indoor air contaminants.'' Comment 3-412
responded ``[o]ur locals have encountered air-pollution problems
ranging from ink mist and photocopier emissions to asbestos and
microbial disease. The level of toxic chemical contaminants is often
alarmingly high in our darkrooms, and carbon-monoxide emissions from
trucks at newspaper loading docks frequently penetrate the ventilation
system. In 1985 microbial contamination from a water tower infected six
New York Times employees with Legionnaires' Disease and 34 others with
less serious respiratory infections.''
Operation engineers are also affected by poor indoor air quality.
Comment 3-452 responded that ``[t]his is particularly important for the
operation engineers who appear healthy and then suffer from respiratory
problems, much like allergic reactions, after working in a building
with poor ventilation.''
2. Environmental Tobacco Smoke
Many case reports of severe material impairment of health due to
occupational exposure to ETS have been reported to OSHA through
submission to the indoor air quality docket [H-122]. Information
contained in these comments indicate that adverse health effects in
workers due to environmental tobacco smoke exposure while at work range
from mucous membrane irritation (eye, nose, and throat effects) to more
severe, life-threatening conditions, such as status asthma, other
chronic lung diseases and heart diseases. For example, comment 3-309
responded [Regarding ETS exposure in a cafeteria], ``By the time I have
finished lunch my eyes are tearing, my nose is plugged, and I have a
headache'' as well as comment 3-315, ``I had fewer headaches and fewer
respiratory ailments; my chronic sore throat disappeared [after a
company-wide no smoking policy was implemented]''. Comment 3-22
responded ``[m]y patients find it hard to obtain smoke free workplaces.
I have seen patients who have suffered status asthma from workplace
smoking, patients who have had to quit their jobs because of ETS in the
workplace. Recently, one of my never smoking patients sustained vocal
cord lesions seen almost entirely in smokers.'' Comment 3-104 continued
that ``[p]assive tobacco smoke (PTS) is the principal indoor air
contaminant in my office building in Rockefeller Center. While smoking
is limited to `private offices', the smoke flows freely from these
private offices throughout the entire general office areas since the
smokers will not keep their doors closed, and even when they do, they
have to come out sometime. And, as soon as the door is opened, the
dense smoke accumulation within the office is diffused to all adjacent
work areas. Because office buildings have closed ventilation systems,
only a `smoke free' office policy can be effective. Half measures only
cause further stress, frustration and irritation to both smokers and
nonsmokers.'' Comment 3-289 responded that ``I have been exposed to
asbestos culminating in my getting asbestosis (plural plaque) of the
lungs. The combination of asbestos exposure plus second-hand smoke from
my smoking co-workers has posed and is currently posing a health risk
to me.''
III. Exposure
Contaminants which contribute to poor indoor air quality can be
attributed to both outside air and inside air. Outside air contaminants
can be introduced into a building through the ventilation intakes,
doors, building envelope, and windows. Outside air contaminants include
vehicular exhausts, industrial emissions, microbiologicals, and pollen.
Inside air contaminants are emitted from building materials and
furnishings, appliances, office equipment and supplies, biological
organisms, and of course, pollutants introduced by the building
occupants themselves. Inside air contaminants include tobacco smoke,
volatile organic compounds, combustion gases such as carbon monoxide,
and occupant-generated bioeffluents. The concentration of these
contaminants in buildings can increase if ventilation systems are
inadequately designed, maintained and operated or if strong local
contaminant sources are not controlled.
A. Sources of Indoor Air Contaminants
A wide variety of substances are emitted by building construction
materials and interior furnishings, appliances, office equipment, and
supplies, human activities, and biological agents. For example,
formaldehyde is emitted from various wood products, including particle
board, plywood, pressed-wood, paneling, some carpeting and backing,
some furniture and dyed materials, urea-formaldehyde insulating foam,
some cleaners and deodorizers, and from press textiles. Volatile
organic compounds, including alkanes, aromatic hydrocarbons, esters,
alcohols, aldehydes, and ketones are emitted from solvents and cleaning
compounds, paints, glues, caulks, and resins, spray propellants, fabric
softeners and deodorizers, unvented combustion sources, dry-cleaning
fluids, arts and crafts, some fabrics and furnishings, stored gasoline,
cooking, building and roofing materials, waxes and polishing compounds,
pens and markers, binders and plasticizers. Pesticides also contain a
variety of toxic organic compounds.
Building materials are point sources of emissions that include a
variety of VOCs (Table III-1). Some of these materials have been linked
to indoor air quality problems. The probability of a source emitting
contaminants is related to the age of the material. The newer the
material, the higher the potential for emitting contaminants. These
materials include adhesives, carpeting, caulks, glazing compounds, and
paints [Ex. 4-33]. These materials, as well as furnishings can act as a
sponge or sink in which VOCs are absorbed and then re-emitted later.
Appliances, office equipment, and supplies can emit VOCs and also
particulates [Ex. 4-33]. Table III-2 lists the many contaminants that
can be emitted from these point sources. There is an indirect
relationship between the age of the point source and the potential rate
of contaminant emission [Ex. 4-33].
Table III-1.--Emissions From Building Materials or Interior Furnishings
------------------------------------------------------------------------
Material Typical pollutants emitted
------------------------------------------------------------------------
Adhesives.......................... Alcohols.
Amines.
Benzene.
Decane.
Dimethylbenzene.
Formaldehyde.
Terpenes.
Toluene.
Xylenes.
Caulking Compounds................. Alcohols.
Alkanes.
Amines.
Benzene.
Diethylbenzene.
Formaldehyde.
Methylethylketone.
Xylenes.
Carpeting.......................... Alcohols.
Formaldehyde.
4-Methylethyl- benzene.
4-Phenylcyclohexene.
Styrene.
Ceiling Tiles...................... Formaldehyde.
Clipboard/Particle Board........... Alcohols.
Alkanes.
Amines.
Benzene.
3-Carene.
Formaldehyde.
Terpenes.
Toluene.
Floor and Wall Coverings........... Acetates.
Alcohols.
Alkanes.
Amines.
Benzenes.
Formaldehyde.
Methyl styrene.
Xylenes.
Paints, Stains & Varnishes......... Acetates.
Acrylates.
Alcohols.
Alkanes.
Amines.
Benzenes.
Formaldehyde.
Limonene.
Polyurethane.
Toluene.
------------------------------------------------------------------------
Table III-2.--Emissions From Appliances, Office Equipment and
Supplies\1\
Appliances......................... Carbon Monoxide.
Nitrogen Dioxide.
Sulfur Dioxide.
Polyaromatic hydrocarbons.
Carbonless Copy Paper.............. Chlorobiphenyl.
Cyclohexane.
Dibutylphthalate.
Formaldehyde.
Computers/Video Display Terminals.. n-Butanol.
2-Butanole.
2-Butoxyethanol.
Butyl-2-Methylpropyl phthalate.
Computer/Video Display Terminals... Caprolactam.
Cresol.
Diisooctyl phthalate.
Dodecamethyl cyclosiloxane.
2-Ethoxyethyl acetate.
Ethylbenzene.
Hexanedioic acid.
3-Methylene-2-pentanone.
Ozone.
Phenol.
Phosphoric Acid.
Toluene.
Xylene.
Duplicating Machines............... Ethanol.
Methanol.
1,1,1-Trichloroethane.
Trichloroethylene.
Electrophotographic Printers, Ammonia.
Photocopiers & Related Supplies. Benzaldehyde.
Benzene.
Butyl methacrylate.
Carbon black.
Cyclotrisiloxane.
Ethylbenzene.
Isopropanol.
Methylmethacrylate.
Nonanal.
Ozone.
Styrene.
Terpene.
Toluene.
1,1,1-Trichloroethane.
Trichloroethylene.
Xylenes.
Zinc stearate combustion Products.
Microfiche Developers/Blueprint Ammonia.
Machines.
Preprinted Paper Forms............. Acetaldehyde.
Acetic Acid.
Acetone.
Acrolein.
Benzaldehyde.
Butanal.
1,5-Dimethylcyclopentene.
2-Ethyl furan.
Heptane.
Hexamethyl cyclosiloxane.
Hexanal.
4-Hydroxy-4-methyl pentanone.
Isopropanol.
Paper dust.
Propionaldehyde.
1,1,1-Trichloroethane.
Typewriter Corrections Fluid....... Acetone.
1,1,1-Trichloroethane.
\1\Source: [Ex. 4-33]
Emissions from equipment, such as computers, will decrease over
time compared to emissions from equipment that continually use
chemicals. Emissions from such equipment (e.g., laser printers) that
use chemicals continually, will obtain a steady state concentration
dependent upon the chemicals used and frequency of equipment use.
B. Microbial Contamination
Three conditions must exist in buildings before microbial
contamination can occur: high humidity (over 60%), appropriate
temperatures (varies according to microbe), and appropriate growth
media [Exs. 3-61, 4-33]. These conditions are found in heating,
ventilating, and air conditioning (HVAC) systems. HVAC systems provide
multiple sites for microbes to grow (reservoir) and also the means to
disperse the microbes throughout the ventilated space. These reservoirs
of microbial growth, if allowed to proliferate unchecked, can lead to
indoor air quality problems once the microbes or microbe-related
products, such as endotoxins, are dispersed.
Building materials that have been soaked with water, such as
fiberglass insulation in air handlers, furnishings and fabrics, ceiling
tiles, and carpeting are excellent media for microbial growth.
Biological organisms, including fungal spores, bacteria, viruses,
pollens, and protozoa derived from mold growth have been identified in
humidifiers with stagnant water, water damaged surfaces and materials,
condensing coils and drip-pans in HVAC systems, drainage pans in
refrigerators, dirty heating coils, and are also associated with
mammals, arthopods and insects. Table III-3 gives examples of
biologicals found in indoor environments.
Various allergens have been associated with the development of
allergic rhinitis, asthma, or airway hyperresponsiveness (Table III-3)
[Ex. 4-33]. Many of these allergens are common to the nonindustrial
work environment. These include chemical volatiles and dusts,
arthropods, and dusts, particulates & fibers.
Table III-3.--Examples of Biologicals Found in Indoor Environments\1\
--------------------------------------------------------------------------------------------------------------------------------------------------------
Class Agent or component Origin
--------------------------------------------------------------------------------------------------------------------------------------------------------
Arthopods and Insects........................... Whole organism, body parts, feces................. Furnishings, building materials, food.
Microbes:
Algae....................................... Whole organism, cellular components............... Outdoor air, HVAC (rare).
Bacteria.................................... Whole organism, spores and cell walls, endotoxin.. Stagnant water, floods, cooling towers, industrial
processes.
Fungi....................................... Whole organism spores and hyphae toxins and Moist surfaces, HVAC system, bird droppings,
volatiles. outdoor air.
Protozoa.................................... Whole organism cellular components................ Water reservoirs, pets (rare).
Viruses..................................... Whole organism.................................... Humans and pets (rare).
Pets............................................ Skin, scales danders, urine, saliva, feces........ Pets, pet litter, pet cages, pet toys, pet
bedding.
Plants.......................................... Stems, leaves and pollens......................... Outdoor and indoor air.
--------------------------------------------------------------------------------------------------------------------------------------------------------
\1\Adapted from Ex. 4-33.
Table III-4.--Indoor Air Allergens Associated With Asthma\1\
------------------------------------------------------------------------
Class Typical examples
------------------------------------------------------------------------
Animal:
Avian................ High and low molecular weight proteins from
feathers and droppings.
Canine and Feline.... High and low molecular weight proteins from
dander, saliva, and feces.
Arthropods:
Mites, Cockroaches, Structural proteins, carbohydrates and
Crickets and Moths. metabolites.
Dusts, Particulates and
Fibers:
Household............ Pollens, fungi, danders and mites.
Metal................ Chromium, cobalt, nickel, platinum, and
vanadium.
Plant................ Castor bean, coffee, cotton, flour, and
grain.
Wood................. Oak, mahogany, redwood, red cedar.
Chemical Volatiles and Acrylates, amines, anhydrides, colophony,
Dusts. enzymes, epoxy resins, freon, furfuryl
alcohol, resins, isocyanates, latex,
organophosphates, polyvinyl chloride,
vegetable gums.
Microbes and Microbial
Products:
Bacteria............. Bacillus spp.
Fungi................ Alternaria spp., Aspergillus spp., Botrytis
spp., Cladosporium spp., Penicillium spp.,
Pullularia spp.
Pollens.................. Agrostis spp., Alopecurus spp., Anthoxanthum
spp. Cynosurus spp., Dactylis spp., Holcus
spp., Lolium spp., Secale spp.
------------------------------------------------------------------------
\1\Source: Ex. 4-33.
Exposures that cause hypersensitivity reactions include
microorganisms, fumes, vapors, and dusts (Table III-5). These exposures
are associated with the development of hypersensitivity pneumonitis or
a less serious variant, humidifier fever [Ex. 4-33]. Many of these
contaminants are found in the nonindustrial workplace. Birds and
rodents are common pests. Air intakes can be contaminated with bird
droppings and other avian-associated problems when used as nesting
sites. These problems can affect the quality of the air being brought
into the ventilation system through these air intakes. Rodent
infestations affect work areas directly. Many of the chemicals listed
in Table III-5 are commonly found in most workplaces.
In summary, exposure to contaminants in nonindustrial workplaces
will vary according to the characteristics of the building. These
include its age, types of materials used in construction and the type
of equipment and supplies that are used by building occupants. The
design, maintenance, and operation of the building's HVAC system as
well as the general housekeeping of the building, can greatly influence
the levels of contaminants that exist.
OSHA requests data on the levels of these contaminants in
nonindustrial workplaces.
Table III-5.--Indoor Air Contaminants Associated With Hypersensitivity
Pneumonitis\1\
------------------------------------------------------------------------
Class Typical examples
------------------------------------------------------------------------
Animals:
Avian................ High and low molecular weight proteins from
feathers and droppings.
Rodent............... Low molecular weight proteins from urine and
feces.
Arthropods:
Weevils.............. Sitophilus spp.
Mites................ Ascaris spp.
Altered Host Proteins or Amines, anhydrides, epoxy resins vegetable
Chemical Hapten-Carrier gums, and isocyanates.
Conjugates.
Microbes:
Bacteria............. Thermoactinomycetes spp., Bacillus spp.
Fungi................ Aspergillus spp., Auerobasillium spp.,
Cephalosporium spp., Penicillium spp.
Organic Dusts &
Particulates:
Wood................. Bark, Sawdust and Pollen.
Grain................ Arthropod- and microbially-contaminated
grains and flours.
Cleaning Products.... Dust residues from carpet cleaning agents.
------------------------------------------------------------------------
\1\Source: Ex. 4-33.
C. Exposure Studies
1. Low-level Contaminants
Experimental studies have demonstrated that exposure of susceptible
people to low level mixtures of VOCs have induced mucous membrane
irritation and pulmonary effects. Some of these studies are discussed
below.
The potential of indoor air contamination to produce adverse
effects in humans was demonstrated by Molhave et al. in Denmark [Ex. 4-
20]. These researchers studied 62 subjects suffering from ``indoor
climate symptoms''. These subjects reported primarily eye and upper
respiratory tract irritation, but were otherwise healthy individuals
that did not suffer from asthma, allergy, or bronchitis. The subjects
were exposed to a mixture of VOCs in concentrations of 0, 5, or 25 mg/
m3. These concentrations respectively represented ``clean'' air,
average polluted air, and the maximum polluted air in Danish
households. After exposure, a Digit Span test was administered. The
study found significant declines in performance on this test;
demonstrating that low-level exposures to volatile organic compounds
had an adverse effect on the ability to concentrate [Ex. 4-20].
Otto et al. [Ex. 4-248], repeating the Molhave et al. (1984)
experiment, studied 66 healthy subjects with no history of eye and
upper respiratory tract irritation. These subjects were exposed at 0
and 25 mg/m3 VOC-contaminated air. Otto et al. reported that while
subjects found the odor of chemicals unpleasant, to degrade indoor air
quality, to increase headache, and produce general discomfort, VOC
exposure for 2.75 hours duration did not affect performance on any
behavioral tests. These results imply that persons who experience
symptoms of SBS may have a lower threshold for certain health effects
compared to nonreactive people. This suggests that those with
compromised immune response (e.g. allergy sufferers) may be at elevated
risk of SBS.
Ahlstrom, et al. [Ex. 4-2] found that synergistic effects may occur
when one strong indoor irritant interacts with other indoor
contaminants present at low-level concentrations. Ahlstrom et al. found
that there was almost a 4-fold increase in the perceived odor strength
of formaldehyde at low concentration (0.08 ppm) when mixed with 100%
indoor air from a building where SBS was reported, relative to 10%
indoor air from the same building.
The Report of the Canadian Interministerial Committee on Indoor Air
Quality [Ex. 4-264] adopts the World Health Organization's definition
of health: ``Health refers to a state of complete physical, mental, and
social well being, and not just the absence of disease or infirmity.''
This definition was adopted to allow the setting of indoor air quality
guidelines based on ``comfort'' as well as ``health''. The report
observes that the symptoms of SBS are sufficiently general or
subjective that they may be indicative of several other medical
conditions. Therefore, perhaps the best indicator that workplace
exposure may play a role in the symptoms reported by an individual is
the observation that symptoms worsen during the work day, and disappear
shortly after leaving work. They state that because there is a wide
variation in individual susceptibility, based on genetics, age,
medication, previous exposure to pollutants, gender, and state of
health, especially those with allergies, that certain individuals may
be more sensitive to SBS than others.
2. Bioaerosols
The levels of bioaerosols in the indoor environment should reflect
those found in the outdoor environment. A rank order assessment,
comparing the abundance of microorganisms in the outdoor versus indoor
environment is one way of assessing this relationship [Exs. 3-61, 4-
229]. If indoor and outdoor sampling results are not comparable, then
it is possible that a reservoir of a particular microbe may be
amplifying in the indoor environment; especially if moisture and a
nutrient-rich substrate are available [Ex. 4-229]. An example of this
would be Legionella. Commonly found in the outdoor environment, the
bacteria are as expected, commonly found in untreated potable and
nonpotable water. Situations can occur that allow these reservoirs to
amplify not only in potable water and hot water service systems but
also water used in cooling towers and evaporative condensers [Ex. 4-
229]. Infection occurs if the bacteria are disseminated, either through
the HVAC system or potable water system (e.g., showers) to the
breathing zone of a susceptible person. A healthy individual may
develop the less severe Pontiac Fever. An individual that smokes or is
older may develop the more serious pneumonia [Exs. 4-33, 4-229].
3. Environmental Tobacco Smoke
The burning of tobacco in enclosed workplaces releases an aerosol
containing a large variety of solid, liquid, and gas phase chemical
compounds. Generation of tobacco smoke is governed by the source
emission characteristics of smokers and their tobacco products, whereas
removal is primarily determined by the rate of replacement of building
air by outside air, with re-emission of surface-sorbed compounds
playing a minor role. Natural and mechanical ventilation systems are
designed primarily to limit the accumulation of the products of human
respiratory metabolism, and secondarily to limit odor; not to control
the byproducts of biomass combustion. Thus, smoking indoors creates air
pollution which is not adequately abated by customary ventilation
systems.
Exposure to tobacco smoke primarily occurs through the inhalation
route. Such an exposure can be measured by the determination of the
absorption, distribution, metabolism and excretion of tobacco smoke
constituents and/or their metabolites. However, relatively few of these
individual constituents have been identified and characterized. Also,
measurement of all components in tobacco smoke is not feasible.
Therefore, it becomes necessary to identify a marker which, when
measured, will accurately represent the frequency, duration and
magnitude of the exposure to environmental tobacco smoke.
This discussion reviews available data for the purposes of
assessing exposure to ETS in the workplace. Nonsmokers are exposed to
mainstream smoke after it has been exhaled by smokers, and to diluted
sidestream smoke. Issues covered include activity patterns affecting
the duration of nonsmokers' exposures, the concentrations of ETS in
buildings, the comparison of ETS components in indoor workplaces,
levels of biomarkers in workers, and the inadequacy of general dilution
ventilation to address ETS exposure control. This discussion will
indicate not only that exposure occurs, but that nonsmokers absorb ETS
components.
(a) Chemistry. Pipe, cigar, and cigarette smoke all contribute to
environmental tobacco smoke (ETS) but cigarette smoke is of principal
interest because it is by far the most common. Tables III-6 and III-7
list some of the known constituents of tobacco smoke.
The combustion of tobacco leads to the formation of mainstream
smoke (MS) and sidestream smoke (SS). MS is generated during puff-
drawing in the burning cone and hot zones; it travels through the
tobacco column and is inhaled by the smoker. The smoke which is exhaled
by the smoker, while different from the inhaled smoke, is also
considered ``mainstream.'' SS is formed in between puff-drawing and is
emitted directly from the smoldering tobacco product into the ambient
air.
Table III-6.--Vapor Phase Constituents of Tobacco Smoke and Related Health Effects
--------------------------------------------------------------------------------------------------------------------------------------------------------
Constituent Amount in MS Ratio in SS/MS Health effects
--------------------------------------------------------------------------------------------------------------------------------------------------------
Carbon monoxide............................................ 10-23 mg............. 2.5-4.7 Nervous system, cardiovascular system.\1\
Carbon dioxide............................................. 20-40 mg............. 8-11 Nervous system, cardiovascular system.\1\
Carbonyl sulfide........................................... 12-42 g..... 0.03-0.13 Irritant, cardiovascular, and nervous systems.\1\
Benzene.................................................... 12-48 g..... 5-10 Known human\3\ carcinogen.
