PRELIMINARY INITIAL REGULATORY FLEXIBILITY ANALYSIS

OF THE PRELIMINARY DRAFT STANDARD FOR OCCUPATIONAL EXPOSURE TO BERYLLIUM

Office of Regulatory Analysis

Directorate of Evaluation and Analysis

Occupational Safety and Health Administration

U.S. Department of Labor

9/17/2007

Introduction

 tc \l1 "Introduction OSHA has initiated rulemaking that could lead to a
comprehensive health standard to address potential occupational hazards
associated with exposure to beryllium.  In developing a proposed rule,
OSHA wishes to ensure that the regulatory requirements will be effective
in reducing potential risks and will not impose any unnecessary burdens
on society.  To that end, OSHA will invite public comment on the
proposal together with a preliminary analysis of its impact and will
base any final rule on the best available evidence in the entire
rulemaking record.  

OSHA also intends to develop an Initial Regulatory Flexibility Analysis
(IRFA) to accompany the proposal to enable particular attention to be
paid to the potential impacts of the proposal on small businesses.  As
described in the Regulatory Flexibility Act (5 USC 601 et. seq.), an
IRFA must contain the following elements:

1)	a description of the reasons why action by the Agency is being
considered;

2)	a succinct statement of the objectives of, and legal basis for, the
proposed rule;

3)	a description of and, where feasible, an estimate of the number of
small entities to which the proposed rule will apply;

a description of the projected reporting, recordkeeping, and other
compliance requirements of the proposed rule, including an estimate of
the classes of small entities that will be subject to the requirements
and the type of professional skills necessary for preparation of the
report or record;

 SEQ 1_0 \* Arabic \r 5 5 )	an identification, to the extent
practicable, of all relevant Federal rules that may

	duplicate, overlap, or conflict with the proposed rule; and

 SEQ 1_0 \* Arabic \n 6 )	a description of any significant alternatives
to the proposed rule that accomplish the stated objectives of applicable
statutes and that minimize any significant economic impact of the
proposed rule on small entities.

In addition to the information and comments to be solicited from all
interested parties in response to the proposal and its accompanying
documents and analyses, OSHA has a particular interest in identifying
and responding to concerns of potentially affected small businesses and
other small entities at an early stage in the rulemaking.  Thus, as part
of this rulemaking, prior to the publication of the proposal, OSHA is
convening a Small Business Advocacy Review Panel (SBARP) in accordance
with Section 609 of Title 5 of the United States Code.  The SBARP
process enables OSHA, with the assistance of the Chief Counsel for
Advocacy of the Small Business Administration, and with the assistance
of the Office of Information and Regulatory Affairs in the Office of
Management and Budget, to obtain advice and recommendations from
affected small entities about the potential impacts of the proposal.

This Preliminary Initial Regulatory Flexibility Analysis (PIRFA) has
been prepared to aid in the SBARP process.  The IRFA for the proposal
will discuss the Panel’s recommendations and OSHA’s responses to
those recommendations.   The next section of this PIRFA provides a
summary of the proposal.  The remaining sections respond to the
requirements of the Regulatory Flexibility Act.  OSHA emphasizes the
preliminary nature of the draft standard and analyses included in this
document.  At this preliminary stage, it should be noted that there are
numerous uncertainties in the preliminary health risk and economic
analyses.  These uncertainties are noted and discussed in the sections
of the PIRFA titled “Employee Risks of Lung Cancer from Exposure to
Beryllium” and “Employee Risk of Beryllium Sensitization and Chronic
Beryllium Disease.”  OSHA, through the SBARP process and other steps
throughout the rulemaking, will be soliciting comment and additional
data and will continue to refine its analyses in order to fully address
these uncertainties.

Summary of the Preliminary Draft of the Occupational Exposure to
Beryllium Standard

OSHA has initiated rulemaking that may lead to a comprehensive health
standard governing employee exposure to beryllium in general industry,
and in the maritime and the construction sectors.  The Agency has
developed a preliminary draft standard that, if promulgated, would
include a permissible exposure limit (PEL) for beryllium and other
requirements,such as conducting exposure assessment, defining a
regulated area, methods of compliance, use of respiratory protection,
provisions for protective work clothing and equipment, hygiene
facilities and practices, housekeeping, medical surveillance, hazard
communication, and recordkeeping.  The requirements contained in this
preliminary draft standard for general, maritime, and construction
industries are described below:

 

(a)	Scope

This preliminary draft standard would apply to all workplaces where
there are occupational exposures to beryllium in all forms within the
general, maritime, and construction industries.   