Toluene.................................................... 100-200 g... 5.6-8.3 Irritant, nervous system.\1\
Formaldehyde............................................... 70-100 g.... 0.1- Probable human carcinogen.\3\
50
Acrolein................................................... 60-100 g.... 8-15 Irritant, pulmonary.\1\
Acetone.................................................... 100-250 g... 2-5 Irritant.\1\
Pyridine................................................... 16-40 g..... 6.5-20 Irritant, nervous system, liver, kidney.\1\
3-methylpyridine........................................... 12-36 g..... 3-13 Irritant.\2\
3-vinylpyridine............................................ 11-30 g..... 20-40 Irritant.\2\
Hydrogen cyanide........................................... 400-500 g... 0.1-0.25 Irritant, nervous, cardiovascular and pulmonary
system.\1\
Hydrazine.................................................. 32 ng................ 3 Probable human carcinogen.\3\
Ammonia.................................................... 50-130 g.... 3.7-5.1 Irritant.\1\
Methylamine................................................ 11.5-28.7 g. 4.2-6.4 Irritant.\1\
Dimethylamine.............................................. 7.8-10 g.... 3.7-5.1 Irritant\1\.
Nitrogen oxides............................................ 100-600 g... 4-10 Pulmonary and cardiovascular system.\1\
N-nitrosodimenthylamine.................................... 10-40 ng............. 20-100 Probable human carcinogen.\3\
N-nitrodiethylamine........................................ ND-25 ng............. <40 Probable human carcinogen.\3\
N-nitrosopyrrolidine....................................... 6-30 ng.............. 6-30 Probable human carcinogen.\3\
Formic acid................................................ 210-490 g... 1.4-1.6 Irritant, skin, kidney, liver\1\.
Acetic acid................................................ 330-810 g... 1.9-3.6 Irritant.\1\
Methyl chloride............................................ 150-600 g... 1.7-3.3 Nervous system.\1\
1,3-butadiene.............................................. 69.2 g...... 3-6 Probable human carcinogen.\3\
--------------------------------------------------------------------------------------------------------------------------------------------------------
\1\NIOSH Pocket Guide to Chemical Hazards. U.S. Department of Health and Human Services. Public Health Services, 1990. Ex. 4-238.
\2\Hazards in the Chemical Laboratory. Ed: L. Bretherick, The Royal Society of Chemistry, 1986. [Ex. 4-137]
\3\EPA: Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders, 1992. [Ex. 4-311]
Table III-7.--Particulate Phase Constituents of Tobacco Smoke and Related Health Effects
--------------------------------------------------------------------------------------------------------------------------------------------------------
Constituent Amount in MS Ratio in SS/MS Health effects
--------------------------------------------------------------------------------------------------------------------------------------------------------
Particulate matter contains di- and polycyclic aromatic 15-40 mg............. 1.3-1.9 Animal carcinogen.\4\
hydrocarbon.
Nicotine................................................... 1-2.5 mg............. 2.6-3.3 Nervous and cardiovascular system.\1\
Anatabine.................................................. 2-20 g...... <0.01-0.5 N/A.\5\
Phenol..................................................... 60-140 g.... 1.6-3.0 Irritant.\1\
Catechol................................................... 100-360 g... 0.6-0.9 Irritant.\3\
Hydroquinone............................................... 110-300 g... 0.7-0.9 N/A.\5\
Aniline.................................................... 360 ng............... 30 Probable human carcinogen.\4\
2-Toluidine................................................ 160 ng............... 19 Irritant, cardiovascular system.\1\
2-Naphthylamine............................................ 1.7 ng............... 30 Known human carcinogen.\4\
4-Aminobiphenyl............................................ 4.6.................. 31 Known human carcinogen.\4\
Benz[a]anthracene.......................................... 20-70 ng............. 2-4 Animal carcinogen.\4\
Benzo[a]pyrene............................................. 20-40 ng............. 2.5-3.5 Probable human carcinogen.\4\
Cholesterol................................................ 22 g........ 0.9 N/A.\5\
-butyrolactone.................................... 10-22 g..... 3.6-5.0 Animal carcinogen.\4\
Quinoline.................................................. 0.5-2 g..... 3-11 Irritant.\3\
Harman [1-methyl-9H-pyrido[3,4-b]-indole................... 1.7-3.1 g... 0.7-1.7 N/A.\5\
N-nitrosonornicotine....................................... 200-3000 ng.......... 0.5-3 Animal carcinogen.\4\
NNK [4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone].. 100-1000 ng.......... 1-4 N/A.\5\
N-nitrosodiethanolamine.................................... 20-70 ng............. 1.2 Probable human carcinogen.\4\
Cadmium.................................................... 110 ng............... 7.2 Probable human carcinogen.\4\
Nickel..................................................... 20-80 ng............. 13-30 Known human carcinogen.\4\
Zinc....................................................... 60 ng................ 6.7 Irritant, nausea, vomiting.\2\
Polonium-210............................................... 0.04-0.1 pCi......... 1.04.0 Known human carcinogen.\4\
Benzoic acid............................................... 14-28 g..... 0.67-0.95 Irritant.
Lactic acid................................................ 63-174 g.... 0.5-0.7 Irritant.\3\
Glycolic acid.............................................. 37-126 g.... 0.60.95 Irritant.\2\
Succinic acid.............................................. 110-140 g... 0.43-0.62 N/A.\5\
PCDD's and PCDF's\6\....................................... 1 pg................. 2 N/A.\5\
--------------------------------------------------------------------------------------------------------------------------------------------------------
\1\NIOSH Pocket Guide to Chemical Hazards. U.S. Department of Health and Human Services. Public Health Services, 1990. Ex. 4-238.
\2\The Merck Index, 10th Edition, Merck & Co., Inc., 1983. Ex. 4-220.
\3\Hazards in the Chemical Laboratory. Ed: L. Bretherick, The Royal Society of Chemistry, 1986. [Ex. 4-137]
\4\EPA: Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders, 1992. [Ex. 4-311]
\5\N/A--Relevant information not available.
\6\PCDDs--Polychlorinated dibenzo-p-dioxins; PCDFs--Polychlorinated dibenzofurans.
MS and SS cigarette smoke are chemically and physically complex
mixtures consisting of electrically charged submicron liquid particles
at very high concentration consisting of permanent gases, reactive
gases, and a large variety of organic chemicals. The composition of the
smoke and especially the total quantities of individual constituents
delivered are dependent on the conditions of smoke generation [Ex. 4-
311].
Nicotine, while found in the particulate phase in MS, is found
predominantly in the gas phase in ETS [Ex. 4-100]. The differences in
size distribution for MS and SS particles, as well as the different
breathing patterns of smokers and nonsmokers, affect deposition of the
produced particle contaminants in various regions of the respiratory
tract.
There are substantial similarities and some differences between MS
and SS emissions from cigarettes [Exs. 3-689D, 4-129, 4-239].
Differences in MS and SS emissions are due to differences in the
temperature of the combustion of tobacco, pH, and degree of dilution
with the air, which is accompanied by a correspondingly rapid decrease
in temperature. SS is generated at a lower temperature (approximately
600 deg.C between puffs versus 800 to 900 deg.C for MS during puffs)
and at a higher pH (6.7-7.5 versus 6.0-6.7) than MS. Being slightly
more alkaline, SS contains more ammonia, is depleted of acids, contains
greater quantities of organic bases, and contains less hydrogen cyanide
than MS. Differences in MS and SS are also ascribable to differences in
the oxygen concentration (16% in MS versus 2% in SS). SS contaminants
are generated in a more reducing environment than those in MS, which
will affect the distribution of some compounds. Nitrosamines, for
example, are present in greater concentrations in SS than in MS.
Many of the compounds found in MS, which were identified as human
carcinogens, are also found in SS emissions [Exs. 3-689D, 4-93, 4-129,
4-239, 4-269] and at emission rates considerably higher than for MS. SS
contains ten times more polycyclic aromatic hydrocarbons, aza-arenes
and amines as compared with MS [Ex. 4-126]. All of the five known
carcinogens, nine probable human carcinogens, and three animal
carcinogens are emitted at higher levels in SS than in MS, several by
an order of magnitude or more. Several toxic compounds found in MS are
also found in SS (carbon monoxide, ammonia, nitrogen oxides, nicotine,
acrolein, acetone, etc.), in some cases by an order of magnitude or
higher (Tables III-6 and III-7).
SS emissions, quantitatively, show little variability as a function
of a number of variables (puff volume, filter versus nonfilter
cigarette, and filter ventilation [Exs. 4-1, 4-34, 4-54, 4-128, 4-129,
4-141]. The lack of substantial variability in SS emissions is related
to the fact that they are primarily related to the weight of tobacco
and paper consumed during the smoldering period, with little influence
exerted by cigarette design [Ex. 4-129].
(b) Human Activity Pattern Studies Used to Assess Workplace
Exposure. Human activity pattern studies utilize random samples of
human activity patterns using questionnaires and time-diary data to
provide detailed generalizable data about human behavior. Such studies
have been used to assess exposure to ETS. In 1987-1988, the California
Air Resources Board sponsored a probability-based cross-sectional
sample of 1,579 Californians aged 18 years and older, called the
California Activity Pattern Survey (CAPS) [Exs. 4-168, 4-271]. The
study was designed to provide information on time spent in various
locations, including indoors, outdoors, and in transit, as well as
specific microenvironments, such as living rooms, kitchens,
automobiles, or buses. The study focused on time spent in activities
such as cooking or playing sports, but more specifically targeted
activities and environments that had implications for air pollution
exposure, such as the presence of smokers, use of cooking equipment or
solvents.
In analyzing the data from CAPS, Jenkins et al. [Ex. 4-168] and
Robinson et al. [Ex. 4-271] found that time spent at work had a high
correlation with exposure to ETS. This association of ETS exposure with
work settings remained strong after controlling for the length of the
activity episode, and hence was not simply a function of longer time
intervals at work. Robinson et al. [Ex. 4-271] also found that men
reported higher levels of exposure than women, even after controlling
for age, employment status, shorter working hours, etc. This finding
suggests that the epidemiological studies of passive smoking and lung
cancer, which have focussed on women, may be underestimating the effect
of ETS on lung cancer.
Further analysis of the CAP study [Ex. 4-169] verifies the high
percentage of nonsmokers who are exposed to ETS while at work. This is
indicated when the data are analyzed by employed nonsmoker status. As
indicated in Table III-8, 51% of male and 38% of female nonsmokers
reported ETS exposure at work. The average duration of this exposure
was 313 minutes for males and 350 minutes for females. When the group
that reported exposure at the workplace is analyzed further it becomes
apparent that the overwhelming exposure location for these employed
nonsmokers is the workplace (Table III-9). As indicated in Table III-9,
77% of males and 85% of females were exposed an average of 313 minutes
and 350 minutes, respectively.
One other finding is that the more time spent at work, the higher
the likelihood of greater ETS exposure. For example, the average
duration of exposure to homemakers was approximately 2 hours a day, for
workers the average duration of exposure was approximately 3 hours a
day.
Table III-8.--Percentage of Employed Nonsmokers Exposed to ETS and
Average Minutes of Exposure (in Parentheses)\1\
------------------------------------------------------------------------
Exposure location Males Females Total
------------------------------------------------------------------------
Home............................. 9(134) 13(109) 11(123)
Work............................. 51(313) 38(350) 46(324)
Other indoor..................... 28(89) 35(77) 31(85)
Outdoor.......................... 12(118) 14(79) 13(104)
------------------------------------------------------------------------
\1\Source: [Ex. 4-169].
Table III-9.--Percentage of Employed Nonsmokers Exposed to ETS and
Average Minutes of Exposure (in Parentheses) of Those Who Reported ETS
Exposure at Work\1\
------------------------------------------------------------------------
Exposure location Males Females Total
------------------------------------------------------------------------
Home............................. 1(147) 2(180) 2(158)
Work............................. 77(313) 85(350) 80(324)
Other indoor..................... 15(92) 9(102) 13(94)
Outdoor.......................... 6(176) 4(140) 5(166)
------------------------------------------------------------------------
\1\Source: [Ex. 4-169].
Work breaks and meals at work were the work activities most closely
associated with ETS exposure, 51% and 35% respectively versus 27% for
work per se [Ex. 4-271]. In other words, nonsmokers experienced ETS
exposure in break areas more than in general work areas.
When white collar versus blue collar workplaces were compared, 37%
of factories/plants versus 22% of offices had episodes of ETS exposure,
suggesting that blue collar nonsmoking workers have a greater exposure
to ETS than white collar workers. For the CAP population, twice as many
workers were employed in offices as were in factories [Ex. 4-271]. The
most ETS exposed nonsmokers were those with 10 or more hours per day of
work (especially at plants/factories), more than 2 hours per day of
restaurant time, and more than 1 hour per day of bar or nightclub time.
Robinson et al. [Ex. 4-271] concluded that the probability of
passive smoking is highest for a combination of various social and work
activities, consistent with the notion that activities that involve
more people involve a greater chance of contact with people who smoke.
A limitation of the CAP survey is that the data do not provide
information on the intensity of exposure in the various
microenvironments [Ex. 4-271].
In summary, the CAP study showed that the most powerful predictor
of potential exposure to ETS was being employed. Respondents who spent
more than ten hours a day at the workplace were found to report more
ETS exposure than those working less than 10 hours a day or not at all.
Further data from this study show that the workplace is the location
with the highest reported exposure to ETS in enclosed environments, and
such exposure is on average nearly three times more prevalent at work
than at home.
Another relevant data source for assessing ETS exposure in the
workplace is the National Health Interview Survey (NHIS) conducted by
the Centers for Disease Control and Prevention (CDC). In its Health
Promotion and Disease Prevention (NHIS-HPDP) supplement, CDC collected
self-reported information on smoking from a representative sample of
the U.S. population [Ex. 4-51]. The results suggest that at least 19%
of employed nonsmokers experience ETS exposure at work. The CDC study
results represent the prevalence of occupational exposure among
nonsmoking adults [see section IV for further discussion of this
study].
In a smaller study, Cummings et al. [Ex. 4-67] studied the
prevalence of exposure to ETS in 663 (44% male) never- and exsmokers
aged 18-84 years, who attended a cancer clinic in Buffalo, New York in
1986 (see Table IV-9). The study employed questionnaires and analysis
of urinary cotinine levels. The subjects were asked if they were
exposed to passive smoke either at home or at work in the four days
preceding the interview. A further analysis of this data focusing on
workers from this survey determined that overall, 339 subjects were
currently employed. Of these 264 (77%) reported ETS exposure at work.
The percentage of subjects exposed to ETS at both work and the home was
29% (n=99). The percentage of subjects exposed at home, but not at work
was 7% (n=23). The percentage of subjects exposed at work, but not at
home was 49% (n=165). The percentage of subjects exposed neither at
home or work was 15% (n=52). This further analysis indicates that the
workplace is a significant source of ETS exposure for nonsmoking,
employed people.
Emmons et al. [Ex. 4-98] reported on a study of 186 nonsmoking
volunteers from workplace settings selected to have a wide range of
exposure to ETS. The subjects were asked to keep a 7-day exposure
diary. The worksites ranged from those with minimal restrictions and
high levels of exposure (long-term care and psychiatric facilities,
chemical dependency and treatment centers, and a VA Hospital) to those
with extensive restrictions and low exposure (e.g., state health
department and community hospitals). Seventy-six percent of the
subjects reported being regularly exposed to ETS in the workplace. The
percentage of subjects reporting exposure at work is similar to that
found by Cummings et al. [Ex. 4-67]. Nonsmokers encountered
significantly more exposure to ETS at work (50%) as compared to home
(10%). When the data set was examined by the presence or absence of
smokers in the home, however, subjects who lived with smokers had
virtually equivalent exposures across all three settings: work (34%),
home (36%), and ``other'' (31%). Nonsmokers living with smokers
received 29 minutes per day of exposure at work and 31 minutes per day
at home and 27 minutes per day in other settings. On the other hand,
subjects who did not live with smokers had the majority of their
exposure at work (36 minutes per day) and very little at other
settings.
Additional studies verify that the workplace is an important source
of exposure to ETS, particularly for nonsmokers unexposed at home [Exs.
4-172, 4-262, 4-315]. A U.K. study of exposure to ETS in 20 nonsmoking
men whose wives smoked showed that 78% of the men's reported hours of
exposure came from outside the home; by contrast, 90% of the ETS
exposure of 101 nonsmoking men whose wives did not smoke was reported
to come from non-domestic microenvironments [Ex. 4-315]. Repace and
Lowrey [Ex. 4-262] estimated that 86% of the U.S. population was
exposed to ETS, and that the workplace was more important than the home
as a source of ETS exposure, when weighted by the duration, exposure
intensity, and probability of exposure. Kabat and Wynder [Ex. 4-172],
in a study of 215 sixty-year-old U.S. women nonsmokers, found that 65%
reported exposure to ETS at home and 67% reported exposure at work,
averaged over adulthood.
The conclusion that can be made from the activity surveys is that
the workplace is a major location of ETS-exposure to nonsmokers. Human
activities that involve contact with a greater number of people
increase the probability of contact with smokers, and thus with ETS.
These studies indicate that the workplace, with its high person
densities relative to other microenvironments, including the home,
appears to be a major factor in the working nonsmoking population's ETS
exposure.
(c) Indoor Levels of Environmental Tobacco Smoke Constituents.
Personal monitoring studies have confirmed the role of the workplace as
an important microenvironment of ETS exposure to nonsmokers. Spengler
et al. [Ex. 4-288] and Sexton et al. [Ex. 4-280] demonstrated by
personal monitoring of respirable suspended particulates (RSP) and the
use of time-activity questionnaires that exposures to ETS both at home
and at work are significant contributors to personal RSP exposures.
Coultas et al. [Ex. 4-66], in a pilot study of 15 nonsmokers in
Albuquerque, New Mexico, collected questionnaires and samples of saliva
and urine to determine workplace ETS exposure. Personal air samples
were obtained pre- and post-workshift. Exposure to ETS was reported by
13 of the 15 subjects. The mean number of hours of exposure was 3.4
(2.1). Basically, although the levels of cotinine,
respirable particles, and nicotine varied with self-reports of ETS
exposure, the general trend was a direct relationship between
increasing incidence of self-reporting of exposure and actual biomarker
data. Coghlin, Hammond, and Gann [Ex. 4-61] found similar results for
53 nonsmoking volunteers studied by use of personal nicotine monitors,
diaries, and questionnaires. They also found that the closer a
nonsmoker was to a smoker, the higher the probability that the
nonsmoker would report exposure.
Presently, vapor phase nicotine and respirable suspended
particulate matter (ETS-RSP) are the most commonly used markers for ETS
because of their ease of measurement, knowledge of their emission rate
from tobacco combustion, and their relationship to other ETS
contaminants [Ex. 4-311]. Controlled experiments have shown that vapor
phase nicotine varies with the source strength, and shows little
variation among brands of cigarettes. Field studies have also shown
that vapor phase nicotine concentrations are correlated with the number
of cigarettes smoked, and further that weekly average nicotine
concentrations are correlated with ETS-RSP [Ex. 4-311].
(d) Levels of Respirable Suspended Particulates and Nicotine Found
in Field Studies. Respirable suspended particulates (RSP) and nicotine
are the most commonly used surrogates for ETS exposure [Ex. 4-239].
Both chamber and field studies have demonstrated that tobacco
combustion has a major impact on indoor RSP mass when particle size is
under 2.5 microns [Ex. 4-239]. A few examples illustrating the impact
of ETS on nicotine and RSP concentrations in workplace and domestic
microenvironments are shown in Tables III-10 and III-11. Studies of RSP
in public access buildings by Leaderer et al. [Ex. 4-190], First [Ex.
4-105], and Repace and Lowrey [Exs. 4-260, 4-261] (a total of 42
smoking buildings and 21 nonsmoking buildings) showed that the weighted
average RSP level during smoking in the smoking buildings was 262
g/m\3\, while in the nonsmoking buildings the RSP level
average 36 g/m\3\.
Leaderer and Hammond [Ex. 4-189] measured weekly average vapor
phase nicotine and RSP concentrations in 96 residences. Vapor phase
nicotine measurements were found to be closely related to number of
cigarettes smoked and highly predictive of RSP generated by tobacco
combustion. The mean RSP background in the absence of measurable
nicotine was found to be 15.27 g/m\3\. The mean
RSP value in the presence of nicotine was 44.130
g/m\3\. The weekly mean nicotine concentration in the 47
residences with detectable nicotine values was 2.17 g/m\3\
(Table III-10).
Summary statistics of additional studies on personal monitoring for
nicotine are shown in Table III-11 [Ex. 4-263]. These studies show that
the median exposures ranged from 5 to 20 g/m\3\.
Summary nicotine data analyzed by the U.S. EPA [Ex. 4-311] suggest
that average nicotine values in residences where smoking is occurring
will average 2 to approximately 10 g/m\3\, with peak values of
0.1 to 14 g/m\3\ as shown in Table III-10. Offices with
smoking occupants show a range of average nicotine concentrations
similar to that of residences, but with considerably higher peak
values. RSP mass concentrations in smoker-occupied residences show
average increases of from 18 to 95 g/m\3\, with individual
increases as high as 560 g/m\3\ or as low as 5 g/
m\3\. ETS-RSP concentrations in offices with smoking occupants on
average appear to be about the same as in residences. Restaurants,
transportation, and other indoor spaces with smoking occupants have a
generally wider range of increases in particle mass concentrations due
to ETS than residential or office environments [Ex. 4-311].
In summary, field data show that RSP is elevated by one to two
orders of magnitude during smoking, and that nicotine released during
smoking is easily detectable in both homes and workplaces by area or
personal monitors. Offices with smoking occupants show a range of
average nicotine concentrations similar to that of residences (2 to 10
g/m\3\), but with considerably higher maximum values. ETS-RSP
concentrations in offices with smoking occupants on average appear to
be about the same as residences (18 to 95 g/m\3\).
Restaurants, transportation, and other indoor spaces with smoking
occupants have a generally wider range of particle mass concentrations
due to ETS than residential or office environments [Ex. 4-311]. It must
be noted that measurements of nicotine and ETS-RSP in indoor spaces do
not constitute a direct measure of total exposure. Concentrations
measured in all microenvironments have to be combined with human
activity pattern studies to determine the time-weighted sum of various
exposures.
(e) Biomarkers of Environmental Tobacco Smoke Exposure. Nicotine,
and its metabolite, cotinine, and other tobacco smoke constituents in
the saliva, blood and urine have been used as biomarkers of active and
passive smoking. Nicotine and cotinine can be used to determine the
integrated short-term exposure of ETS across all microenvironments [Ex.
4-311].
Table III-10.--Mean Nicotine Levels in Home and Workplace Air: Area Monitors\1\
----------------------------------------------------------------------------------------------------------------
g/
Study and location Sample m\3\ Comment
----------------------------------------------------------------------------------------------------------------
Leaderer and Hammond 1991, homes, NY State........ 47 2.17 7-day average smoking.
Hammond [3-1096] Mass., industrial................ ........... 24 9-hour average workshift
(nonsmoker's air; smoking allowed
on premises).
White collar.................................. 60 21.5 ..................................
Blue collar................................... 123 8.9 ..................................
Food service.................................. 51 10.3 ..................................
Carson (1988), offices, Canada.................... 31 11 Workday samples.
Miesner (1989) workplaces, MA..................... 11 6.6 Workweek average.
Oldaker (1990), restaurants, NC................... 33 10.5 1-hour average (range).
Jenkins (1991), Knoxville, TN, metro.............. ........... ........... 1-hour average.
Restaurants................................... 7 3.4 ..................................
Cocktail lounges.............................. 8 17.6 ..................................
Bowling alleys................................ 4 10.7 ..................................
Gaming parlors................................ 2 10.7 ..................................
Laundromats................................... 3 2.0 ..................................
Airport gates................................. 2 6.0 ..................................
Office........................................ 1 6.0 ..................................
Nagda (1989), U.S. aircraft--in-flight average:
All flights................................... 69 13.4 Smoking section.
Domestic...................................... 61 0.11 Nonsmoking section.
International................................. 8 0.33 Nonsmoking section.
Vaughn (1990), highrise office building........... 1 2.0 Nonsmoking air; 9-hour average.
----------------------------------------------------------------------------------------------------------------
\1\Adapted from Repace and Lowrey 1993 [Ex. 4-263].
Table III-11.--Nicotine in Nonsmokers' Air: Personal Moitors\1\
----------------------------------------------------------------------------------------------------------------
g/
Study and Location Sample m3 Comment
----------------------------------------------------------------------------------------------------------------
Schenker (1990), railroad clerks, NE.............. 40 6.9 Workshift median.
Coultas (1990), white collar, NM.................. 15 20.4 Workshift mean SD.
Mattson (1989); flight attendants................. 4 4.7 4 flights, mean SD.
----------------------------------------------------------------------------------------------------------------
\1\Adapted from Repace and Lowrey 1993 [Ex. 4-263].
Both nicotine and cotinine are tobacco-specific. Cotinine in
saliva, blood, and urine is the most widely accepted biomarker for
integrated exposure to both active smoking and ETS by virtue of its
longer half-life than nicotine in body fluids. The half-life of
cotinine in nonsmokers is of the order of a day, making it a good
indicator of integrated ETS exposure over the previous day or two [Ex.
4-311]. Although intersubject variability exists for both nicotine
absorption and cotinine metabolism [Exs. 4-156, 4-162], cotinine is a
good indicator that ETS exposure has taken place [Ex. 4-311]. Further,
studies show that cotinine levels correlate with levels of recent ETS
exposure [Ex. 4-311].
In summary, nonsmokers' exposure to ETS has been characterized by a
database of widely used atmospheric and biological markers which have
been measured in a number of workplaces, such as offices, restaurants,
commercial buildings, and on trains and in planes. OSHA believes that
this database is sufficient to support the risk assessment which
follows. ETS-nicotine exposures of the average worker appear to be of
the order of 5 to 10 micrograms per cubic meter (g/m\3\), and
for the most-exposed workers, 50 to 100 g/m\3\). For ETS-RSP,
exposures are about tenfold that of the nicotine levels. The
concentrations of various ETS atmospheric markers to which nonsmokers
are exposed in the workplace, such as nicotine, respirable suspended
particulate matter (RSP) and carbon monoxide, are linearly correlated
with the amount of tobacco burned. Studies of human activity patterns
show that the workplace is the largest single contributor to ETS
exposure. Air exchange rates in nonindustrial workplaces are not
designed to control the risks of ETS exposure.
(f) Inadequacy of General Dilution Ventilation to Address
Environmental Tobacco Smoke Exposure Control. A primary function of
heating, ventilating, and air-conditioning (HVAC) systems is to
circulate air throughout a building to achieve thermal and sensory
comfort for the building occupants. The general ventilation function of
the HVAC system is to dilute and remove occupant generated bioeffluents
and other contaminants from the space. However, from the industrial
hygiene perspective, general ventilation as delivered by a HVAC system,
is not an acceptable engineering control measure for controlling
occupational exposures to ETS.
Dilution ventilation offers no protection in those cases where, due
to the close proximity to a smoker (e.g., contaminant point source),
the nonsmoking employee may be exposed to large amounts of sidestream
smoke and exhaled mainstream smoke (ETS). Due to the limitations of
general ventilation, the smoke cannot be removed from the air before
reaching the breathing zone of nearby employees. The carcinogenicity of
ETS discounts the use of general ventilation as an engineering control
for this contaminant.
The major ventilation guidance document available to HVAC
practioners (e.g., designers, maintenance, and operators), is Standard
62-1989 titled ``Ventilation for Acceptable Indoor Air Quality'' [Ex.
4-333]. The standard is published by the American Society of Heating,
Refrigerating and Air-Conditioning Engineers, Inc. (ASHRAE) and it
specifies recommended minimum design outside air ventilation rates for
91 different applications. Based on this current ventilation standard,
a typical commercial HVAC system serving general office space should
prescriptively deliver 20 cubic feet per minute per person (cfm/person)
of outside air to the occupied space to dilute occupant generated
contaminants like carbon dioxide (CO2) and body odors. This
ventilation rate would provide what ASHRAE defines as ``acceptable
indoor air quality'' (e.g., sensory comfort) to satisfy at least 80% of
the building occupants. The prescribed ventilation rates in ASHRAE
Standard 62-1989 are proportional to the occupants in the space (e.g.,
cfm/PER PERSON) because of the presumption that the contamination
produced is in proportion to the occupant density.
The foreword of ASHRAE Standard 62-1989 states ``with respect to
tobacco smoke and other contaminants, this standard does not, and
cannot, ensure the avoidance of all possible adverse health effects,
but it reflects recognized consensus criteria and guidance.'' As
published, ASHRAE Standard 62-1989 did not include any summary and/or
explanation documentation which would explain the basis of the
consensus standard. Without this documentation, it can only be inferred
that the standard was mostly based on satisfaction of sensory comfort
rather than based on the control of contaminants like ETS which may
contribute to adverse health effects like lung cancer and heart
disease.
The method of room air distribution found in most HVAC systems is a
mixing system that attempts to create an environment of uniform air
velocities, temperatures and humidities in the occupied zone of a room
(e.g.; floor to 6 feet above floor). In this occupied zone, air
velocities less than 50 feet per minute (fpm) and minimization of
temperature gradients will promote occupant comfort. In a conventional
mixing system where the supply air diffusers (outlets) and the return
air grilles are both located in the ceiling, the air motion in the
occupied zone could be characterized as ``gentle drift'' toward the
ceiling where the room air is then mixed with the conditioned air being
delivered to the room through the supply air diffusers [1993 ASHRAE
Handbook, Ch.31]. Because of natural convection currents and thermal
buoyancy forces it is common, especially during heating season, to have
stagnant zones. In a mixing room air distribution system, the emphasis
is on comfort.
There are other room air distribution schemes which consider
contaminant control and have been used in the industrial environment
like displacement ventilation and unidirectional (plug-flow) airflow
ventilation. In these schemes, there is an attempt to move contaminants
directionally along a clean to less clean gradient. These schemes are
seldom used in conventional HVAC systems due to their cost, feasibility
and compromise of comfort issues.
From the industrial hygiene perspective, local exhaust ventilation,
specific to each source, would be the preferred and recommended method
for controlling occupational exposures to contaminant point sources
like ETS. Such specific ventilation is effective because the
contaminant is captured or contained at its source before it is
dispersed into the work environment where only ineffective general
dilution ventilation is available to control exposures.
A designated smoking area which is enclosed, exhausted directly to
the outside, and maintained under negative pressure is sufficient to
contain tobacco smoke within the designated area. Such areas could be
considered an application of local exhaust ventilation because the
contaminant is being exhausted from a confined source without dispersal
into the general workspace.
IV. Preliminary Quantitative Risk Assessment
A. Introduction
The determining factor in the decision to perform a quantitative
risk assessment is the availability of suitable data for use in such an
assessment. A wide spectrum of health effects have been associated with
exposure to indoor air pollutants and ETS. These effects range from
acute irritant effects to cancer. In the case of ETS, OSHA has
determined that data are available to quantify two types of risk: lung
cancer and heart disease. For this risk assessment, OSHA defines
``heart disease'' to be coronary heart disease excluding strokes, as
defined in the Framingham study [Ex. 4-108]. In the case of indoor air
pollutants, the only data available to OSHA were on specific acute
health effects, such as severe headaches, excluding migraines, and
other respiratory conditions, such as ``stuffy nose'', ``runny nose'',
etc. OSHA is aware that there are more serious conditions such as
legionellosis and hypersensitivity diseases associated with poor indoor
air and suspected to be potential occupational hazards. However, the
Agency currently does not have adequate data to conduct a quantitative
risk assessment addressing these risks in the workplace. OSHA is
continuing to develop appropriate methodology to address risk
estimations for conditions related to poor indoor air quality in the
workplace and is requesting input on data sources relevant to these
efforts.
There is uncertainty associated with the quantification of any kind
of risk. In this risk assessment, OSHA has tried to describe many of
the sources of uncertainty and to address their implications for OSHA's
estimates of risk.
For the purpose of this rulemaking and for deriving a quantitative
estimate of occupational risk, OSHA has concentrated on information and
data concerning heart disease and lung cancer as potential effects
associated with exposure to ETS.
B. Review of Epidemiologic Studies and Published Risk Estimates
As a first step in this risk assessment, OSHA critically reviewed
epidemiologic studies associating exposure to ETS or indoor air
pollutants with adverse health effects. The purpose of such a critical
evaluation was to determine whether exposure to ETS is a causal factor
in cancer and heart disease and whether exposure to indoor air
pollutants has caused a significant increase in acute irritant effects.
The critical review also enables OSHA to select those studies that have
potential for use in a quantitative risk assessment. Tables IV-1 and
IV-2 contain a summary of OSHA's assessment of several epidemiologic
studies of ETS exposed individuals.
OSHA evaluated studies on exposure to ETS to determine the
importance and weight of each study in the overall hazard
identification process. Of those, it was determined that fourteen
showed a statistically strong association between exposure to ETS and
lung cancer and four showed a significant association between ETS
exposure and heart disease. Studies that were determined to be
``positive'' by OSHA's review standards met standard epidemiologic and
statistical criteria to support causation.
Overall, on the basis of the studies reviewed, OSHA concludes that
the relative risk of lung cancer in nonsmokers due to chronic exposure
to ETS ranges between 1.20 and 1.50 and the relative risk for heart
disease due to ETS exposure ranges between 1.24 and 3.00.
Table IV-1.--Epidemiologic Studies Reviewed by OSHA--Lung Cancer
----------------------------------------------------------------------------------------------------------------
Positive Equivocal positive trend Equivocal
----------------------------------------------------------------------------------------------------------------
Brownson et al. (1992)..................... Akiba et al..................... Brownson et al. (1987).
Correa et al............................... Butler.......................... Buffler et al.
Fontham et al.............................. Gao et al....................... Chan and Fung.
Garfinkel et al............................ Gillis et al.................... Hole et al.
Geng et al................................. Kabat and Wynder................ Janerich et al.
Hirayama 1984a............................. ................................ Katada et al.
Humble..................................... ................................ Koo et al.
Inoue et al................................ ................................ Lee et al.
Kalandidi et al............................ ................................ Shimizu et al.
Lam et al.................................. ................................ Sobue et al.
Pershagen et al............................ ................................ Svenson et al.
Sandler et al.............................. ................................ Wu et al.
Stockwell et al............................ ................................ .................................
Trichopoulos et al......................... ................................ .................................
----------------------------------------------------------------------------------------------------------------
Table IV-2.--Epidemiologic Studies Reviewed by OSHA Heart Disease
----------------------------------------------------------------------------------------------------------------
Positive Equivocal positive trend Equivocal
----------------------------------------------------------------------------------------------------------------
Dobson et al............................... Gillis et al.................... Garland et al.
He 1989.................................... Hole et al...................... Lee et al.
Helsing et al.............................. Humble et al.................... .................................
Sandler et al.............................. Svendsen et al.................. .................................
Hirayama 1964.............................. ................................ .................................
----------------------------------------------------------------------------------------------------------------
Other relative risk estimates based on summaries of studies on ETS
exposure performed by independent scientists and other government
agencies are found in Tables IV-3 and IV-4. OSHA is not aware of any
published risk assessments for overall exposure to indoor air
pollutants.
Table IV-3.--Published Risk Estimates for Lung Cancer
------------------------------------------------------------------------
Estimates of relative
Study risk\1\
------------------------------------------------------------------------
Daleger et al. [Ex. 4-78]................. 1.47(.076-2.83)
NRC 1986 [Ex. 4-239]...................... 1.34(1.18-1.53)
Repace and Lowry [Ex. 4-263].............. 2.4
Vainio and Partanen [Ex. 4-312]........... 1.25-1.30
Wald et al. [Ex. 4-315]:
Case-control studies.................. 1.27(1.05-1.53)
Prospective studies................... 1.44(1.20-1.72)
Combined.............................. 1.55(1.19-1.54)
Wells [Ex. 4-319]......................... 2.10(1.30-3.20)
EPA 1992 [Ex. 4-311]...................... \2\1.19
------------------------------------------------------------------------
\1\Numbers in parenthesis indicate published 95 percent confidence
intervals.
\2\Pooled studies.
Table IV-4.--Published Risk Estimates for Heart Disease
------------------------------------------------------------------------
Study Estimates of relative risk
------------------------------------------------------------------------
Steenland [Ex. 4-292]..................... \1\1.51
\2\1.37
Wells [Ex. 4-319]......................... \3\1.32
------------------------------------------------------------------------
\1\Represents risk to nonsmoking men with spousal exposure.
\2\Represents risk to nonsmoking women with spousal exposure.
\3\Women.
Most published risk assessments are based on spousal exposure to
ETS. These studies have examined the lung cancer risk in nonsmoking
housewives, using spousal smoking as a surrogate for the wife's
exposure to ETS. The size of the association between these health
effects and ETS exposure in the workplace is expected to be at least as
large as the association seen between these health effects and ETS
exposure in residential settings or public places. As noted by Meridian
Research in their 1988 report, ``. . . it is the exposure to
environmental tobacco smoke, and not the environment in which that
exposure occurs, that is the important risk factor'' [Ex. 4-221].
Therefore, health effects observed and the risk estimates calculated
from studies of the general population, or of selected subgroups, such
as nonsmoking wives of smoking husbands, are relevant to the working
nonsmoking population.
In developing risk estimates for disease attributable to
occupational exposure, reliance is placed on exposure encountered in
the workplace to the extent possible. However, in the absence of purely
occupational data, information derived in environments other than work
sites is also considered. OSHA believes that there is no physiological
difference related to exposure (or its outcome) regardless of where it
is experienced. This is true regardless of whether the endpoint is lung
cancer, heart disease, or indoor air related acute irritant effects.
The only difference is that the degree of exposure may be greater in
one place than in the other. Available information which uses nicotine
concentration as an index of exposure suggests that the differences in
exposure between office workplaces and residences lie well within the
uncertainties of the determinations and for some workplaces, such as
restaurants and transportation facilities, exposures are significantly
higher than the average exposures found in residences. Thus, risk
estimates based on residential exposures are expected to accurately
reflect occupational risks in most workplaces and possibly
underestimate the risk in some workplaces.
In developing its risk assessment for lung cancer, the EPA reviewed
19 studies which investigated nicotine concentrations in various
environments [Ex. 4-311]. EPA's analysis showed that the range of
average nicotine concentrations in office workplaces is very similar to
that of homes. However, in some workplaces, such as restaurants and
transportation facilities, exposures are significantly higher. It is
true that there are many complicating factors in such determinations
which could affect any final conclusions. For example, it is important
to consider the duration of exposure, the intensity of exposure, the
distance from the sources and other factors as well. However, EPA's
analysis suggests that risk assessments based on home exposures are
relevant to workplaces as well and, in comparison to some workplaces,
may even result in an underestimate of the true occupational risk.
In addition, other studies substantiate the magnitude of workplace
exposures. For example, Emmons et al. [Ex. 4-98] found that the
majority of ETS exposure occurred in the workplace. Study subjects were
selected from workplace settings with a wide range of ETS exposure. The
work sites ranged from those with minimal restriction of smoking and
high levels of exposure to work sites with extensive smoking
restrictions and low exposure. Ninety percent of the subjects worked
outside the home. Eighty-four percent of those who worked outside the
home (75.6% of the total sample) reported being regularly exposed to
smoking in the workplace. While the most highly exposed individuals in
the study were those who had both home and work exposures, it is clear
that workplace exposure constituted a significant component of overall
exposure. Subjects who did not live with smokers reported that the
majority of their exposure was in the workplace (mean=36.1 min/day),
home (mean=1.4 min/day) or in other locations (mean=13.1 min/day).
Subjects who lived with smokers reported receiving slightly more
exposure at home than the workplace, however the difference between
home exposure and workplace exposure was not substantial (work:
mean=29.4 min/day, home: mean=31.2 min/day, other: mean=27.1 min/day).
These results are shown in Table IV-5. The importance of the findings
from this study is twofold. First, it indicates that the workplace is
the primary source of ETS exposure for nonsmokers, who do not live with
smokers. Secondly, it shows that for nonsmokers living with smokers,
even though their household environment becomes their primary source of
exposure, the workplace still contributes a substantial amount of
exposure, comparable to that experienced by the nonsmoker living with
nonsmokers (29.4 min/day v. 36.1 min/day).
Table IV-5.--Exposure to ETS by Location\1\
------------------------------------------------------------------------
Exposure 95 percent confidence
Subject Category (min/day) interval
------------------------------------------------------------------------
Living with a smoker:
Workplace..................... 29.4 (7.01-51.80)
Home.......................... 31.2 (21.60-40.80)
Other......................... 27.1 (15.10-39.10)
Living without a smoker:
Workplace..................... 36.1 (22.70-49.50)
Home.......................... 1.4 (0.05-2.75)
Other......................... 13.1 (8.75-17.40)
------------------------------------------------------------------------
\1\Source: Emmons et al. [Ex. 4-98]
Cummings et al. [Exs. 4-67], Hudgafvel-Pursiainen et al. [Ex. 4-
152], and Marcus et al. [Ex. 4-205] also present results to show
significant workplace exposures to ETS. A re-analysis of the CAPS data
(a detailed description of this study is found in the EXPOSURE section)
shows that the workplace contributes on the average 46 percent to the
total ETS exposure experienced by a nonsmoking worker.
C. Data Sources
As mentioned previously, only diseases that have been reported to
be significantly associated with ETS exposure and for which OSHA has
access to data will be used in calculating health risk due to
occupational exposure to ETS. These will be referred to as the
``diseases of interest'' and include coronary heart disease (excluding
strokes) as defined in the Framingham study and lung cancer.
Ideally, data on the incidence of the diseases of interest in the
U.S. population were needed to estimate the number of cases of disease
in employed nonsmokers. Since nationwide incidence data were not
available for nonsmokers, several survey sources were used to estimate
the mortality rates for heart disease (Framingham Community Study) [Ex.
4-108], and lung cancer (Cancer Prevention Survey conducted by the
American Cancer Society) [Ex. 4-7]. Data on the U.S. workforce were
obtained from the Bureau of Labor Statistics [Ex. 4-39]. Based on the
1993 annual averages, as estimated by the Household Survey, BLS reports
that the U.S. workforce for sectors covered by this standard is
estimated to be 101,631,300 (men: 54.36%, women: 45.64%). Information
on the proportion of employed adults who smoke was obtained from the
National Health Interview Survey and is found in Table IV-7 [Ex. 4-
235]. It is estimated that 74,201,000 adults (73.01% of the U.S. labor
force), employed in sectors covered by this standard, are nonsmokers.
[Editorial note: No Table IV-6 is included in this preamble.]
Table IV-7.--Percent Estimates of Adults Employed in the United States
by Smoking Status\1\
------------------------------------------------------------------------
Smoker Nonsmoker
------------------------------------------------------------------------
Currently employed................................ 26.99 73.01
Unemployed........................................ 40.38 59.62
Not in labor force................................ 21.50 78.50
------------------------------------------------------------------------
\1\National Health Interview Survey [Ex. 235].
In an effort to characterize prevalence of occupational exposure,
OSHA considered several sources. To determine the prevalence of smoking
among U.S. adults during 1991, the National Health Interview Survey-
Health Promotion and Disease Prevention (NHIS-HPDP) supplement
collected self-reported information on smoking exposure at work from a
representative sample of the U.S. civilian, non-institutionalized
population greater than 18 years of age [Ex. 4-51]. In particular,
employed individuals were asked whether, during the past two weeks,
anyone had smoked in their immediate work area. Based on results
adjusted for nonresponse and weighted to reflect national estimates,
18.81 percent of nonsmokers reported exposure to smoke in their
immediate work area as shown in Table IV-8. OSHA believes that 18.81
percent may be an underestimate of frequency of exposure in the
workplace because it is based solely on self-reported information and
the question was not very specific in defining immediate work area.