(b)	Definitions

The definitions section explains important terms used in the preliminary
draft standard, such as “action level,” “employee exposure,”
“emergency,” “regulated area,” “physician or other licensed
health care professional,” and others.  

(c)	Permissible Exposure Limit (PEL)

g/m3 as an 8-hour TWA, 5 g/m3 as a ceiling value and 25 g/m3 as
a maximum peak above the ceiling value.    The PEL is usually expressed
in terms of X microgram(s) per cubic meter of air (X μg/m3),
calculated as an 8-hour time-weighted average (TWA).    OSHA has not
included a specific PEL in its preliminary draft standard.  The Agency
is currently considering 5 PELs:  2 μg/m3, 1 μg/m3, 0.5 μg/m3, 0.2
μg/m3, and 0.1 μg/m3.   

The health risk data and analyses published in the scientific literature
indicate that exposure over a working lifetime to OSHA’s current PEL
of 2 μg/m3 may not adequately protect employees against the risk of
chronic beryllium disease (CBD). (See the section of the PIRFA entitled
“Reasons Why Action By the Agency Is Being Considered.”) Although
OSHA is still evaluating the scientific evidence underlying these risk
analyses, OSHA has made a preliminary decision not to consider PELs
higher than 2 μg/m3.

OSHA is also considering setting an action level.  The action level is
typically established as half of the PEL, thus possible action levels
under consideration include, 1 μg/m3, 0.5 μg/m3, 0.25 μg/m3, 0.1, and
0.05 μg/m3 .  Generally, an action level is intended to assure that
employers with workplace exposures just below the PEL continue to act to
ensure that such exposures do not exceed the PEL 

	While not in the preliminary draft standard, OSHA is also considering
several regulatory alternatives with regard to the health-based
permissible exposure limits, including a PEL-based on respirable mass
concentration rather than total mass concentration and/or a short-term
exposure limit.

(d)  Exposure Assessment

This paragraph of the preliminary standard covers provisions for
conducting an initial exposure assessment, performing periodic and
additional monitoring, and the observation of monitoring. 

Each employer is required to conduct an assessment of his or her work
site to determine if employees are exposed to levels of beryllium at or
above the action level or are likely to be exposed to beryllium by skin
or eye contact through routine handling of beryllium powders or dusts or
contact with contaminated surfaces.   The purpose of this assessment is
to determine whether or not certain provisions of the standard such as
medical surveillance, periodic monitoring, training or respiratory
protection would be needed.  Airborne exposures would be determined by
personal breathing air zone samples.  Representative sampling is allowed
to assess employee exposures. Eye and skin contact would be determined
by using professional judgment on the degree to which certain jobs or
work practices could reasonably be anticipated to involve routine
contact with beryllium powders or dusts or contact with contaminated
surfaces.  In cases where the employer has conducted exposure monitoring
within the past 12 months, and has satisfied all other requirements
within this section, the results of previous monitoring may be used to
satisfy the initial monitoring provision.  In addition, in cases where
the employer has objective data demonstrating that a material containing
beryllium on a specific process, operation, or activity involving
beryllium cannot release dust, fumes, or mist of beryllium in
concentrations that exceed the action level, the employer may rely upon
such data to satisfy the initial exposure monitoring provision of the
section.  The data must reflect work place conditions closely resembling
the processes, types of material, control methods, work practices, and
environmental conditions in the employer’s current operations. 

Periodic monitoring of the work site is required if the initial
monitoring indicates that employee exposures are at or above the action
level or PEL.  For airborne levels above the action level but below the
PEL, monitoring is required every 6 months.  For airborne levels above
the PEL, monitoring is required every 3 months.  However, if the
periodic monitoring indicates that exposures are below the action level
and the employer can confirm the results by two consecutive air samples
taken seven days apart, then the employer may discontinue monitoring at
the work site.  

Additional monitoring is required when:  (1) there is a change in
production process, raw materials, equipment, personnel, work practices,
or control methods that may result in new or additional exposures to
beryllium, or (2) when the employer has reason to believe that new or
additional exposures have occurred at the work site. 