Table IV-8.--Percent Estimates of Responses to Question 6a in the NHIS
by Smoking Status\1\
------------------------------------------------------------------------
Smoker Nonsmoker
------------------------------------------------------------------------
Yes............................................... 37.58 18.81
No................................................ 60.81 79.79
Unknown........................................... 1.61 1.39
------------------------------------------------------------------------
\1\Question 6a was: ``During the past 2 weeks, has anyone smoked in your
immediate work area?''
Another source considered by OSHA for defining nonsmoker ETS
exposure in the workplace was the work published by Cummings et al.
[Ex. 4-67]. A recent re-analysis of the data file showed that among the
nonsmoking, currently employed subjects, 48.67 percent (165 out of 339)
reported exposure to ETS at work and not at home (Table IV-9) [Ex. 4-
69]. Based on the data sources mentioned above, OSHA assumes that the
percent of nonsmoking workers who are potentially exposed to ETS at
their worksite ranges between 18.81 and 48.67.
Table IV-9.--Prevalence of ETS Exposure for Nonsmoking Workers\1\
------------------------------------------------------------------------
Subject category Count Percent
------------------------------------------------------------------------
Exposed at work and home.......................... 99 29.22
Exposed at home, not at work...................... 23 6.78
Exposed at work, not at home...................... 165 48.67
Not exposed at work or home....................... 52 15.34
------------------------------------------------------------------------
\1\Data source: Cummings reanalysis [Ex. 4-69].
D. OSHA's Estimates of Risk--Environmental Tobacco Smoke Exposure
The incidence of disease due to occupational exposure in nonsmokers
was estimated using the following methodology: The expected number of
cases, Ne, in nonsmoking workers who are occupationally exposed to
ETS is expressed by:
Ne=Nd - N * Iu = N * (Ip - Iu)
where:
Ne is the cases in nonsmoking exposed workers attributable to ETS
per year
Nd is the estimated number of cases per year in nonsmoking workers
N is the number of nonsmoking workers in the U.S.
Iu is the incidence rate of disease among the unexposed workers
Ip is the U.S. population incidence rate for nonsmokers
The number of nonsmoking workers (N) was estimated by multiplying
the percent of currently employed adults who report to be nonsmokers by
the number of adult, employed, civilian noninstitutional population, as
reported by BLS.
The number of nonsmoking workers with disease per year (Nd)
was estimated as Nd=N * Ip. The U.S. population incidence
rate of lung cancer for nonsmoking women is reported to be 0.121 per
one thousand nonsmoking women. The lung cancer incidence for nonsmoking
males is estimated to be higher. For the purpose of this risk
assessment, OSHA used 0.121 as the population incidence rate of lung
cancer for nonsmokers. This will most likely result in an underestimate
of the true risk for male workers. The average annual incidence rate
for death from coronary heart disease excluding strokes for nonsmokers
age 35 to 64 is estimated to be 4 per one thousand men and 2 per one
thousand women, as reported by the Framingham study. This results in an
overall weighted average of 3 deaths per one thousand individuals.
The incidence rate of disease (Iu) among the unexposed workers
is estimated using the relationship:
Iu = Ip / [RR * pe + (1-pe)]
where:
RR is the observed relative risk of disease for nonsmokers exposed to
ETS
pe is the proportion of nonsmoking workers exposed to ETS while at
work.
OSHA used 1.34 as an observed estimate of relative risk (RR) for
lung cancer among nonsmokers with occupational exposure as reported by
Fontham et al. [Ex. 4-106]. Estimates of observed relative risk for
heart disease in nonsmokers, as reported by Helsing et al. (1.24 for
females and 1.31 for males), were used in calculating an overall
adjusted relative risk estimate of 1.28 [Ex. 4-139]. The adjusted
relative risk was a weighted average of the reported relative risks
using the gender composition of the U.S. workforce as weights
((1.24*45.64 + 1.31*54.36/100) = 1.28). The proportion of nonsmoking
workers exposed to ETS while at work (pe) was assumed to range
from 18.81 to 48.67 as stated previously.
OSHA chose to rely on the Fontham and Helsing studies for estimates
of the observed relative risks for several reasons. Both studies were
conducted in the U.S. Both are large, population-based studies whose
results can be generalized to the general public. Both studies, by
design, controlled for misclassification to a large degree. The Helsing
study, which was done in the 60's--a time when smoking was more
acceptable than more recently, and being a prospective cohort study,
was less prone to misclassification and other sources of bias. The
Fontham study used multiple sources to ascertain nonsmoking status and
validate subject response. Study subjects were questioned twice; the
self-reported nonsmoking status was corroborated by urinary cotinine
measurements; and medical records were cross-referenced with the
physician's assessment. In addition, in the Fontham study, information
on occupational exposure was collected and an estimate of lung cancer
risk attributable to the workplace exposure was ascertained.
The annual risk of disease attributable to occupational exposure to
ETS was estimated by dividing the expected number of cases (Ne) by
the number of nonsmoking workers in the U.S. population. Table IV-10
presents the annual risk attributable to occupational exposure to ETS
per 1,000 exposed employees. Because section (6)(b)(5) of the OSH Act
states that no employee shall suffer ``material impairment of health or
functional capacity even if such an employee has regular exposure to
the hazard dealt with * * * for the period of his working life'', OSHA
has converted the attributable annual risk into an attributable
lifetime risk on the assumption that a worker is employed in his or her
occupation for 45 years. Lifetime estimates of risk attributable to
occupational ETS are presented in Table IV-10. Information contained in
Table IV-10 indicates that for every 1,000 workers exposed to ETS,
approximately 1 will most likely develop lung cancer and 7 to 16 will
develop heart disease if they are exposed to ETS at their workplace in
the course of a 45-year working lifetime. The formula used to calculate
lifetime risk estimates the probability of at least one occurrence of
disease in 45 years of continuous exposure and assumes independence of
events from year to year. It also assumes that the worker's exposure
profile and working conditions that may affect the level and intensity
of exposure remain constant throughout a working lifetime.
Table IV-10.--Estimates of Risk For Nonsmoking Workers Exposed to ETS at
the Workplace1,2
------------------------------------------------------------------------
Lifetime
Annual occupational
risk\2\ risk\3\
------------------------------------------------------------------------
Lung cancer.................................. 0.01-0.02 0.4-1
Heart disease................................ 0.15-0.36 7-16
------------------------------------------------------------------------
\1\Risks are expressed as number of cases per 1,000 workers at risk.
\2\The annual risk for nonsmoking workers is estimated assuming the
proportion of nonsmoking workers exposed to ETS at the workplace
ranges from 18.81 to 48.67.
\3\Assumes 45 years of occupational exposure and is calculated as 1-(1-
p)\45\, where p is the annual risk.
E. OSHA's Risk Estimates--Indoor Air Quality
Adverse health effects associated with poor IAQ are described as
Building-Related Illness (BRI) and Sick Building Syndrome (SBS). SBS
related conditions are not easily traced to a single specific
substance, but are perceived as resulting from some unidentified
contaminant or combination of contaminants. Symptoms are relieved when
the employee leaves the building and may be reduced by modifying the
ventilation system.
Research in Britain [Ex. 4-44], Denmark [Ex. 4-284] and the United
States [Ex. 3-745] indicates that about 20% of all office workers are
afflicted with such symptoms. If the 20% level were to be considered as
``background'', a simple approach would be to determine that any
building, more than 20% of whose occupants report the symptoms, would
be considered to be ``sick''. However, the question then arises as to
how much greater than 20% would the incidence have to be to be
considered excess and how would one address such issues as statistical
significance for any one building. Furthermore, the definition used in
assessing symptom occurrence can cause substantial variations in
estimating symptom prevalence, even in the same building. The problem
with many investigations of ``sick'' buildings is that rarely have
``non-sick'' or control buildings been used to determine background
prevalence of the symptoms. Until now, it appears that limited research
has been done to address the issue of background levels of symptoms.
OSHA seeks input on data sources to address expected background levels
of SBS related conditions.
Mendell and Smith [Ex. 4-218] examined symptom reports compiled in
a number of individual studies for a number of buildings which had
different types of ventilation. On the basis of the information
gathered in the individual studies, Mendell and Smith compared the
prevalence of sick building symptoms in buildings with five types of
ventilation: natural only; fans only; air conditioned with no
humidification; air conditioned with steam humidification; and air
conditioned with water-based humidification. Overall, they found the
prevalence of work-related headache, lethargy, upper respiratory/mucous
membrane, lower respiratory and skin symptoms significantly increased
in buildings with any type of air conditioning as compared to buildings
with no air conditioning. Thus, according to this analysis, a basic
problem with SBS appears to reside in the air conditioning system or,
in some building aspect associated with the presence of air-
conditioning.
Building-related illness (BRI) describes those specific medical
conditions of known etiology which can often be documented by physical
signs and laboratory findings. Symptoms may or may not disappear when
the employee leaves the building. Currently, OSHA does not have any
data on BRI related symptoms to conduct a quantitative risk assessment.
The number of cases of illness in the United States related to poor
indoor air quality has not yet been quantified; however OSHA has made
an attempt to develop a preliminary risk estimate of SBS using a
similar methodology as was done for ETS. The National Health Interview
Survey was the primary data source for U.S. population frequency rates
for acute upper respiratory symptoms other than the common cold,
influenza, acute bronchitis, and pneumonia and frequency rates on
severe headaches other than migraines. For this preliminary risk
assessment, OSHA used the reported frequency rates as representative of
population incidence rates for upper respiratory conditions and severe
headaches. OSHA seeks comment on the use of frequency data in place of
incidence data.
Observed relative risks for comparable conditions were estimated by
Mendell [Ex. 4-219]. Mendell's data source was the California Healthy
Building Study. This study surveyed a representative sample of 12
public office buildings in Northern California to ascertain the
occurrence of work-related symptoms associated with air-conditioned
office buildings. All buildings were either smokefree or had separately
ventilated designated smoking areas. The sample included 6 buildings
with air- conditioning systems, 3 buildings with mechanical ventilation
and no air-conditioning, and 3 buildings with natural ventilation. The
study included 880 workers. Mendell estimated relative risks for
several building related symptoms and a subset of these estimates are
shown in Table IV-11. In an effort to define comparable symptoms
between the reported national statistics from NHIS and Mendell's study
and for computational ease OSHA grouped ``runny nose'', ``stuffy
nose'', ``dry/irritated throat'', and ``dry/irritated/itching eyes'' as
upper respiratory/mucous membrane symptoms. Mendell reported relative
risks for upper respiratory conditions and frequent headaches in air-
conditioned buildings as compared to naturally ventilated buildings.
The relative risk for frequent headaches was reported to be 1.5. For
upper respiratory conditions, such as ``stuffy nose'', ``runny nose'',
etc., the relative risks ranged from 1.4 to 1.8. OSHA used 1.4 as an
observed relative risk for upper respiratory conditions.
CDC reports in the ``Current Estimates from the National Health
Interview Survey, 1992'' that the annual rate for severe headaches,
requiring medical attention or activity restriction, is at least 5 per
thousand and the rate for upper respiratory conditions is at least 9
per thousand. In addition, it is estimated that the proportion of
office buildings in the U.S. with air-conditioning is 70 percent (see
Preliminary Regulatory Impact Analysis section). Using the above
information and the same methodology as described in section IV-D, OSHA
estimated that the lifetime excess burden for severe headaches
experienced in air-conditioned office buildings is 57 per one thousand
exposed employees and the lifetime risk for acute upper respiratory
conditions is 85 per one thousand exposed employees. OSHA's risk
estimates for indoor air are shown in Table IV-12. OSHA used data
derived from a study of air-conditioned office buildings to make an
assessment of the occupational risk in all air-conditioned buildings.
Furthermore, OSHA made an implicit assumption that an increase in work-
related headaches associated with an air-conditioned office environment
occurs in the same proportion as headaches which can be severe enough
to affect work activity. OSHA seeks comment on the applicability of the
Mendell study for estimating occupational risk in air-conditioned
buildings due to poor indoor air quality. In addition, OSHA seeks
comment on its methodology of developing annual and lifetime risk
estimates attributable to occupational exposures.
Table IV-11.--California Healthy Building Study Comparing Buildings With
Natural Ventilation to Buildings With Air-Conditioning\1\
------------------------------------------------------------------------
Relative Confidence
Health outcome risk interval
------------------------------------------------------------------------
Upper respiratory symptoms:
Runny nose..................................... 1.5 (0.9-2.5)
Stuffy nose.................................... 1.8 (1.2-3.7)
Dry/irritated throat........................... 1.6 (0.9-2.7)
Dry/irritated/itchy eyes....................... 1.4 (0.9-2.2)
Frequent headaches............................... 1.5 (0.9-0.3)
------------------------------------------------------------------------
\1\Study subjects were asked whether the symptoms were occurring often
or always at work and improving when away from work.
Table IV-12.--OSHA's Estimates of Risk for Workers in Air-Conditioned
Buildings\1\
------------------------------------------------------------------------
Annual Lifetime
risk\2\ occupational
risk\3\
------------------------------------------------------------------------
Severe headaches\4\.............................. 1.296 57
Upper respiratory symptoms\5\.................... 1.969 85
------------------------------------------------------------------------
\1\Risks are expressed as number of cases per 1,000 workers at risk.
\2\The annual risk is estimated assuming that the prevalence of air-
conditioned office buildings in the U.S. is 70 percent.
\3\Assumes 45 years of occupational exposure and is calculated as 1-(1-
p45, where p is the annual risk.
\4\Defined as headaches that either require medical attention or
restrict activity.
\5\Defined as runny nose, stuffy nose, dry/irritated throat and dry/
irritated/itchy eyes and being severe enough to either require medical
attention or restrict activity.
F. Pharmacokinetic Modeling of ETS Exposure
In developing a final rule, OSHA would like to consider the use of
a physiologically based pharmacokinetic (PBPK) model in an effort to
develop a clear and complete picture of factors that may affect
environmental exposure measurements, internal dose estimates and
ultimately estimates of expected risk attributed to ETS exposure at the
workplace. OSHA is seeking comment on appropriate methodology,
available data, etc. The following discussion offers an explanation of
OSHA's approach to this issue and an opportunity for the Agency to
solicit comment on specific points of concern as they relate to the use
of pharmacokinetics in estimating occupational risk from exposure to
ETS.
Estimating the risk from exposure to ETS requires the use of some
measure of the extent of exposure. Possible measures, or metrics, can
range from categorical ranking based on survey responses to direct
measurement of ETS-related chemicals in the body fluids of exposed
individuals. In general, the use of an internal measure of individual
exposure would be preferred over measurements of environmental
contamination, such as airborne chemical or particulate concentrations.
In particular, considerable attention has been given in the scientific
literature to the possible use of cotinine concentrations in body
fluids as a biomarker of ETS exposure [Exs. 4-24, 4-146, 4-165, 4-263,
4-316]. However, obtaining a dependable estimate of exposure from
measurements of a chemical's concentration in body fluids requires a
quantitative understanding of the chemical's pharmacokinetics; its
uptake, distribution, metabolism, and excretion. Following is a review
of the evidence concerning the suitability of cotinine as an internal
biomarker for ETS exposure.
1. Considerations for Selection of a Biomarker for ETS
A biomarker should, to the greatest extent possible, accurately
represent the individual's exposure to the substance of concern and
have relevance to a specific endpoint. In the case of ETS, there are
several relevant endpoints, with principal attention being given to
heart disease and lung cancer. Each different endpoint may be mediated
by a different subset of the components of ETS, and therefore the
appropriate biomarker(s) for each endpoint could be different.
2. Cardiovascular Effects
Cardiovascular effects resulting from exposure to ETS have been
associated with carbon monoxide (CO), nicotine, and more recently with
polycyclic aromatic hydrocarbons (PAHs) [Ex. 4-123]. Each of these is
associated with a different fraction of ETS; CO is a gas phase
constituent, nicotine is a low volatility vapor, and PAHs are absorbed
on particulates. Because of the significant differences in physical
fate and transport, a strategy for the use of biomarkers for
cardiovascular effects of ETS would ideally make use of separate
markers for CO, nicotine, and PAHs.
The most common internal measure of CO exposure is blood
carboxyhemoglobin (HbCO). Blood HbCO provides a useful measure of
exposure to CO, and can be related to the cardiovascular effects of CO.
A way to determine the occupational component of one's total CO
exposure is to measure workplace CO levels and predict blood HbCO with
a physiologically based pharmacokinetic model for CO [Ex. 4-11]. A
difficulty associated with the use of CO or HbCO as a biomarker for ETS
effects is the presence of other sources of CO in the workplace.
Nicotine can be measured directly in body fluids and the
circulating concentration can be related to physiological effects, such
as heart rate [Ex. 4-26]. Alternatively, measurements of nicotine in
air or cotinine in body fluids can be measured, and the circulating
concentration of nicotine can be inferred using a pharmacokinetic
model. The use of a pharmacokinetic model to relate inhaled nicotine to
circulating nicotine and cotinine levels is the main focus of this
section.
PAHs are inhaled in the form of particulates on which they are
adsorbed. Developing an appropriate biomarker for ETS-associated PAHs
is complicated by the presence of PAHs on particulates not associated
with ETS, and by the low, and variable, composition of PAHs adsorbed to
particulate matter. One candidate material which has been suggested as
an environmental marker for ETS-associated particulates is solanesol, a
non-volatile tobacco constituent. However, the pharmacokinetic
information necessary for use of solanesol as an internal biomarker is
not currently available.
The use of these three different biomarkers (CO, PAHs, and
solanesol) does not appear to be practical. It appears that the most
effective strategy currently achievable would be to rely on nicotine
(or cotinine) measurement as a specific marker of ETS exposure as well
as a direct measure of nicotine exposure.
3. Carcinogenicity
The mechanism of carcinogenicity from exposure to ETS is not known,
but it has been established that ETS includes a number of chemicals
which have been identified as carcinogens (see Tables II-2, III-6, and
III-7), although most of the identified carcinogenic components of ETS
are not unique to ETS. Therefore, direct measurement of the
carcinogenic components or related biomarkers in biological fluids
would not provide a unique measure of exposure from ETS. The
potentially carcinogenic components of ETS include highly volatile
chemicals such as formaldehyde and benzene, lower volatility chemicals
such as the nitrosamines, and non-volatile chemicals such as PAHs and
metal compounds, which are bound to particulates. Given the current
lack of information on the mechanism of carcinogenicity of ETS it is
impossible to identify which components of ETS should be targeted for
exposure estimation. The most prudent choice for a biomarker in this
case would be one which provides the most general representation of all
the components of ETS, and which is itself unique to ETS. In an
experimental study of potential ETS-unique environmental markers of
exposure, only nicotine was found to represent both the gas phase and
particulate phase organic constituent of ETS [Ex. 4-97]. Several
studies have shown a strong correlation between measurements of
nicotine in the air and the mutagenicity of ETS [Exs. 4-198, 4-215]. In
these studies, the relationship of nicotine to mutagenicity was as good
as or better than the relationship of RSP to mutagenicity (RSP is
assumed to be the major contributor of the carcinogenic effects of
ETS). Therefore, since measurements of nicotine in the air correlate
better than measurements of RSP to mutagenicity of ETS, and there is a
positive correlation between short-term mutagenicity tests and
carcinogenicity, the use of nicotine as an exposure marker for the
carcinogenic effects of ETS appears to be justified.
4. Evaluation of Cotinine as a Biomarker for ETS
The purpose of this section is to discuss the use of cotinine, a
metabolite of nicotine, as an internal biomarker for inhalation
exposure to nicotine, and, as such, its usefulness as a metric for the
health effects of ETS. Cotinine is preferred over nicotine as an
internal biomarker because of its slower clearance from the body [Ex.
4-71].
There is a strong correlation between nicotine intake and plasma
cotinine levels [Ex. 4-115]. There is also a strong correlation between
cotinine measured in body fluids and ETS exposure. In a controlled
study, urinary cotinine was found to be a reliable marker for long-term
ETS exposure, and plasma and salivary cotinine were found to be good
indicators of short- as well as long-term exposure [Ex. 4-73]. Several
studies have also demonstrated a positive relationship between self-
reported exposure to ETS and cotinine in serum [Exs. 4-166, 4-250, 4-
301], saliva [Ex. 4-166], and urine [Exs. 4-166, 4-211, 4-316]. In
general, the currently available data support the assumption that
nicotine and cotinine kinetics parameters for smokers can be
extrapolated to nonsmokers for estimating exposures to ETS in
nonsmokers [Ex. 4-24]. Studies have also demonstrated that salivary
levels of cotinine are directly proportional to plasma levels [Ex. 4-
73], and that urinary excretion of cotinine is linearly related to
plasma levels [Ex. 4-82]. Thus all three biological fluids provide a
reasonable metric for nicotine intake, and thus can serve as biomarkers
of ETS exposure in nonsmokers.
There are two potential difficulties associated with the use of
cotinine as a biomarker for ETS. The first is the presence of nicotine
in the diet. Several foods, including tea, tomatoes, and potatoes, have
been shown to contain nicotine in measurable quantities [Exs. 4-49, 4-
81, 4-281]. However, a study of 3,383 nonsmokers was unable to
substantiate an effect of tea drinking on serum cotinine levels for
self-reported daily tea consumption [Ex. 4-301]. The same study did
find a strong correlation between self-reported ETS exposure and serum
cotinine level.
OSHA seeks comment and data on whether dietary intake of nicotine
should be considered a significant factor in modelling nicotine
metabolism for assessing risk due to ETS exposure.
The second issue associated with the use of cotinine as a biomarker
is the possibility that there is a longer half-life for the elimination
of cotinine at very low biological concentrations, associated with the
slow release of nicotine from binding sites [Exs. 4-28, 4-24, 4-167, 4-
254]. This longer half-life at very low concentrations could have the
effect of overestimating exposure to ETS in the lowest exposed
population. At this time there is not sufficient evidence to quantify
the potential magnitude of this effect, but it is likely to be small.
OSHA seeks comment on this issue.
5. Description of Pharmacokinetic Models for Nicotine and Cotinine
For many purposes, an essentially first order process such as the
kinetics of cotinine can be effectively modeled with a simple
compartmental kinetic analysis [Exs. 4-27, 4-24, 4-73, 4-82]. The
compartmental approach has been used to relate steady-state urinary
cotinine levels to atmospheric nicotine concentrations [Ex. 4-263]. For
investigating some of the concerns associated with the use of cotinine
as a biomarker, however, a physiologically based pharmacokinetic (PBPK)
description would be preferred. The advantage of the PBPK approach
stems from its biologically motivated structure, which permits the
direct incorporation of biochemical data and the biologically
constrained comparison of model predictions with experimental
timecourses to investigate such issues as dose-rate effects, exposure-
route differences, pharmacodynamic processes, and other potential
nonlinearities [Ex. 4-57]. PBPK models of nicotine and cotinine have
been described for both rats [Exs. 4-112, 4-255] and humans [Exs. 4-
254, 4-270].
A physiological model of cotinine disposition [Ex. 4-112] was
developed to analyze intravenous infusion of nicotine and cotinine and
bolus dosing of cotinine in rats. In general, the observed cotinine
time profiles in blood and tissues were consistent with linear
kinetics, but the distribution of cotinine into all tissues appeared to
be roughly three-fold greater following infusion of nicotine than
following infusion of cotinine, and the clearance of cotinine following
bolus and infusion dosing was significantly different.
A more recent rat model [Ex. 255] featured a physiologically based
description of nicotine kinetics and a compartmental description of
cotinine. This model provided a successful description of the plasma
kinetics of both nicotine and cotinine for intraarterial or intravenous
bolus dosing of nicotine. The timecourse of nicotine in most tissues
was also consistent with first order kinetics; however, it was
necessary to include a description of saturable nicotine binding in the
brain, heart, and lung to adequately reproduce nicotine concentration
profiles in these tissues. This rat model has also been scaled for use
in predicting mouse and human pharmacokinetics [Ex. 4-254]. The human
model has recently been expanded to include a physiological description
of cotinine as well as a forearm compartment, and is now able to
describe nicotine and cotinine kinetics following intravenous infusion
of nicotine in humans [Ex. 4-266]. Another human model [Ex. 4-270] has
also been developed which includes physiological descriptions of both
nicotine and cotinine. This model, which assumes linear kinetics,
predicts results which agree with published data on the kinetics of
nicotine and cotinine in blood following nicotine infusion as well as
cotinine in the blood following the infusion of cotinine.
6. Application of Pharmacokinetic Modeling for ETS Exposure Estimation
Both of the human models described above possess a reasonable
biologically based structure, and either model would provide a useful
starting point for the development of a PBPK model which could be of
use in examining the relationship between cotinine concentrations in
body fluids and inhaled nicotine. However, neither of the models
currently possesses all of the features which would be necessary for
such an analysis. The most useful application of PBPK modeling would
appear to be to support an analysis of four issues related to the use
of cotinine as a biomarker of ETS exposure: (1) Estimation of the
contribution of dietary intake of nicotine to cotinine levels in the
plasma, saliva and urine of nonsmokers; (2) Estimation of a plausible
upper bound for cotinine concentrations in plasma, saliva and urine
associated with ETS exposure (to identify individuals wrongfully
identifying themselves as nonsmokers). This can be viewed as a way to
validate misclassification results derived from surveys; (3) Evaluation
of the potential impact of high affinity, low capacity binding of
nicotine and cotinine in nonsmokers with low exposure to ETS; and (4)
Evaluation of the potential impact of pharmacokinetic uncertainty and
variability on the use of cotinine concentrations in plasma, saliva or
urine to infer an individual's ETS exposure. The necessary features for
accomplishing these analyses include both inhalation and oral routes of
nicotine exposure, a salivary compartment, and a description of
nicotine binding in the brain, heart and lung.
In evaluating the use of cotinine as a biomarker of ETS exposure,
two kinds of uncertainty must be considered. The first kind of
uncertainty embraces those factors which could tend to bias a risk
estimate. Two such factors are dietary intake of nicotine and nicotine
binding. In both of these cases, the impact of ignoring the effect, if
it were significant, would be to overestimate exposure (and therefore
risk) for the least exposed individuals. The second kind of uncertainty
includes those factors which tend to broaden the confidence interval
for the risk estimate. The most significant factors in this category
are uncertainty in the fraction of nicotine converted to free cotinine,
and the rates of metabolic and urinary clearance of nicotine and
cotinine. An example of such uncertainty is results reported for half-
lives of cotinine in nonsmokers [Ex. 4-24, 4-73, 4-82, 4-184, 4-186],
showing a mean of 16.2 hours, with a coefficient of variation of 0.22.
7. Analysis of Uncertainty
It is useful in this evaluation to distinguish uncertainty from
variability. As it relates to the issue of using pharmacokinetic
modeling in risk assessment, uncertainty can be defined as the possible
error in estimating the ``true'' value of a parameter for a
representative (``average'') individual. Variability, on the other
hand, represents differences from individual to individual.
For the purpose of evaluating the usefulness of pharmacokinetic
modeling for estimating exposure, the uncertainty and variability in
the various parameters for the pharmacokinetic models can be grouped
into four classes: the physiological parameters (volumes and flows),
the tissue distribution parameters (partitioning and binding), and the
kinetic parameters (absorption, metabolism, and clearance).
(a) Physiological Parameters. The physiological parameters include
(1) the body weight and the weights of the individual organs or tissue
groups, (2) the total blood flow and flows to each organ or tissue
group, and (3) the alveolar ventilation rate. These quantities have
been reasonably well established for the human [Exs. 4-155, 4-309] and
the chief effort associated with pharmacokinetic model parameterization
in the human is the determination of the necessary level of detail for
the physiological description, grouping of the tissues not meriting a
separate description into pharmacokinetically similar groups, and the
association of the proper volume and flow data with the selected
groupings. Existing models for nicotine and cotinine contain a fairly
detailed physiological structure and differ only slightly in their
assignment of tissues. The model of Plowchalk and deBethizy [Ex. 4-254]
includes separate compartments for the brain, heart, and skin. The
first two of these tissues are lumped into a ``vessel-rich'' tissue
compartment in the model of Robinson et al. [Ex. 4-270], and the skin
is lumped in with the muscle. Conversely, the gastrointestinal tract is
given a separate compartment in the Robinson model but is lumped into a
``slowly perfused'' tissue compartment in the Plowchalk model. These
differences mainly reflect the different interests of the modeling
groups in terms of target organs and routes of exposure. The Robinson
model contains a venous infusion compartment to accommodate the mixing
time for arterial administration. The published Plowchalk model does
not include this feature, but a forearm compartment has since been
added to provide a similar function [Ex. 4-83]. Neither model appears
to contain an explicit description of inhalation or oral exposure, but
the necessary equations could easily be added to the existing
physiological structures. A salivary fluid compartment could also be
added to either model if desired. Experience with other chemicals has
shown that uncertainty in the physiological parameters generally has
much less impact on overall model uncertainty because they are known
relatively well and are not as influential on model behavior as the
distribution and kinetic parameters [Ex. 4-56].
(b) Distributional Parameters. In both of the published human
models, the tissue partitioning was initially estimated on the basis of
steady-state tissue/blood concentration ratios measured in animals. The
partitioning parameters in the Robinson model were then iteratively
adjusted to fit other timecourse data. The resulting partition
coefficients in the two models differ by a factor from two to five in
corresponding tissues. The partitioning data for cotinine, determined
by Gabrelsson and Bondesson [Ex. 4-112], show a similar level of
uncertainty; partitions for cotinine following infusion of nicotine
were two- to five-fold higher than the same partitions following
infusion of cotinine. The lack of reproducibility of these data
represents a deficiency in the development of PBPK modeling for these
chemicals. Fortunately, the partition coefficients tend to be less
important than the kinetic parameters in terms of overall model
performance. To a large extent, as long as the volume of distribution
associated with the physiological structure and partition coefficients
is in agreement with the apparent pharmacokinetic volume of
distribution for each chemical, the model will perform adequately in
terms of timecourses in blood and urine. This was evidenced by the
ability of the Robinson model to reproduce published nicotine and
cotinine pharmacokinetic data [Ex. 4-270]. A potentially more
significant uncertainty associated with distribution is the possibility
of pharmacokinetically significant tissue binding of nicotine.
Satisfactory description of the timecourse of nicotine in the brain,
lung, and heart of the rat required the inclusion of binding in these
tissues [Ex. 4-255]. Clearly, the relatively low capacity, high
affinity binding associated with nicotine is unlikely to effect total
systemic clearance except at very low concentrations. However, the
existence of nonlinear pharmacokinetics at low concentrations could
lead to a miscalculation of exposure for the least exposed individuals.
It has been suggested that there is a longer clearance half-life for
nicotine, and therefore cotinine, associated with low circulating
concentrations, and that this longer half-life is due to the slower
release of nicotine bound to tissues [Exs. 4-28, 4-24, 4-167]. To date,
no careful pharmacokinetic investigation of this possibility has been
performed in the human model, and adequate nicotine-specific tissue
binding information does not appear to have been collected except
perhaps in the brain.
(c) Kinetic Parameters. By far the most significant parameters in
the models are those describing the absorption, metabolism, and
clearance of nicotine and cotinine. The Robinson model uses reported
human hepatic and renal clearance values for nicotine and cotinine. The
sensitivity of this model to these input parameters was investigated by
varying them within the range of reported clearance values from
infusion studies in humans. The resulting model predictions for post-
infusion blood levels, urinary output, and the elimination half-lives
of both nicotine and cotinine were found to be well within the ranges
of those observed in human studies. Thus the model structure does not
produce an exaggerated response to variation of the input parameters,
and reflects the natural interaction between measures of clearance,
volume of distribution, and rates of elimination. In the case of the
physiological parameters, variability dominates over uncertainty, while
for the distributional parameters, uncertainty dominates. In the case
of the kinetic parameters describing clearance, it appears that
variability again dominates. For example, the mean values for the
terminal half-life of cotinine reported in different studies range from
12 to 21 hours in non-smokers [Exs. 4-24, 4-73, 4-82, 4-184, 4-186].
The coefficient of variation in these same studies, a measure of
interindividual variability, ranges from 17-22%, and the coefficient of
variation for the entire collection of reported individual values is
similar: 22% (N=35, mean=16.2). A review of the published data on
infusion of nicotine and cotinine in humans [Ex. 4-270] found a 3-fold
variation in reported half-lives for cotinine. For comparison, the
variation in the volume of distribution for cotinine was 5-fold, while
for the half-life and volume of distribution of nicotine, the variation
was 8-fold and 6-fold, respectively. An even greater level of
variability can be expected for the kinetic parameters for the renal
clearance of nicotine and cotinine.
OSHA considers the use of pharmacokinetics and specifically PBPK
models an important tool in characterizing and quantifying internal
dose for evaluation potential exposures and seeks comment on the
applicability of this approach in ascertaining the relationship between
adverse health effects and exposure to ETS.
V. Significance of Risk
Before the Secretary can promulgate any permanent health or safety
standard, he must find that a significant risk of harm is present in
the workplace and that the new standard is reasonably necessary to
reduce or eliminate that risk. Industrial Union Department, AFL-CIO v.
American Petroleum Institute, 444 U. S. 607, 639-642 (1980) (Benzene).
In the Benzene case, the Supreme Court held that section 3(8) of the
Act, which defines a ``occupational safety and health standard'' as a
``requirement reasonably necessary or appropriate'' to promote safety
or health requires that, before promulgating a standard, the Secretary
must find, ``on the basis of substantial evidence, that it is at least
more likely than not that long-term exposure to [the hazard without new
regulation] presents a significant risk of material health
impairment.'' 444 U. S. at 653.
In the Benzene decision, the Supreme Court indicated when a
reasonable person might consider the risk significant and take steps to
decrease it. The Court stated:
It is the Agency's responsibility to determine in the first
instance what it considers to be a ``significant'' risk. Some risks
are plainly acceptable and others are plainly unacceptable. If, for
example, the odds are one in a billion that a person will die from
cancer by taking a drink of chlorinated water, the risk clearly
could not be considered significant. On the other hand, if the odds
are one in a thousand that regular inhalation of gasoline vapors
that are 2% benzene will be fatal, a reasonable person might well
consider the risk significant and take the appropriate steps to
decrease or eliminate it. (IUD v. API, 448 U. S. at 655).
A. Environmental Tobacco Smoke
Two of the adverse health effects associated with exposure to ETS
are lung cancer and heart disease (coronary heart disease, excluding
strokes). Clinically, lung cancer is almost always fatal. However,
heart disease runs the gamut from severe to disabling to fatal. Both of
these diseases then constitute the type of ``material impairment of
health or functional capacity'' which the Act seeks to reduce or
eliminate. Therefore a standard aimed at reducing the incidence of
these impairments is an appropriate exercise of the Secretary's
regulatory authority.
In the case before us the Agency estimates that there will be
approximately between 144 and 722 cases of lung cancer per year among
nonsmoking American workers exposed to ETS in the workplace. When
considered over a working lifetime, this translates into an excess lung
cancer rate in the workplace of one per thousand. As noted above, the
Benzene court clearly indicated that a risk of one in a thousand could
be considered significant and that the Agency would be justified in
prescribing reasonable efforts to reduce such a risk.
Therefore, the risk from lung cancer associated with worker
exposure to ETS in the workplace meets the Benzene court's
characterization of what could be considered significant.
In addition, in evaluating the significance of the risk posed by
any particular workplace hazard, the Secretary is entitled to take into
consideration not only the rate of risk but the total number of workers
exposed to such risk and the absolute magnitude of effects. In this
case, evidence in the record shows that approximately between 144 and
722 lung cancer deaths per year are attributable to ETS and that there
are presently over 74 million nonsmoking American workers exposed to
ETS in their places of employment. On the basis of these data, it would
also be reasonable to conclude that Agency action is warranted to
reduce this widespread and significant risk, although the Agency would
reach this conclusion even without the great magnitude of effects.
As noted above, cancer is not the only serious adverse health
effect associated with exposure to ETS. Preliminary estimates indicate
that the risk of mortality from heart disease due to ETS exposure is
even greater than that of cancer. The Agency estimates that there will
be between 2,094 and 13,000 deaths from heart disease per year among
nonsmoking American workers exposed to ETS in the workplace. When
considered over a working lifetime, this translates into an excess
death rate of approximately between 7 and 16 cases of heart disease per
thousand attributed to workplace exposure to ETS. Clearly, this risk is
significant in itself and combined with the lung cancer risk, the
significance of risk is very great.
The proposal seeks to protect nonsmoking employees from the hazards
of exposure to ETS in the workplace. It does this by prescribing the
conditions under which employees would be allowed to smoke in the
workplace, that is, only in separately enclosed designated areas which
are separately ventilated. No employee can be required to work in an
area where there will be contamination from ETS. This in OSHA's view
reduces significant risk to only a small percentage of the current
risk. To the extent that there are failures of enforcement of the
smoking limitation and of the ventilation system, the risk will not be
totally eliminated. Since there is no definition of, nor an established
method for quantifying, exposure, it is not possible to determine a
``dose limit'' that would eliminate significant risk. Even if that were
possible, it is not clear it would be the correct policy approach.
29 CFR Part 1990--Identification, Classification and Regulation of
Potential Occupational Carcinogens sets forth certain procedures for
regulating occupational carcinogens. Those procedures may not allow for
the level of public input and policy review that is appropriate for
this rulemaking, involving many different types of health effects and a
broad range of employers and workers. Accordingly, the Assistant
Secretary finds pursuant to 29 CFR Section 1911.4 that ``in order to
provide greater procedural protections to interested persons or for
other good cause consistent with the applicable laws'' ``it is found
necessary or appropriate'' to adopt different procedures here.
B. Indoor Air Quality
Poor indoor air quality creates a variety of material impairments
of health, two aspects of which are Building-Related Illness and Sick
Building Syndrome.
One of the most severe health effects associated with Building-
Related Illness is legionellosis, a disease associated with microbial
contamination of water sources which is commonly found in the water
present in heating and cooling systems of buildings. Legionnaire's
disease, caused by the Legionella organism, results in pneumonia which
is fatal in approximately 20% of the cases. Even when not fatal, it is
usually very severe, requiring substantial treatment or
hospitalization. As many as 5% of those exposed to Legionella will get
sick1. Legionnaire's disease and other illnesses associated with
microbial contamination due to poor indoor air quality are serious
health effects that constitute material impairment. Compliance with the
indoor air quality provisions set forth in the proposal will
substantially reduce these illnesses.
---------------------------------------------------------------------------
\1\Raw figures from 1992 show approximately 1300 cases of
Legionella reported although this is most certainly a gross under-
estimation of the scope of the problem, since the disease resembles
others and is frequently misdiagnosed.
---------------------------------------------------------------------------
There are numerous other adverse health effects such as nausea,
dizziness, fatigue, pulmonary edema, asthma and aggravation of existing
cardiovascular disease, which have been associated with poor indoor air
quality. Evidence in the record indicates that between 20 and 30% of
office buildings are ``sick'', having environments which may lead to a
variety of these effects. Unfortunately, quantitative data are not
systematically available on all of these effects.
For purposes of risk evaluation, however, as explained more fully
in the risk assessment discussion, the Agency has primarily focussed on
two health effects commonly associated with poor indoor air quality:
upper respiratory symptoms and severe headaches. The upper respiratory
symptoms associated with poor indoor air quality (sick building
syndrome) include stuffy nose, runny nose, dry itchy eyes, nose and
throat. For purposes of our evaluation, ``severe headaches'' are
defined as those serious enough to require medical attention or
restrict activity, but excludes migraines.
Unlike lung cancer and heart disease (health effects associated
with exposure to ETS), these effects will not lead to death. There is
no doubt, however, that OSHA does have the authority to regulate
working conditions that lead to the type of upper respiratory effects
and severe headaches described herein.
Clearly the upper respiratory effects and severe headaches
associated with poor indoor air quality are of the type that interfere
with the performance of work. The severe headaches were such that
medical treatment had to be sought; certainly such headaches were
impairing at the time they occurred, even though they were not
permanent. The upper respiratory symptoms were also severe enough to
either require medical attention or restrict activity.
There is ample precedent in OSHA rulemaking proceedings for the
regulation of working conditions to avoid health impairments that are
material but not life threatening. The Supreme Court in the cotton dust
case,2 concluded that OSHA had the authority to promulgate
regulations that would avoid Byssinosis, a respiratory disease which in
the large majority of cases is not deadly or disabling, and is
reversible if the employee left the cotton mills. Stage \1/2\
byssinosis, the most frequent type, has relatively mild symptoms. In
the case of occupational exposure to formaldehyde, the regulation was
designed to avoid, among other things, sensory irritation.3
---------------------------------------------------------------------------
\2\AFL-CIO v. Marshall, 452 U. S. 490 (1981)
\3\ See 52 FR 46168, 46235 (12/4/87)
---------------------------------------------------------------------------
Moreover in the ``Air Contaminants'' standard, OSHA regulated many
chemicals, such as acetone, gypsum and limestone which caused less
severe impairments of health.4 In promulgating the final air
contaminant rule OSHA analyzed which sorts of conditions would
constitute material impairment, concluding that ``. . . the OSH Act is
designed to be protective of workers and is to protect against
impairment with less impact than severe impairment.''5 The less
severe conditions, such as upper respiratory symptoms and severe
headaches, caused by poor indoor air quality are the same type as the
PELs preamble concluded were material impairments. These specific
conclusions of the Agency with respect to what constitutes material
impairments were upheld by the Court of Appeals on review6
although the Court disagreed with OSHA on other matters.
---------------------------------------------------------------------------
\4\See 54 FR 2332, 2361 (1/19/89)
\5\ See discussion, 54 FR at 2361-2362
\6\See AFL-CIO v. OSHA, 965 F. 2d 962, 975 (11th Cir., 1992).
The Court noted that ``section 6(b)(5) of the Act charges OSHA with
addressing all forms of `material impairment of health or functional
capacity,' and not exclusively `death or serious physical harm' . .
. from exposure to toxic substances.''
---------------------------------------------------------------------------
Therefore OSHA concludes that the adverse health effects caused by
poor indoor air quality, which range from legionellosis to severe
headaches to upper respiratory symptoms are material impairments of
health which the Act allows the Agency to regulate.
The effects of the pneumonia caused by Legionella are deadly or
severe. Although the rate of risk may not be as large as 1/1000 because
the number of employees at risk is large. This effect alone makes a
substantial contribution to a finding of significant risk, especially
when taking into account the large number of cases.
As to the severe headaches, the Agency estimates that the excess
risk of developing the type of non-migraine headache which may need
medical attention or restrict activity which has been associated with
poor indoor air quality is 57 per 1,000 exposed employees. In addition
the excess risk of developing upper respiratory symptoms which are
severe enough to require medical attention or restrict activity is
estimated to be 85 per 1,000 exposed employees. These numbers are
extrapolated from actual field studies and therefore show the magnitude
of the problem at present. There is no doubt that better maintenance of
ventilation systems such as required in the proposal will improve the
quality of air in covered workplaces and reduce the number of cases. In
addition the types of good practices prescribed in the proposal will
substantially reduce the type of microbial contamination associated
with Legionnaire's disease. Therefore, OSHA concludes that this number
of less severe effects along with the severe effects from Legionnaire's
disease, together, constitute a significant risk. Accordingly, OSHA
preliminarily concludes that, the proposal will substantially reduce a
significant risk of material impairment of health from poor indoor air
quality.
VI. Preliminary Regulatory Impact Analysis
A. Introduction
Executive Order 12886 requires a Regulatory Impact Analysis and
Regulatory Flexibility Analysis to be prepared for any regulation that
meets the criteria for a ``significant regulatory action.'' One of
these criteria, relevant to this rulemaking is that the rule have an
effect on the economy of $100 million or more per year. Based upon the
preliminary analysis presented below, OSHA finds that the proposed
standard will constitute a significant regulatory action.