Employers are required to notify employees within 15 days of the receipt
of the results of any monitoring performed under this standard.  This
notification may be made individually in writing or by posting the
results in a location that is accessible to all affected employees. 
Where exposure levels are above the PEL, the employer is required to
describe in the employees’ written notification the corrective action
taken to reduce the exposure levels to the PEL or lower.   

Employers are required to use a method of monitoring that produces
results that are accurate to within a statistical confidence level of 95
percent.  The analytical method used should measure airborne beryllium
to within an accuracy of plus or minus 25 percent at or above the action
level.  

The employers are required to provide affected employees or their
designated representatives an opportunity to observe any monitoring of
employee exposure to beryllium.  Plus, the employer is required to
provide PPE to all observers during the observation periods.

 

(e) Regulated Areas

To minimize any unnecessary employee exposures, the preliminary draft
standard requires employers to establish a regulated area wherever an
employee’s exposure to airborne concentrations of beryllium is or can
be expected to be above the PEL.  The regulated area would be
demarcated, and the employer would be required to limit entry to
employees working in the area and other authorized personnel (e.g.,
designated employee representative for purpose of observing monitoring;
and NIOSH or OSHA representatives).

(f)  Methods of Compliance

If an operation has exposures above the action level or there is
potential for routine handling of beryllium powders or dusts on contact
with contaminated surfaces, the preliminary draft standard would require
that the employer develop a written exposure control plan.  The purpose
of the plan is to have procedures in place to minimize employee exposure
to airborne beryllium and exposure to eye or skin contact with
beryllium.  The written exposure control plan must contain at least the
following elements: (1) an inventory of locations of beryllium
operations and other locations of employees who are, or who are likely
to be exposed at these locations; (2) procedures for minimizing
cross-contamination; (3) procedures for cleaning beryllium-contaminated
products or equipment before release from beryllium-contaminated work
areas; (4) procedures for cleaning beryllium-contaminated work surfaces;
and (5) procedures for removal, storage and cleaning or disposal of
personal protective equipment.  Each employer is required to ensure a
copy of the written exposure control plan is accessible (i.e.,
available) to all employees. 

The preliminary draft standard would require each employer to use
engineering controls and work practices to maintain exposures to levels
at or below the PEL, unless the employer can demonstrate that such
controls are not feasible.  Wherever engineering and work practice
controls are not sufficient to reduce employee exposure to the PEL, the
employer must use these control methods to reduce employee exposure to
the lowest level achievable and then supplement them with the use of
respiratory protection.   

Employers are prohibited from rotating employees to different jobs to
achieve compliance with the PEL.

(g) Respiratory Protection

The preliminary draft standard makes reference to OSHA’s Respiratory
Protection Standard (29 CFR 1910.134), which employers must comply with
when employees must use respirators for protection against beryllium
exposure.  The Respiratory Protection Standard includes provisions for
written procedures for the proper selection, use, cleaning, storage, and
maintenance of respirators.

(h)  Protective Work Clothing and Equipment

When employees are exposed to beryllium above the PEL or there is
anticipated skin exposure from routine handling of beryllium powder or
dusts or contact with contaminated surfaces, employers are required to
provide appropriate personal protective clothing and equipment and
ensure that employees use it.  The employer is responsible for ensuring
that employees remove all protective clothing and equipment contaminated
with beryllium in change rooms at the end of the work shift or at the
completion of their tasks (whichever comes first), and leave their
contaminated clothing at the work site.  The preliminary draft standard
requires that all employers provide the protective clothing and
equipment at no cost to the employee.  Employers are required to ensure
that all employees use the protective clothing and equipment in
accordance with the written exposure control plan. 

The preliminary draft standard requires employers to clean, launder,
repair, or replace work clothing and equipment as needed to maintain its
effectiveness.  Employers are prohibited from removing beryllium
contamination by blowing air on or shaking the clothing.  Employers must
ensure that the clothing and equipment is stored and transported in
sealed, impermeable bags or in other closed, impermeable containers. 
Bags or containers of contaminated protective clothing or equipment that
are removed from change rooms for laundering, cleaning, maintenance, or
disposal must be labeled in accordance with paragraph (l).

Employers are also required to inform persons responsible for laundering
contaminated work clothing of the hazards associated with exposure to
beryllium.  The laundering and cleaning must be performed in such a
manner to minimize eye or skin contact and to prevent release of
airborne beryllium in excess of PEL.