The estimates presented in this Phase 1 Preliminary Regulatory
Impact Analysis demonstrate technological and economic feasibility of
the proposed standard. The analysis provides a non-detailed preliminary
count of the affected employees and buildings, the associated costs,
and benefits of the proposed standard provisions.
OSHA estimates the annual cost of compliance with the IAQ standard
to be $8.1 billion, of which the most costly provision will be for the
building systems operation and maintenance, $8.0 billion. The cost for
eliminating exposure to ETS may range from $0 to $68 million depending
on whether establishments ban smoking or allow smoking in designated
areas. In order to assess the overall economic impact of the rule, OSHA
also estimated the cost savings to employers, or cost savings that will
result from the implementation of the proposed standard. The major
forms of these savings are efficiency and productivity improvements,
cost reductions in operations and maintenance, and reduced incidence of
property damage. Cost savings associated with productivity improvements
are estimated to be $15 billion annually.
OSHA preliminarily estimates that the proposed standard will
prevent 3.0 million severe headaches and 4.5 million upper respiratory
symptoms over the next 45 years. This is, approximately, 69,000 severe
headaches and 105,000 upper respiratory symptoms per year. These
estimates understate the prevalence of building-related symptoms since
they reflect excess risk in only air conditioned buildings. In
addition, 5,583 to 32,502 lung cancer deaths and 97,700 to 577,818
coronary heart disease deaths related to occupational exposure to ETS
will be prevented over the next 45 years. This represents 140 to 722
lung cancer deaths per year and 2,094 to 13,001 heart disease deaths
per year.
B. Industry Profile
The environmental concern for air pollution has been largely
focussed on questions of outdoor air contamination. Recently, however,
attention has begun to shift to concerns about the quality of air
within buildings since people spend 80 to 90 percent of their time
indoors [Ex. 3-1075H].
Indoor air is a variable complex mixture of chemicals and airborne
particles. Its composition largely depends on the outdoor environment
(urban or rural area), the shelter itself (age, construction material,
electric equipment, heating, cooling, and ventilation systems), the
activities of the occupants (smoking, nonsmoking, cooking by gas, oil
or electricity) and the presence of plants and animals.
The Industry Profile chapter characterizes the building stock and
describes the factors that affect indoor air quality. This section also
presents the number of employees who work in buildings whose indoor air
will be affected by the proposed standard.
1. Affected Industries
The standard covers all OSHA regulated industries: Agriculture, Oil
and Gas Extraction (SIC 13), Manufacturing, Transportation,
Communications, Wholesale Trade, Retail Trade, Finance, Insurance and
Real Estate and Services. The scope of the proposal is twofold. The
proposed indoor air quality compliance provisions would only cover
employers with non-industrial work environments. This includes public
and private buildings, schools, healthcare facilities, offices and
office areas. Coverage also applies to nonindustrial work environments
that are part of industrial worksites (e.g., an office, cafeteria, or
break room located at a manufacturing facility).
The provisions for protecting the nonsmoking employees from
exposure to ETS apply to all indoor or enclosed work environments, in
industrial and nonindustrial establishments. This would include
maritime, construction, and agricultural workplaces.
2. Indoor Contaminants-Sources
Indoor air contaminants emanate from a broad array of sources that
can originate both outside of structures as well as from within a
building. When a building is new, some contaminants are given off
quickly and soon disappear. Others continue off-gassing at a slow pace
for years. Common office supplies and equipment have been found to
release hazardous chemicals--especially duplicators and copiers. Bulk
paper stores have been found to release formaldehyde [Ex. 3-1087A20].
Some typical contaminants are listed below:
(a) Gases and Vapors (organic/inorganic):
--Radon
--Sulfur dioxide
--Ammonia
--Carbon Monoxide
--Carbon Dioxide
--Nitrous Oxides
--Formaldehyde
(b) Fibers:
--Asbestos
--Fiberglass/Mineral Wools
--Textiles/Cotton
(c) Dusts:
--Allergens
--Household dust (mites)
--Pollens:
--Feathers
--Danders
--Spores
--Smoke/Fume
--Environmental Tobacco Smoke
--Coal
--Wood
(d) Microbes:
--Bacteria
--Fungi
--Viruses
People contribute millions of particles to the indoor air primarily
through the shedding of skin scales. Many of these scales carry
microbes, most of which are short lived and harmless. Clothing,
furnishings, draperies, carpets, etc. contribute fibers and other
fragments. Cleaning processes, sweeping, vacuuming, dusting normally
remove the larger particles, but often increase the airborne
concentrations of the smaller particles. Cooking, broiling, grilling,
gas and oil burning, smoking, coal and wood generate vast numbers of
airborne indoor pollutants in various classifications.
3. Controlling Indoor Air
Control of pollutants at the source is the most effective strategy
for maintaining clean indoor air. However, control or mitigation of all
sources is not always possible or practical. In the case of ETS, this
means restricting smoking to separately ventilated spaces. General
ventilation is, therefore, the second most effective approach to
providing acceptable indoor air [Exs. 3-1061G, 3-1075J].
Outside air dilutes and removes contaminants through natural
ventilation, mechanical ventilation or through infiltration and
exfiltration. Natural ventilation occurs when desired air flows occur
through windows, doors, chimneys and other building openings.
Mechanical ventilation is the mechanically induced movement of air
through the building. Mechanical systems usually condition and filter
the air and allow for the entry of outdoor air through outdoor dampers.
Infiltration is the unwanted movement of air through cracks and
openings into the building shell.
The outside air ventilation rate of a building affects indoor air
quality. It determines the extent to which contaminants are diluted and
removed from the indoor environment. The extent to which outside air
ventilation is effective in diluting indoor contaminants depends on how
well outside air is mixed with indoor air and is reflected by
ventilation efficiency. Ventilation efficiency can be reduced by air
short-circuiting from the supply diffusers to the return inlets, by
modular furniture partitions, and differences between the supply air
temperature and the room air temperature.
The rate at which outside air is supplied to a building is
specified by the building code at the design stage. Outside air
ventilation rates are based primarily on the need to control odors and
carbon dioxide levels (e.g., occupant-generated contaminants or
bioeffluents). Carbon dioxide is a component of outdoor air whose
excessive accumulation indoors can indicate inadequate ventilation.
Lack of adequate ventilation contributes to indoor air related
health complaints. Specific deficiencies that produce air quality
problems include inadequate outside air supply, poor air distribution,
poor air mixing (and therefore poor ventilation efficiency), inadequate
control of humidity, insufficient maintenance of the ventilation
system, inadequate HVAC system capacity and inadequate exhaust from
occupied areas. Inadequate outdoor air supply and distribution and
insufficient control of thermal conditions can result from strategies
to control energy consumption. In approximately 500 indoor air quality
investigations conducted in the late 1970's and early 1980's, the
National Institute for Occupational Safety and Health (NIOSH) found
that the primary causes of indoor air quality problems were inadequate
ventilation (52%), contamination from outside the building (10%),
microbial contamination (5%), contamination from building fabric (4%)
and unknown sources (13%) [56 FR 47892]. To date, NIOSH has conducted
over 1,100 IAQ related investigations, but has not yet evaluated them
to provide updated estimates.
OSHA, therefore, believes that it is necessary to require
maintenance of the HVAC system components that directly affect IAQ,
since failure to do so results in the degradation of IAQ. Standards of
HVAC maintenance vary and sometimes are deficient where untrained
personnel are designated to maintain complex systems. It is, also,
customary for companies to defer maintenance for economic and budgetary
reasons, with adverse impacts on IAQ. Some examples of maintenance
deficiencies include: plugged drains on cooling coil condensate drip
pans (resulting in microbial contamination); failed exhaust fans in
underground parking garages; microbial fouling of cooling tower water
from lack of water treatment with biocides resulting in legionellosis
cases; and failure of the automatic temperature control system
resulting in lack of outside ventilation air.
4. Building Characteristics
During the last 25 years, technical and socioeconomic changes have
profoundly influenced the methods employed to plan, design, construct
and operate buildings. Buildings system design, maintenance and
operation can, and regularly do, provide acceptable indoor
environments. However, neglect or disregard of the sources of indoor
air contaminants, or of the proper design, operation and maintenance of
building system components which influence indoor air quality can
create an uncomfortable and unhealthy indoor atmosphere [Ex. 3-1075H2].
The oil embargo of 1973 brought about the realization that
considerable savings could be made in reducing the consumption of
energy used to heat and cool buildings. Prior to 1973, the energy to
heat and cool buildings was much cheaper and the buildings reflected
that reality. Building enclosures had lower insulating values and
allowed more infiltration. More air was circulated to the occupied
spaces and more outdoor air was provided for ventilation. This resulted
in a lower concentration of pollutants and higher velocities of air
motion in indoor air. Office buildings were divided into individual
rooms with their own walls as opposed to the current practice of open
spaces with movable screens [Ex. 4-74].
The centralization of services and the expanding economy have led
to concentration of office space in the cities. The cost of land has
shaped buildings into high-rise structures. The cost of materials and
popularity of mirror glass has led to the sprouting of hundreds of what
may be termed ``glass boxes''. These boxes are sealed to keep out noise
and pollution--mainly from traffic.
Buildings designed after 1973 have incorporated many energy
conservation measures that range from adjusting thermal comfort zones
to increased awareness of lighting efficiency, to designing new
operating methods for ``sealed building'' [Ex.3-1159, p.1]. In large
buildings, outside air ventilation rates were also reduced by closing
outside air dampers in mechanical ventilation systems at nights, on
weekends and sometimes even during occupancy. As a result of these
measures, which primarily reduced costs for conditioning outdoor air as
opposed to increasing energy efficiency, considerable energy savings
have been achieved in buildings.
In addition, during the 1970's variable air volume (VAV) HVAC
systems became widely accepted. VAV systems condition supply air to a
constant temperature and insure thermal comfort by varying the airflow.
Early VAV systems did not allow control of the outside air quantity, so
that a decreasing amount of outside air was provided as the flow of
supply air was reduced.
In some cases, building design flaws contribute to the poor quality
of indoor air, such as locating air intake vents near to a loading dock
or parking garage. Design flaws of interior space also contribute to
indoor air problems. Most building cooling systems are designed to
remove the heat generated by office machines, employees and light. The
heat generated by these sources often exceeds the capacity of the HVAC
system to remove it [Ex.3-1159C1]. Ideally with effective filtration
and management systems, the air indoors should be cleaner than the air
outdoors.
5. Profile of Affected Buildings
Estimates of the number of buildings potentially affected by the
indoor air standard were developed by OSHA based on Department of
Energy's commercial building energy consumption survey (CBEC) 1989\7\
[Ex. 4-303]. There is a total of 4.5 million commercial buildings in
the United States. Commercial buildings are defined as all non-
manufacturing/industrial and non-residential structures. Table VI-1
presents the distribution of buildings by use, occupancy and thermal
conditioning. Approximately 28 percent of all buildings are for
mercantile or services. Other uses include offices (15 percent),
assembly and warehouses (14 percent each), food service (5 percent),
lodging (3 percent) and food sales and healthcare (2 percent each). The
``other'' category (1 percent) covers buildings such as public
restrooms and buildings that are 50 percent or more commercial but
whose principal activity is agricultural, industrial/manufacturing or
residential.
---------------------------------------------------------------------------
\7\The commercial building and energy consumption survey is a
triennial national sample survey of commercial buildings and their
energy suppliers. This survey is the only source of national level-
data on both commercial building characteristics and energy
consumption.
---------------------------------------------------------------------------
On average, the largest types of buildings are for education and
health care. Mercantile and service buildings account for the greatest
number and floorspace of any single activity category. Office buildings
account for nearly as much floorspace, but far fewer buildings.
Together office and mercantile buildings represent almost 40 percent of
all buildings and floorspace. Warehouses and assembly buildings both
are almost as numerous as office buildings, but account for less
floorspace. Over 62 percent of buildings have only one floor and 13
percent have three or more floors. Most buildings (69%) house single
establishments. Government occupied buildings represent 13 percent.
Table VI-1.--Employees Working in Buildings and Other Building Characteristics
----------------------------------------------------------------------------------------------------------------
Percent of
Principle building activity Number of all Total number
buildings buildings employees
----------------------------------------------------------------------------------------------------------------
Principal building activity:
Assembly................................................. 615,000 14 4,012,000
Education................................................ 284,000 6 7,204,000
Food sales............................................... 102,000 2 844,000
Food service............................................. 241,000 5 1,943,000
Health care.............................................. 80,000 2 4,225,000
Lodging.................................................. 140,000 3 3,092,000
Mercantile and service................................... 1,278,000 28 12,414,000
Office................................................... 679,000 15 27,780,000
Parking garage........................................... 45,000 1 332,000
Public order and safety.................................. 50,000 1 861,000
Warehouse................................................ 618,000 14 4,377,000
Other.................................................... 62,000 1 2,111,000
Vacant\1\................................................ 333,000 7 1,472,000
--------------------------------------------------
Total.................................................. 4,527,000 ........... 70,667,000
Building occupants:
Single establishments--owner occupied.................... 2,445,000 54
Multiple establishments--owner occupied.................. 369,000 8
Single establishments--nonowner occupied................. 672,000 15
Multiple establishments--nonowner occupied............... 259,000 6
Vacant................................................... 206,000 5
Government buildings..................................... 577,000 13
Thermal conditioning:
Heated................................................... 3,865,000 85
--------------------------------
Entire building...................................... 2,739,000 60
Part of building..................................... 1,126,000 25
Cooled................................................... 3,184,000 70
--------------------------------
Entire building...................................... 1,550,000 34
Part of building..................................... 1,634,000 36
----------------------------------------------------------------------------------------------------------------
\1\Vacant buildings may contain occupants who are using up to 50 percent of the floorspace.
Source: U.S. Energy Information Administration, Commercial Buildings Characteristics 1989. Washington, DC. June
1991.
The survey also provides information on the number of buildings
with heating and air conditioning systems. Total number of heated
buildings is estimated to be 3.9 million. Heating systems include
boilers, furnaces, individual space heaters, and packaged heating
units. Almost one-half of all the buildings are heated by forced-air
central systems. Air-distributing heat and cooling systems are most
prevalent in office, mercantile and service buildings. The survey
reveals that 70 percent of the buildings have air conditioning. It also
shows that 80 percent of the buildings have heat and air conditioning,
and 12 percent have heat, but no air conditioning.
Over 40 percent of the floorspace built since 1986 was in a
building with a computerized energy management and control systems
(EMCS). EMCS is an energy conservation feature that uses mini/micro
computers, instrumentation, control equipment and software to manage a
building's use of energy for heating, ventilation, air conditioning,
lighting and/or business related processes. These systems can also
manage fire control, safety and security. Overall, EMCS are present in
buildings accounting for 23 percent of floorspace. EMCS controls HVAC
in only 251,000 buildings or 6 percent of total number of buildings.
However, the DOE survey [Ex. 4-303] does not provide data by two-
digit Standard Industrial Classification (SIC). The number of buildings
by SIC will determine subsequent costs. OSHA applied the DOE estimates
of the number of buildings by type of occupancy (single or multi-
tenant) to the number of establishments by two-digit SIC given by the
Bureau of Labor Statistics. First, OSHA allocated non-government single
tenant buildings (estimated at 3.1 million) across the relative two-
digit SIC using the relative two-digit SIC distribution of the number
of establishments. Then, OSHA allocated the 0.8 million non-government
multi-establishment buildings across two-digit SIC using the relative
two-digit SIC distribution of the number of establishments in multi-
establishment buildings (2.8 million). All government buildings were
considered single tenant buildings. OSHA recognizes that this
methodology of classification of buildings by two-digit SIC code may
not reflect the fact that establishments in multi-tenant buildings
should be allocated across several SICs or the fact that some single
establishment buildings may be concentrated in certain SICs instead of
all SICs. This is particularly true for the agricultural sector for
which farms and farm buildings (silos, grain elevators and barns) are
outside the scope of the IAQ portion of the proposal. However, OSHA
does not have the data to provide such delineation at this point. Table
VI-2 presents OSHA's estimate of the number of buildings by two-digit
SIC and by characteristics of occupancy and ventilation system.
Table VI-2.--Number of Buildings and Establishments Affected by IAQ Proposed Standard
----------------------------------------------------------------------------------------------------------------
Number of
Buildings with Buildings with Total Number of Number of naturally
SIC industry single multiple number of heated cooled ventilated
establishments establishments buildings buildings buildings buildings\1\
----------------------------------------------------------------------------------------------------------------
Agriculture, forestry,
fishing................... 136,629 36,557 173,186 147,806 124,312 10,564
Mining..................... 11,976 3,204 15,181 12,956 10,897 926
Construction............... 336,841 90,127 426,968 364,398 306,475 26,045
Manufacturing.............. 203,995 54,582 258,577 220,684 185,605 15,773
Transportation............. 127,706 34,170 161,876 138,154 116,193 9,874
Wholesale and retail trade. 1,011,035 270,518 1,281,553 1,093,747 919,889 78,175
Finance, insurance, real
estate.................... 275,760 73,784 349,544 298,320 250,900 21,322
Services................... 1,013,057 271,058 1,284,115 1,095,934 921,729 78,331
Government................. 577,000 .............. 577,000 505,000 348,000 35,197
------------------------------------------------------------------------------------
Total................ 3,694,000 834,000 4,528,000 3,877,000 3,184,000 276,208
----------------------------------------------------------------------------------------------------------------
\1\Based on estimate of 6.1 percent of floorspace without HVAC.
Source: OSHA, Office of Regulatory Analysis, 1994.
6. Buildings With Indoor Air Problems
Many published reports on building wellness describe buildings in
terms of two general categories, sick or well buildings. Some of the
published categories, in addition to the terms sick or well are:
problem buildings and non-problem buildings, healthy buildings;
buildings with high and low rates of IAQ related complaints; sick
building syndrome (SBS).
The SBS symptom complex is characterized by a range of symptoms
including but not limited to, eye, nose and throat irritation, dryness
of mucous membranes and skin, nose bleeds, skin rash, mental fatigue,
headache, cough, hoarseness, wheezing, nausea and dizziness [Ex. 4-
159]. Within a given building there will usually be some commonality
among the symptoms manifested as well as temporal association between
occupancy in the building and appearance of symptoms. Many people who
work in buildings characterized as having SBS typically exhibit health
symptoms that disappear when the person is no longer in the building.
In most cases, a physical basis for the occurrence of the SBS can be
found: lack of proper maintenance, changes in thermal or contaminant
loads imposed during the building's life, changes in control strategies
to meet new objectives (e.g., energy conservation) or inadequate
design.
Building-related illnesses (BRI), on the other hand, are medically
diagnosed diseases that present symptoms that can last for weeks,
months, years or even a lifetime. Examples include nosocomial
infections, humidifier fever, hypersensitivity pneumonitis, and
legionellois. BRI can develop as a result of poor building systems
operation and maintenance and uncontrolled point sources of
contaminants.
No building has a complete absence of problems, but those that
function with minimal occupant complaints and comply with acceptable
criteria for occupant exposure, system performance, maintenance
procedures and economic objectives may be characterized as healthy
buildings. Figure VI-1 below presents the classification of buildings
by stages of performance.
Based on the information submitted to the docket, OSHA assumed that
30 percent of the buildings have indoor air quality problems [Ex. 3-
745].
BILLING CODE 4510-26-P
TP05AP94.000
BILLING CODE 4510-26-C
Therefore, as presented in Table VI-3, the total number of problem
buildings is estimated to be 1.4 million buildings.
7. Number of Employees Affected
The commercial building energy consumption survey estimates that
there are 70.7 million employees. However, survey data do not provide
information by two-digit SIC. OSHA examined data obtained through the
Bureau of Labor Statistics to estimate the number of employees by two-
digit SIC affected by the proposed standard. The data from the Bureau
provided occupational breakdown of the labor force by detailed industry
categories (two-digit SIC) and major occupational groupings.
Table VI-3.--Number of Problem Buildings and Number of Employees Exposed to Indoor Air Quality Problems\1\
----------------------------------------------------------------------------------------------------------------
Number of
Employees Number of employees
working buildings with exposed to IAQ
indoors\2\ IAQ problems problems\3\
----------------------------------------------------------------------------------------------------------------
Agriculture, forestry, fishing.................................. 279,050 51,956 83,715
Mining.......................................................... 180,700 4,554 54,210
Construction.................................................... 1,643,750 128,091 493,125
Manufacturing................................................... 5,748,000 77,573 1,724,400
Transportation.................................................. 3,412,350 48,563 1,023,705
Wholesale and retail trade...................................... 15,744,000 384,466 4,723,200
Finance, insurance, real estate................................. 7,248,150 104,863 2,174,445
Services........................................................ 26,926,000 385,235 8,077,800
Government...................................................... 9,473,561 173,100 2,842,068
-----------------------------------------------
Total..................................................... 70,655,561 1,358,400 21,196,668
----------------------------------------------------------------------------------------------------------------
\1\Exclusive of exposure to ETS.
\2\OSHA estimate based upon BLS's 1993 employed persons by detailed industry and major occupation.
\3\Based on OSHA estimate of 30 percent employee exposure to poor IAQ.
Source: OSHA, Office of Regulatory Analysis, 1994.
OSHA classified employees according to whether or not they work
primarily in indoor areas, e.g., areas with possible exposures, by
developing percentages of employees in each occupational category who
might be working indoors. For example, personnel in the transportation
industries were apportioned according to those potentially exposed to
indoor air pollution (office workers) and those who are not (truck
drivers). Table VI-3 presents the distribution of the 70.7 million
employees who work indoors.
No data are available as to the number of employees exposed to poor
indoor air quality. Based on OSHA's percentage of problem buildings (30
percent), OSHA assumed that 30 percent of employees working indoors are
exposed to poor indoor air quality. Therefore, the number of employees
potentially affected is 21 million.
8. Environmental Tobacco Smoke
Environmental Tobacco Smoke (ETS) represents one of the strongest
sources of indoor air contaminants in buildings where smoking is
permitted. ETS is a mixture of irritating gases and carcinogenic tar
particles and is considered one of the most widespread and harmful
indoor air pollutants.
(a) Smoking ordinances\8\ and policies. State and Local Governments
have adopted an increasing number of ordinances and regulations
limiting smoking in public and private worksites. The restrictiveness
of these laws varies from simple, limited prohibitions to laws that ban
smoking. Forty-five states and the District of Columbia restrict
smoking in public workplaces and 19 states and the District of Columbia
restrict smoking in private workplaces.
---------------------------------------------------------------------------
\8\A smoking ordinance may mean any local law which addresses
public smoking in some fashion to protect non-smokers.
---------------------------------------------------------------------------
There are 397 city and county smoking ordinances covering 22
percent of the total population [Ex. 4-305]. A total of 297 cities and
counties mandate the adoption of workplace smoking policies. Typically
these provisions require employers (private and public) to maintain a
written smoking policy. Ordinances range from requirements for written
smoking policies to the total elimination of smoking in the workplace.
A total of 505 cities and counties limit smoking, specifically in
restaurants. The requirements range from a nonsmoking section of
unspecified size to the banning of all smoking [Ex. 4-305].
A 1991 survey of company smoking policies shows that of the 85
percent of firms with smoking policies, 34 percent have complete bans
and another 34 percent prohibit smoking in all open work areas. Over 90
percent of non-manufacturing establishments have smoking policies [H-
030 Ex. 77].
Workplace smoking policies are more common in larger businesses. In
a survey of personnel managers, 63 percent of those with 1,000 or more
employees reported having a smoking policy compared with 52 percent of
companies with fewer employees. In the same survey, smaller companies
were half as likely as larger ones to have a policy under
consideration. Similar findings were reported by the National Survey of
Worksite Health Promotion Activities, in which larger worksites were
more likely than smaller ones to report smoking control activities. In
a survey of private New York city businesses, only 4 percent of
companies with fewer than 100 employees had a written smoking policy
[Ex. 3-1030Q].
(b) Number of nonsmokers working indoors. Based on the National
Health Interview Survey, OSHA estimated that 74.2 million employees or
73.01 percent of the U.S. labor force covered by OSHA are nonsmokers.
Table VI-4 presents the distribution of nonsmoking employees by two
digit SIC.
Results of population based surveys show that 88 percent of
nonsmokers are aware of the negative health consequences of ETS.
Despite this general awareness, exposure to ETS is pervasive [Ex. 4-
98]. To determine the occupational exposure of nonsmoking employees to
ETS, OSHA used the estimate provided by the 1991 National Health
Interview Survey. The survey, requested information from employed
individuals on whether during the past two weeks anyone smoked in their
immediate work area. Based on results adjusted for non-response and
weighted to reflect national estimates, 18.81 percent reported exposure
to ETS. OSHA believes that the 18.8 percent is an underestimate since
it is based solely on self reported information and the question was
not very specific in defining ``immediate'' work area. A recent
reanalysis of a study by Cummings et al. [Ex. 4-68] shows that 48.67
percent of currently employed nonsmokers reported ETS exposure at work
and not at home [Ex. 3-442F].
Table VI-4.--Employees Exposed to Environmental Tobacco Smoke
----------------------------------------------------------------------------------------------------------------
Number of employees exposed to
ETS
SIC industry Nonsmoker -------------------------------
employees\1\ Lower bound Upper bound
(18.81%) (48.67%)
----------------------------------------------------------------------------------------------------------------
Agriculture, forestry, fishing.................................. 1,008,007 189,606 490,597
Mining.......................................................... 249,256 46,885 121,313
Construction.................................................... 3,479,876 654,565 1,693,655
Manufacturing................................................... 13,050,099 2,454,724 6,351,483
Transportation.................................................. 3,953,337 743,623 1,924,089
Wholesale and retail trade...................................... 19,041,884 3,581,778 9,267,685
Finance, insurance, real estate................................. 3,995,180 751,493 1,944,454
Services........................................................ 21,687,986 4,079,510 10,555,543
Government...................................................... 7,735,393 1,455,027 3,764,816
-----------------------------------------------
Total..................................................... 74,201,019 13,957,212 36,113,636
----------------------------------------------------------------------------------------------------------------
\1\Based on 73.01 percent nonsmoking employees.
Source: OSHA, Office of Regulatory Analysis, 1994.
By applying the lower and upper ranges of exposure, OSHA estimates
that the number of nonsmoking employees exposed to ETS to be 13.9 to
36.1 million employees.
C. Nonregulatory Alternatives
(1) Introduction
The declared purpose of the Occupational Safety and Health (OSH)
Act of 1970 is ``* * * to assure so far as possible every working man
and woman in the Nation safe and healthful working conditions and to
preserve our human resources. * * *'' Thus, the Act requires the
Secretary of Labor, when promulgating occupational safety and health
standards for toxic materials or harmful physical agents, to set the
standard ``* * * that most adequately assures, to the extent feasible,
on the basis of the best available evidence, that no employee will
suffer material impairment of health or functional capacity. * * *'' It
is on the basis of this congressional directive that OSHA has initiated
regulatory actions to reduce the adverse health effects associated with
occupational exposure to indoor air pollutants.
The discussion below assesses the requisite preconditions for
optimal safety in the context of a free market economy, and real world
economic factors are compared with the free market paradigm to
illustrate the shortcoming of the nonregulatory environment.
(2) Market Imperfections
Economic theory suggests that the need for government regulation is
greatly reduced where private markets work efficiently and effectively
to allocate health and safety resources. The theory typically assumes
perfectly competitive labor markets where employees, having perfect
knowledge of job risks and being perfectly mobile among jobs, command
wage premiums that fully compensate for any risk of future harm. Thus,
theoretically, the costs of occupational injury and illness are borne
initially by the firms responsible for the hazardous workplace
conditions and ultimately by the consumers who pay for the final goods
and services produced by these firms. With all costs internalized,
private employers have an incentive to reduce hazards wherever the cost
of hazard abatement is less than the total cost to the firm, the work
force, and society of the expected injury or illness.
The conditions of perfect competition do not need to be completely
satisfied in order for the forces of the market to approximate an
efficient outcome. However, some market imperfections can produce sub-
optimal results that can be improved upon with regulatory action. In
the case of this rulemaking, employees face a significant health risk
which is not adequately addressed by current nonregulatory
alternatives. OSHA, therefore, believes that it must take appropriate
actions to provide greater health protection for workers exposed to
toxic substances.
Although OSHA believes that adequate job safety and health could
exist in the private market under perfect conditions, the private
market often fails to provide acceptable levels of safety and health in
instances where these conditions are not met. It appears that at least
two of several conditions traditionally considered essential components
of perfect markets are absent from the environment in which employees
are exposed to hazards associated with exposure to indoor pollutants:
(1) Perfect employee knowledge of risks and (2) perfect employee
mobility between jobs.
First, evidence on occupational health hazards in general suggests
that in the absence of immediate or clear-cut danger, employees and
employers have little incentive to seek or provide information on the
potential long-term effects of exposure. Employers faced with
potentially high compensatory payments may, in fact, have a
disincentive to provide information to employees. When relevant
information is provided, however, employers and employees might still
find informed decisionmaking a difficult task, especially where long
latency periods precede the development of chronic disabling disease.
Moreover, if signs and symptoms are nonspecific--that is, if an illness
could be job-related or could have other causes--employees and
employers may not link disease with such occupational exposure.
Second, even if workers were fully informed of the health risks
associated with exposure to hazardous substances, many face limited
employment options. Nontransferability of occupational skills and high
national unemployment rates sharply reduce a worker's expectation of
obtaining alternative employment quickly or easily.
In many regions of the country, the practical choice for workers is
not between a safe job and a better paying but more hazardous position,
but simply between employment and unemployment at the prevailing rates
of pay and risk. In addition to the fear of substantial income loss
from prolonged periods of unemployment, the high costs of relocation,
the reluctance to break family and community ties, and the growth of
institutional factors such as pension plans and seniority rights serve
to elevate the cost of job transfer. Thus, especially where wages are
more responsive to the demands of more mobile workers who tend to be
younger and perhaps less aware of job risks, hazard premiums for the
average worker will not be fully compensated. Where this is the case,
labor market negotiations are unlikely to reflect accurately the value
that workers place on health.
In addition to these market imperfections, externalities occur if
employers and employees settle for an inefficiently low level of
protection from hazardous substances. For the competitive market to
function efficiently, only workers and their employers should be
affected by the level of safety and health provided in market
transactions. In the case of occupational safety and health, however,
society shares part of the financial burden of occupationally induced
diseases, including the costs of premature death, chronic illness, and
disability. Those individuals who suffer from occupationally related
illnesses are cared for and compensated by society through taxpayer
support of social programs, including welfare, Social Security, and
Medicare.
If private employers do not have to pay the full cost of
production, they have no economic incentive to reduce hazards whenever
the cost of hazard abatement is greater than the cost of the expected
illness. In this way, the private market fails to produce optimal
levels of safety.
(3) Alternative Non-regulatory Options
Based on the above evidence, OSHA has concluded that the private
market has failed to provide optimal levels of safety to employees.
Consequently, some form of intervention that fosters safer work
environments must be used to reduce occupational exposure. Because such
intervention need not occur through government regulation, OSHA has
considered the effectiveness of other non-regulatory options: (1)
relying on tort litigation and (2) relying on workers' compensation
programs.
(a) Tort Liability. The use of liability under tort law is one
nonregulatory alternative that has been increasingly used in litigation
concerning occupationally related illnesses. Prosser [Ex. 4-256]
describes a tort, in part, as a ``civil wrong, other than a breach of
contract, for which the court will provide a remedy in the form of an
action for damages''.
If the tort system applies, it would allow a worker whose health
has been adversely affected by occupational exposure to a hazardous
substance to sue and recover damages from the employer. Thus, if the
tort system is effectively applied, it might shift the liability of
direct costs of occupational disease from the worker to the firm under
certain specific circumstances.
With very limited exceptions, however, the tort system is not a
viable alternative in dealings between employees and employers. All
states have legislation providing that Workers' Compensation is either
the exclusive or principal remedy available to employees against their
employers. Thus, under tort law, workers with an occupational disease
caused by exposure to a hazardous substance can only file a product
liability suit against a third party manufacturer, processor,
distributor, sales firm, or contractor. It is often difficult, however,
to demonstrate a direct link between an exposure to a hazardous
substance and the illness.
In order to pursue litigation successfully, there must be specific
knowledge of the magnitude and duration of a worker's exposure to a
hazardous substance, as well as the causal link between the disease and
the occupational exposure. Usually, it is extremely difficult to
isolate the role of occupational exposures in causing the disease,
especially if workers are exposed to many toxic substances and the
exposure is not necessarily limited to the workplace such as the case
for ETS. This difficulty is further compounded by the long latency
periods that are frequently involved. In addition, the liable party
must be identifiable, but workers may have several employers over a
working lifetime. The burden of proof that an occupational exposure to
a hazardous substance occurred, that a specific employer is the liable
party, and that the exposure level was significant may prohibit the
individual from initiating the suit.
There are an increasing number of lawsuits that are related to
health effects to building occupants from poor indoor air quality.
These lawsuits are typically filed after the illness or health effect
has been diagnosed. In this sense, increasing pressure is being placed
on businesses. However, the legal pressure currently does not relate to
the implementation of a clean indoor air policy (e.g., legal action is
not currently being taken just because a company does not have a clean
indoor air policy. These actions are event related as opposed to being
policy related). IAQ litigation is growing rapidly and the focus is
shifting from residential to commercial facilities. Examples to
emphasize that are the recent $12.5 million claims against the Social
Security Administration for the Richmond, California episode of
Legionnaire's disease, the Call versus Prudential case in which
building tenants settled with the defendants in what may have been the
first jury trial in sick building litigation, and a suit by Hamilton,
Ohio, county employees against their office building owners alleging
exposure to fumes, bacteria, fungi, dust and irritants [Ex. 3-575].
Legal proceedings do not internalize occupational illness costs
because they involve substantial legal fees associated with bringing
about court action. In deciding whether to sue, the tort victim must be
sure that the size of the claim will be large enough to cover legal
expenses. In effect, the plaintiff is likely to face substantial
transaction costs in the form of a contingency fee, commonly 33
percent, plus additional legal expenses. The accused firm must also pay
for its defense. The high costs and uncertainties associated with tort
law make it an inefficient mechanism for ensuring adequate protection
of workers' health.
Insurance and liability costs are not borne in full by the specific
employer responsible for the risk involved. For firms that are insured,
the premium determination process is such that premiums only partially
reflect changes in risk associated with changes in exposure to
hazardous substances. This lack of complete adjustment is the so-called
``moral hazard'' problem, which is the risk that arises from the
possible imprudence of the insured. As the insured firm has paid an
insurance company to assume some of the risks, that firm has less
reason to exercise the diligence necessary to avoid losses. Transfer of
risk is a fundamental source of imperfection in markets.
There is a growing number of state and local laws and ordinances
controlling smoking. Armed with new data that show health effects from
indoor air pollutants, plaintiffs who believe that they have been
injured by the air inside their workplaces are beginning to take the
offensive. They are lobbying on the local, state and federal levels for
protective legislation, and in the absence of such legislation, they
are suing for damages to their health. These cases are complex not only
in the nature of the technical proof that must be developed and
presented, but also in the number of parties involved. Suits have been
filed against architects, builders, contractors, building product
manufacturers and realtors [Ex. 3-662].
(b) Workers' Compensation. The Workers' Compensation system is a
result of the perceived inadequacies in liability or insurance systems
to compel employers to prevent occupational disease or compensate
workers fully for their losses. The system was designed to internalize
some of the social costs of production, but in reality it has fallen
short of compensating workers adequately for occupationally related
disease. Thus, society shares the burden of occupationally related
health effects, premature mortality, excess morbidity, and disability
through taxpayer support of social programs such as welfare, Social
Security disability payments, and Medicare.
Compensation tends to be inadequate especially in permanent
disability cases, in view of the expiration of benefit entitlement and
the failure to adjust benefits for changes in a worker's expected
earnings over time. As of January 1987, eight states restricted
permanent disability benefits either by specifying a maximum number of
weeks for which benefits could be paid or by imposing a ceiling on
dollar payments [Ex. 4-302].
At present, time and dollar restrictions on benefit payments are
even more prevalent in the area of survivor benefits. The duration of
survivor benefits is often restricted to 10 years, and dollar maximums
on survivor payments range from $7,000 to $60,000. In addition, it
should be noted that if the employee dies quickly from the occupational
illness and has no dependents, the employer need pay only nominal
damages under Workers' Compensation (e.g., a $1,000 death benefit).
Finally, in spite of current statutory protection, disability from
occupational diseases represents a continuing, complex problem for
Workers' Compensation programs. Occupational diseases may take years to
develop, and more than one causal agent may be involved in their onset.
Consequently, disabilities resulting from occupationally induced
illness often are less clearly defined than those from occupationally
induced injury. As a result, Workers' Compensation is often a weak
remedy in the case of occupational disease. Indeed, there is some
evidence indicating that the great majority of occupationally induced
illnesses are never reported or compensated [Ex. 4-84].
The insurance premiums paid by a firm under the Workers'
Compensation system are generally not experience rated; that is, they
do not reflect the individual firm's job safety and health record.
About 80 percent of all firms are ineligible for experience rating
because of their small size. Such firms are class rated, and rate
reductions are granted only if the experience of the entire class
improves. Even when firms have an experience rating, the premiums paid
may not accurately reflect the true economic losses. Segregation of
loss experience into classes is somewhat arbitrary, and an individual
firm may be classified with other firms that have substantially
different normal accident rates. An experience rating is generally
based on the benefits paid to workers, not on the firm's safety record.
Thus, employers may have a greater incentive to reduce premiums by
contesting claims than by initiating safety measures.
In summary, the Workers' Compensation system suffers from several
shortcomings that seriously reduce its effectiveness in providing
incentives for firms to create safe and healthful workplaces. The
scheduled benefits are significantly less than the actual losses to the
injured workers, and recovery is often very difficult in the case of
occupational diseases. Thus, the existence of a Workers' Compensation
system limits an employer's liability significantly below the actual
costs of the injury. In addition, premiums for individual firms are
unlikely to be specifically related to that firm's risk environment.
The firm, therefore, does not receive the proper economic signals and
consequently fails to invest sufficient resources in reducing workplace
injuries and illnesses. The economic costs not borne by the employer
are borne by the employee or, as is often the case, by society through
public insurance and welfare programs.
(4) Conclusion
OSHA believes that there are no nonregulatory alternatives that
adequately protect workers from the adverse health effects associated
with exposure to indoor air pollution. Tort liability laws and Workers'
Compensation provide some protection, but due to market imperfections
they have not been sufficient. Some employers have not complied
voluntarily with standards recommended by professional organizations.
The deleterious health effects resulting from continued exposure to
hazardous substances require a regulatory solution.
D. Benefits
In this chapter, OSHA presents its preliminary estimates of the
expected reduction in fatalities and illnesses among the employees
affected by the proposed IAQ standard. A qualitative description of the
non-quantifiable additional cost savings to employers, is also
provided.
1. Indoor Air Quality
Health effects typically caused by poor IAQ have been categorized
as Sick Building Syndrome (SBS) or Building-Related Illness (BRI). Some
of the symptoms that characterize SBS include: irritation of eyes, nose
and throat, dry mucous membranes and skin and coughs, hoarseness of
voice and wheezing, hypersensitivity reaction, nausea and dizziness.
BRI describes specific medical conditions of known etiology such
as: Respiratory allergies, legionellosis, humidifier fever, nosocomial
infections, sensory irritation when caused by known agents and the
symptoms and signs characteristic of exposure to chemical or biologic
substances such as carbon monoxide, formaldehyde, pesticides,
endotoxins or mycotoxins. BRIs do not disappear when the person leaves
the building.
The Centers for Disease Control Prevention estimate that over
25,000 cases of the pneumonia caused by Legionella occur each year with
more than 4,000 deaths. It has been suggested that a large number of
these cases occur as the result of workplace exposure [Exs. 4-33, 4-
318]. However, specific data on the occurrence of Legionella-related
cases due to workplace exposure were not available.
Some of the reductions attributable to the proposed standard, such
as decreases in the number of upper respiratory symptoms (nose, throat
and eye symptoms) and severe headaches have been estimated. Other
reductions, however, have not been quantified at this time.
OSHA's estimates are based upon the exposure profile (presented in
Table VI-5) and OSHA's quantitative risk assessment discussed in detail
in the preamble to the proposal). OSHA preliminarily estimates the risk
of working in mechanically ventilated workplaces to be 57 severe
headaches and 85 upper respiratory symptoms per 1,000 employees over a
45 year work lifetime. By applying these rates to the affected
population at risk, OSHA estimates that 3.8 million severe headaches
and 5.6 million upper respiratory symptoms will develop in employees
over the next 45 years who work in buildings with mechanical
ventilation (with the worker population held constant).
A common theme that runs through the literature and the OSHA docket
indicates that the principal factor associated with indoor air quality
complaints is inadequate ventilation. However, information available
does not quantify the effectiveness of ventilation improvements. NEMI
reports that: ``ventilation system modifications and improvements are
key elements of solving existing IAQ problems and reducing IAQ
complaints. In every case where recommended ventilation system
modifications and improvements are implemented, the frequency and
severity of complaints are reduced significantly'' [Ex. 3-1183].
Some of the submissions base the effectiveness of ventilation
improvements on the NIOSH analysis of indoor air quality investigations
[Exs. 3-1183, 3-1090]. In approximately 500 indoor air quality
investigations, NIOSH found that the primary causes of indoor air
quality problems were inadequate ventilation (52%), contamination from
outside the building (10%), microbial contamination (5%), contamination
from building fabric (4%), and unknown sources (13%). Excluding
contamination from building fabric and unknown sources, this suggests
that 83 percent of complaints related to IAQ problems would be
eliminated by the proposed OSHA standard. For purposes of this
analysis, OSHA assumes that the overall effectiveness is, therefore, 80
percent. As shown in Table VI-5, OSHA estimates that the proposed
standard will prevent 3.0 million severe headaches and 4.5 million
upper respiratory symptoms over the next 45 years. This is,
approximately, 69,000 severe headaches and 105,000 upper respiratory
symptoms per year. These estimates understate the prevalence of
building-related symptoms since they only reflect excess risk in only
air conditioned buildings. OSHA believes that the standard will also
prevent severe headaches and upper respiratory symptoms in heated (but
not air conditioned) buildings, and that it will prevent various other
adverse health effects. OSHA is seeking additional information upon
which to base quantifiable estimates of the other known adverse health
effects.
OSHA requests comment on the methodology of estimating the benefits
for the IAQ portion of the proposal. Specifically, OSHA requests any
studies which document (in quantitative terms) the effectiveness of
HVAC maintenance on the decline of indoor air related ailments.
2. Environmental Tobacco Smoke
Tobacco smoke has been classified as a carcinogen by the
International Agency for Research on Cancer, the Surgeon General,
NIOSH, and the U.S. Environmental Protection Agency. The National
Health Interview Survey of Cancer Epidemiology and Control (NHIS-CEC)
shows that the prevalence of cigarette smoking continues to decline in
smoking among adults by approximately 0.50 percent per year. Despite
these declines, smoking is responsible for an estimated 390,000 deaths.
Exposure to ETS has been associated with the occurrence of many
diseases, such as lung cancer and heart disease in nonsmokers and low
birthweight in the offspring of nonsmokers.
Table VI-5.--Cases and Cases Avoided of Occupationally Developed Upper
Respiratory Symptoms and Headaches in Buildings With HVAC Systems Over a
Working Lifetime of 45 Years
------------------------------------------------------------------------
Headaches Upper respiratory
-------------------------- symptoms
-------------------------
Cases Cases
Baseline avoided due Baseline avoided due
cases\1\ to IAQ cases\2\ to IAQ
standard standard
------------------------------------------------------------------------
Agriculture,
forestry, fishing.. 14,936 11,948 22,272 17,818
Mining.............. 9,672 7,737 14,423 11,538
Construction........ 87,978 70,383 131,196 104,957
Manufacturing....... 307,650 246,120 458,777 367,021
Transportation...... 182,639 146,111 272,357 217,885
Wholesale and retail
trade.............. 842,666 674,133 1,256,607 1,005,286
Finance, insurance,
real estate........ 387,943 310,354 578,511 462,809
Services............ 1,441,160 1,152,928 2,149,099 1,719,279
Government.......... 507,053 405,643 756,132 604,906
---------------------------------------------------
Total......... 3,781,698 3,025,358 5,639,374 4,511,499
------------------------------------------------------------------------
\1\Based on OSHA estimate of occupational headache risk of 57 per 1,000
employees over a working lifetime of 45 years.
\2\Based on OSHA estimate of occupational upper respiratory symptoms
risk of 85 per 1000 employees over a working lifetime of 45 years.
OSHA estimate for cases prevented through proposed standard is 80
percent.
Source: OSHA, Office of Regulatory Analysis, 1994.
OSHA's estimates are based upon the exposure profile (presented in
Table VI-3) and OSHA's quantitative risk assessment (discussed in
detail in the preamble to the proposal). The OSHA estimates of lifetime
risk of death attributable to exposure to ETS in the workplace range
between 0.4 and 1 for lung cancer and between 7 and 16 for coronary
heart disease, per 1,000 exposed employees. OSHA's estimate of the
attributable risks suggest that all baseline cases of lung cancer and
coronary heart disease will be prevented due to elimination of exposure
of nonsmokers to ETS in the workplace.
Table VI-6 presents estimates of the incidence of work-related
cases avoided of lung cancer and heart disease following either the
banning of smoking in the workplace or limiting smoking to designated
smoking areas. OSHA estimates that approximately between 5,583 and
32,502 cancer deaths and 97,700 to 577,818 coronary heart disease
deaths related to occupational exposure to ETS will be prevented over
the next 45 years. This represents 140 to 722 cancer deaths per year
and 2,094 to 13,001 heart disease deaths per year.
3. Costs Savings
OSHA has also preliminarily determined that the estimated number of
deaths or illnesses prevented understates the actual benefits that
would occur under the proposed standard. Significant additional
economic benefits, apart from the lives saved and illnesses averted,
are anticipated most of which can not be quantified at this time.
Table VI-6.--Cases Avoided of Occupationally Developed Lung Cancer and Coronary Heart Disease Per Employees
Exposed to ETS Over a Working Lifetime of 45 Years
----------------------------------------------------------------------------------------------------------------
Number of non-smoking Coronary heart\1\ Lung cancer\2\ deaths
employees exposed to ETS disease avoided avoided
at work ---------------------------------------------------
----------------------------
Lower bound Upper bound Lower bound Upper bound Lower bound Upper bound
----------------------------------------------------------------------------------------------------------------
Agriculture, forestry, fishing.. 189,606 490,597 1,327 7,850 76 442
Mining.......................... 46,885 121,313 328 1,941 19 109
Construction.................... 654,565 1,693,655 4,582 27,098 262 1,524
Manufacturing................... 2,454,724 6,351,483 17,183 101,624 982 5,716
Transportation.................. 743,623 1,924,089 5,205 30,785 297 1,732
Wholesale and retail trade...... 3,581,778 9,267,685 25,072 148,283 1,433 8,341
Finance, insurance, real estate. 751,493 1,944,454 5,260 31,111 301 1,750
Services........................ 4,079,510 10,555,543 28,557 168,889 1,632 9,500
Government...................... 1,455,027 3,764,816 10,185 60,237 582 3,388
-------------------------------------------------------------------------------
Total..................... 13,957,212 36,113,636 97,700 577,818 5,583 32,502
----------------------------------------------------------------------------------------------------------------
\1\OSHA estimate of occupational coronary heart disease risk for lower and upper bound exposure of 7 to 16 per
1,000 employees over a working life of 45 years.
\2\OSHA estimate of occupational lung cancer risk for lower and upper bound exposure of 0.4 to 0.9 per 1,000
employees over a working life of 45 years.
Source: OSHA, Office of Regulatory Analysis, 1994.
The major forms of these savings are efficiency and productivity
improvements, cost reductions in operations and maintenance, and
reduced incidence of property damage.
(a) Worker Productivity. Productivity gains are realized when less
labor input is required per unit of production. A productivity gain
can, therefore, take the form of either a decrease in the labor hours
needed to maintain the level of production or in the form of increased
production and net income for the establishment.
Productivity losses due to indoor air quality may take several
forms: employees may be less effective because they feel fatigued or
suffer from headaches, eye irritation or other effects. Employees may
accomplish less per hour worked or may spend more time away from their
work location (e.g., taking breaks or walks outdoor). One company
indicated that ``since two of my employees have refrained from smoking
while working . . ., their production has increased and their overall
health seems better to say nothing of the health of those working
around them'' [Ex. 3-192]. In addition to individual productivity, the
quality of indoor air affects organizational productivity such as the
visitor and customer satisfaction, impact on sales and revenue and
repeat customers.
Little data exist on productivity lost due to poor indoor air
quality. A survey of 94 state government office buildings attributes an
average productivity loss of 14 minutes per day or 3.0 percent to poor
indoor air quality [Ex. 3-1075H2]. Based on information gathered from
published resources, the National Energy Management Institute estimates
that there is an increase in productivity of 3.5 percent or
approximately 15 minutes per day for employees in a building that
starts as an unhealthy building, and after IAQ improvements, becomes a
healthy building [Ex. 4-240].
To monetize the productivity improvements resulting from
implementation of the proposed IAQ standard, OSHA multiplied the
average employee payroll by 3.0 percent. As shown in Table VI-7,
monetized productivity improvements is estimated at an annual $15
billion.
OSHA requests any studies relating to productivity effects relevant
to the proposal be submitted.
(b) Property Damage, Maintenance and Cleaning Costs. High
concentrations of contaminants in indoor air can have adverse effects
on materials and equipment. Damages may include corrosion of electronic
components and electrical current leakage, which may eventually result
in equipment malfunction. The costs of materials and equipment damage
by indoor air pollutants include maintenance, repair, and/or
replacement costs resulting from (1) soiling or deterioration of a
materials's appearance, or (2) reduced service life for corroded or
degraded appliances, furnishings, and equipment [Ex. 3-1075H2].
Bell Communications Research reported that the seven regional
telephone companies have spent large sums ranging from $10,000 to
$380,000 per event to replace, clean and repair switches and other
electronic equipment malfunctioning as a result of indoor air
contaminants.
Table VI-7.--Average Annual Cost Savings From Compliance With the IAQ
Proposed Standard Due to Productivity Gains
------------------------------------------------------------------------
Number of Average Annual
employees annual productivity\1\
exposed to payroll per improvements
poor IAQ employee (million)
------------------------------------------------------------------------
Agriculture, forestry,
fishing.................. 83,715 $16,290 $41
Mining.................... 54,210 32,375 53
Construction.............. 493,125 25,286 374
Manufacturing............. 1,724,400 28,376 1,468
Transportation............ 1,023,705 29,655 911
Wholesale and retail trade 4,723,200 20,405 2,891
Finance, insurance, real
estate................... 2,174,445 28,377 1,851
Services.................. 8,077,800 20,811 5,043
Government................ 2,842,068 32,570 2,777
---------------- ----------------
Total............... 21,196,668 15,409
------------------------------------------------------------------------
\1\Based on productivity loss of 3.0 percent.
Sources: U.S. Department of Labor, OSHA, Office of Regulatory Analysis,
1994. U.S. Department of Labor, Bureau of Labor Statistics. Employment
and Wages Annual Averages, 1991. U.S. Bureau of the Census, County
Business Patterns, 1990. January 1993.
Microbial contamination can cause significant damage to buildings
and equipment and there is anecdotal evidence that damage can be so
severe as to make a building unfit for human occupation. OSHA requests
comment on the explicit or implicit rental value affected in buildings
with such problems.
No quantitative estimates are available on the effects of indoor
air on equipment. OSHA requests more information on the effects of
indoor air on materials and equipment.
Indoor air pollutants and in particular ETS contribute to increased
maintenance and cleaning expenses. Increased maintenance and cleaning
costs include: the need to paint walls more frequently, need to clean,
repair and replace furniture, upholstery, carpeting and curtains or
drapes that have cigarette burns and or odors; the need to wash
windows, showcases, and other surfaces that attract ash and dust; and
the need to clean ashtrays. A survey of 2,000 companies that had
adopted no-smoking policies found that 60 percent of these companies
were able to reduce their cleaning and maintenance costs. The savings
have been estimated at about $500 per smoker per year (3).
If establishments decide to ban smoking in the workplace, the
proposed standard would result in virtually eliminating all smoking
related fires, fire fatalities and injuries and direct property damage.
Smoking is a leading cause of fire related fatalities. During the
1980's, the National Fire Protection Association reports that smoking
materials were the cause of over 200,000 fires per year. This resulted
in more than 1,000 civilian fatalities and 3,000 civilian injuries and
approximately $300 million in direct property damage. During the period
of 1989 to 1990, there was an average of $115 million in direct
property damage due to non-residential smoking related fires which
resulted in 36 fatalities and 3,212 injuries. OSHA will further
investigate this issue and requests available data from the public.
E. Technological Feasibility and Compliance Costs
This section presents OSHA's preliminary compliance cost estimates
for the proposed standard on indoor air quality. The cost analysis
covers the major proposed provisions for which data are available.
OSHA requests more information on the consideration for the
relationship of employers and facility owners. The decision to
implement any IAQ improvements will be greatly influenced by the
relationship between employers and landlords. Since changes in building
ventilation systems will be made by landlords, employers may have to
negotiate agreements to ensure that they can meet the OSHA standard. On
the requirement for ETS, landlords in turn are likely to pressure
employers to ban smoking; thereby, forestalling any need for
construction of designated smoking rooms. This section also examines
the technological feasibility of complying with proposed regulation.
1. Technological Feasibility
As interpreted in the Benzene and Cotton Dust cases, the
Occupational Safety and Health Act of 1970 requires that the Agency,
with regard to exposure to toxic substances, is to reduce significant
risk of material health impairment to the extent feasible. Accordingly,
as part of the investigation of the potential effects of the OSHA
proposal, OSHA has examined both the technological and economic
feasibility of the proposal. The economic feasibility assessment
appears later.
OSHA's assessment of the technological feasibility is based on an
examination of what would be required to comply with the proposal,
along with a review of existing practices among affected
establishments. With regard to this proposal, problems with
technological feasibility, by and large, are not evident. Employers are
required to operate their HVAC systems within those parameters
originally designated for the equipment. While many employers may
choose to provide separately ventilated smoking areas, this is an
option, not a requirement, under the proposed regulation. This
technology is widespread currently and can be used to achieve
compliance with the proposed standard.
For example, in some situations, such as hotels and prisons,
employees have as their workplace the residence of others who live in
that building. Restaurants, bars and other ``public'' places expose
employees to customer's tobacco smoke. While it is technologically
feasible to ban smoking in those establishments, there may be other
problems, legal and economic. While it is theoretically possible to
minimize employee exposure to ETS in such a work environment through
special ventilation, in the absence of modified customer service
arrangements, actually eliminating worker exposure to ETS would likely
prove difficult. Consequently, the selection process for one of the
smoking policy alternatives for a particular workplace must consider
both the physical limitations of the building or firm and the
building's use. In addition, some employers may be using their building
facilities for purposes for which the original design did not intend,
and for which retrofitting might prove difficult. OSHA requests comment
on those workplaces for which compliance with the proposed standard
would prove technologically challenging. OSHA will consider additional
information on the ability of firms to implement IAQ programs.
2. Compliance Costs
OSHA estimated preliminary costs of complying with the proposed
standard. OSHA's cost assumptions and methodologies are based on
information available from the rulemaking record. Further detailed
industry analysis will be developed by the Agency.
Table VI-8 contains OSHA's estimates of the annualized first-year
and the annual recurring costs of full compliance with the proposed
rule. The annualized first-year cost of compliance is $1.4 billion. The
cost for eliminating exposure to ETS may range from $0 to $68 million
depending on whether establishments shall ban smoking or allow smoking
in designated areas. OSHA estimated that the annual cost of compliance
with the IAQ standard will be $8.1 billion, of which the most costly
provision will be for the building systems operation and maintenance,
$8.0 billion.
OSHA developed cost estimates for the affected industries using the
following categories of information: (1) Provisions of the proposed
standard requiring activities; (2) the number of potentially affected
buildings, establishments and employees; (3) the percentage of
establishments or buildings in each industry currently in compliance
with each proposed requirement; and (4) the unit costs for bringing
establishments into compliance with the various provisions of the
proposed standard. These four items were combined to produce OSHA's
estimated costs of compliance.
Costs were estimated on an annual basis, with total annual costs
calculated as the sum of annualized initial costs and annual recurring
costs. All capital costs and non-recurring first year costs were
annualized over the service life of the equipment or administrative
activity, at a discount rate of 10 percent.
(a) Developing Indoor Air Quality Compliance Programs. The proposed
standard requires establishments to prepare written operations plans
which would describe information required for the daily operation and
management of the building systems9 and maintenance. The plan
should provide an overview of the building and system, using a short
text description and single-line schematics or as-built construction
documents. The operations information would also describe how to
operate the HVAC systems so that it performs with the reported design
criteria. In addition, the operations information should include: (1)
Special procedures like seasonal start-ups and shutdowns, and (2) a
list of operating performance criteria such as minimum outside air
ventilation rates, potable hot water storage and delivery temperatures,
range of space relative humidities and any space pressurization
requirements, (3) an evaluation of the need to retrofit the HVAC system
when the design occupancy levels are exceeded, and (4) a checklist for
visual inspection of building systems.
---------------------------------------------------------------------------
\9\ Building systems include but are not limited to the heating
and air conditioning (HVAC) system, the potable water systems, the
energy management system and all other systems in a facility which
may impact IAQ.
Table VI-8.--Summary of Compliance Costs for Proposed OSHA Indoor Air
Quality Standard
------------------------------------------------------------------------
Annualized Recurring Annual
cost cost cost
($million) ($million) ($million)
------------------------------------------------------------------------
IAQ written compliance program...... $21.1 -- $21.2
IAQ maintenance and operation
program............................ 1,281.1 $6,697.4 7,978.5
Information and Training:
Maintenance workers............. 0.5 0.8 1.3
All employees................... -- -- --
Controls for environmental tobacco
smoke\1\........................... 0-68.1 -- 0-68.1
-----------------------------------
Total......................... 1,371.0 6,698.2 8,069.1
------------------------------------------------------------------------
\1\Costs incurred are dependent on whether establishments will totally
ban smoking or allow smoking in designated areas.
Source: U.S. Department of Labor, OSHA, Office of Regulatory Analysis,
1994.
The maintenance written plan will also include a description of the
equipment to be maintained and the recommended maintenance procedures
and frequency of performance. Preferably, the plan should contain the
equipment maintenance manuals issued upon completion of facility
construction. For establishments in buildings with natural ventilation,
employers will develop a plan to assure that windows, doors, vents,
stacks and other portals designed or used for natural ventilation are
in operable condition.
The cost associated with compiling such information will vary
depending on the size of the establishment building, the complexity of
the building system, the extent to which such information is already
available, and type of occupancy (e.g., single establishment or multi-
establishment). In some cases some establishments especially the large
ones may already have developed such information. For example, in 1986,
IBM initiated a program by first evaluating building design, operation
and maintenance and as a result, an IAQ program was devised to include:
a model operation/maintenance and IAQ awareness program for building
operation/maintenance personnel, an updated building commissioning
document and appropriate building lease and contracted operation/
maintenance agreements [Ex. 3-904]. There are no data on the number of
establishments with IAQ programs. Based on information in the docket,
OSHA assumed that 95 percent of all establishments are required to
develop the IAQ compliance program information.
In addition, employers are required to: (1) identify a designated
person who is given the responsibility of the IAQ compliance program,
(2) keep written records of employee complaints of building-related
illness and maintenance records, and (3) set up procedures to be
utilized during renovation and modeling to minimize degradation of the
indoor air quality of employees performing such activities and
employees in other areas of the building.
The cost equation for developing the written IAQ compliance
program:
Co=En x Pc x ((Wt x T1)+(Wm x T2))
where
Co=the cost of developing operation and maintenance
information
En=the number of establishments
Pc=the percentage of establishments to develop operation
and maintenance information (95%)
Wt=the technician wage rate ($15.51 hourly compensation
rate)
T1=the technician time required to compile and develop
building system operation and maintenance information (1 hour)
Wm=the managerial wage rate ($30.48 hourly compensation
rate)
T2=the managerial time required to develop some
requirements of the written plan (15 minutes)
As presented in Table VI-9, the one time annualized cost of
compiling and developing the written IAQ compliance program is $21.2
million.
Table VI-9.--Cost of Compliance for Developing a Written IAQ Program
------------------------------------------------------------------------
Annualized
Total no. of first year
establishments cost\1\
($million)
------------------------------------------------------------------------
Agriculture, forestry, fishing............. 260,801 $0.91
Mining..................................... 22,861 0.08
Construction............................... 642,972 2.24
Manufacturing.............................. 389,392 1.35
Transportation............................. 243,769 0.85
Wholesale and retail trade................. 1,929,891 6.71
Finance, insurance, real estate............ 526,378 1.83
Services................................... 1,933,750 6.73
Government................................. 135,496 0.47
----------------------------
Total................................ 6,085,310 21.17
------------------------------------------------------------------------
\1\Based upon 15 minutes of managerial time estimated at $30.48/hr and
one hour of technician time estimated at $15.51/hour. Assumes 5
percent existing compliance. Cost is annualized over 10 years at a 10
percent interest rate.
Source: U.S. Department of Labor, OSHA, Office of Regulatory Analysis,
1994.
(b) IAQ Operation and Maintenance Program. The proposed standard
requires maintenance and inspection of the building system components
that directly affect IAQ. Specifically, the HVAC system should provide
at least the outside air ventilation rate based on actual occupancy,
building code, mechanical code or ventilation code and that carbon
dioxide concentration does not exceed 800 parts per million. In
approximately 500 indoor air quality investigations, NIOSH found that
the primary cause of indoor air quality problems is inadequate
ventilation (52 percent).
Other actions required include: (1) Control of humidity in
buildings with mechanical cooling systems, (2) implementing the use of
general or local exhaust ventilation where maintenance and housekeeping
activities involve use of equipment or products which emit air
contaminants in other areas of the facility, (3) maintain mechanical
equipment rooms and any non-ducted air plenums or chases in a clean
condition.
OSHA recognizes that not every building will have to make all
recommended changes to improve operation and maintenance of the HVAC
system. In the majority of the cases, some improvements can be
accomplished by changing the setting on a control device or centralized
control system. Depending on the condition of the HVAC equipment,
inspection and maintenance may include simple housekeeping of equipment
and air transport pathways and/or catastrophic failure maintenance to
repair/replace failed equipment. Also, there may be cases where a
number of buildings will require major changes in the HVAC system such
as enlarging the size of the outside air intake.
The cost for providing maintenance first requires an estimate of
the number of buildings without regular HVAC maintenance. The 1989
Commercial Buildings Characteristics survey by the Department of Energy
estimates that 46 percent of the buildings have regular HVAC
maintenance. Therefore, the total number of buildings requiring
maintenance is estimated at 2.3 million. OSHA then determined the
number of problem buildings without HVAC maintenance by applying the
OSHA estimate of 30 percent (presented in section B). The number of
problem buildings without HVAC maintenance is estimated at 0.7 million.
In general, the average cost per year to maintain a commercial HVAC
system is a function of a number of factors. These factors include the
type of system, the age of the system, the size of the system, layout
of the system, reliability of the equipment installed. In addition to
the physical characteristics of the system, the cost per year to
maintain the system also depends on the operation of the system, the
maintenance policies of the owner, the skill levels of the operating
engineers and maintenance workers, and whether the maintenance is
carried out by employees of the building owner or is the responsibility
of an outside company.
Bank of America's maintenance costs for its 2,000 worksites
averaged $4 million per year or an average of $2,000 per worksite [Ex.
3-552]. One facility, a high-rise office building, reported an annual
cost of approximately $0.6 million [Ex. 3-448]. DOW Chemical Company's
estimate for ventilation systems maintenance ranges from $0.17 to
$0.25/sq.ft/yr [Ex. 3-502]. Therefore, OSHA used an average of $0.21/
sq.ft/yr to compute the cost of HVAC maintenance.
In addition to regular HVAC maintenance, buildings with known IAQ
problems will require other improvements such as (1) relocating air
intakes and other pathways of building entry to restrict the entry of
outdoor air contaminants, or (2) installing local source capture
exhaust ventilation or substitution within workspaces where air
contaminants are being emitted, or (3) increasing ventilation
effectiveness, or (4) reduce unwanted infiltration, or (5) monitor
outside air quantity to meet ventilation requirements. The National
Energy Management Institute developed a cost model for implementing IAQ
improvements which is based on the distribution of buildings with IAQ
problems by climate zone, building activity and size, and
characteristics of ventilation systems. The average cost to implement
the actions listed above are estimated to be $1.14 per square foot.
These improvements will only be required for the initial year.
The cost equation for implementing the compliance program is as
follows:
Cp=Ms
(Nh x Ca+Np x Ca+Np x Ci x A20)
where
Cp=cost for providing regular HVAC maintenance
Ms=mean square footage per building (14,000)
Nh=number of buildings without HVAC maintenance
Ca=cost per square foot for providing HVAC maintenance
($0.21)
N