(i)  Hygiene Areas and Practices

The preliminary draft standard requires employers to provide change
rooms, showers, handwashing facilities, and eating and drinking
facilities in conformance with 29 CFR 1910.141 if employees’ exposures
are above the PEL or if there is anticipated skin exposure. 

The preliminary draft standard requires that employers provide a clean
change room equipped with separate storage facilities for protective
clothing and equipment and for street clothes.  These facilities must
prevent cross-contamination.   

The employer must also provide accessible hand washing facilities and
ensure that employees use them when necessary.  Employees who have eye
or skin contact from beryllium exposure must wash their hands and face
at the end of the work shift and prior to eating, drinking, smoking,
chewing tobacco or gum, applying cosmetics, or using the toilet.  All
employees exposed above the PEL are required to shower at the end of
their work shift.

Where the employer allows employees to consume food or beverages at the
work site, the employer is responsible for ensuring that all eating and
drinking area surfaces are maintained and cleaned enough that no
employee is exposed at any time to beryllium to levels at or above the
action level.  Employees are prohibited from entering the eating or
drinking area while wearing contaminated protective clothing or
equipment.  The employer is also required to ensure that employees do
not eat, drink, smoke, chew tobacco or gum or apply cosmetics in
regulated areas, or in areas where skin or eye contact is likely. 

 (j) Housekeeping

The preliminary draft standard requires that employers keep all surfaces
as clean as practicable.  The standard requires the clean-up of
accumulations of beryllium on surfaces.  All spills and releases of
beryllium-containing materials are to be cleaned up promptly and in
accordance with procedures developed under the written exposure control
plan.   Surfaces must be cleaned with a HEPA – filtered vacuum cleaner
or equally effective filtration or dust collection method.   Shoveling,
sweeping, and brushing may be used only where HEPA-filtered vacuuming or
other methods that minimize the likelihood of exposure to beryllium have
been tried and found not to be effective.  The use of compressed air is
prohibited, unless the compressed air is used in conjunction with a
ventilation system designed to capture the dust cloud created by the
compressed air.   Cleaning equipment must be handled in a manner that
minimizes the re-entry of beryllium into the work site.  Finally, all
waste, scrap, debris, and any other materials contaminated with
beryllium must be disposed of in sealed, impermeable bags or other
closed, impermeable containers.  These bags or containers must be
labeled in accordance with paragraph (l) (Communication of beryllium
hazards) of this section.

(k) Medical Surveillance

The preliminary draft standard requires that the employer must make
medical surveillance available at no cost to the employee and at a
reasonable time and place for: (1) those employees who are
occupationally exposed to beryllium to levels at or above the action
level, or who can be expected to have skin exposure from routine
handling of dusts or powders or contact with contaminated surfaces; (2)
those employees exposed in emergency situations such as spills, or (3)
those experiencing signs or symptoms of the adverse health effects
associated with beryllium exposure.  All medical examinations are to be
performed by or under the supervision of a physician or licensed health
care professional (PLHCP).   As in the case of most OSHA standards,
employees may decline to participate in the employer-provided medical
surveillance programs required by the draft preliminary standard (or
medical tests required under medical surveillance).  The obligation
under the draft preliminary standard is that the employer offers the
medical surveillance at a reasonable time and place and pays for it.  It
is the employer’s prerogative (not a requirement of the draft
preliminary standard) as to whether he or she chooses to make
participation in their medical surveillance program a condition of
employment.  

All medical examinations must be given within 30 days after initial
assignment, unless the employee can demonstrate that he or she has
already had a medical examination for beryllium exposure within the past
12 months.  Otherwise, medical examinations must be given: (1) annually,
(2) 30 days after a written medical opinion by the PLHCP to have
additional examinations, (3) when the employee shows signs and symptoms
of adverse health effects associated with beryllium exposure, (4) within
30 days after exposure during an emergency situation of an uncontrolled
release of beryllium, or (5) at the termination of employment unless the
last examination was less than six months prior to the date of
termination.  

	The medical examinations must consist of a physical examination of the
skin and respiratory tract, a beryllium lymphocyte proliferation test
(BeLPT) for those employees who have not previously been confirmed
positive, and any additional tests deemed appropriate by the examining
PLHCP.   At the present time, alternative tests to the BeLPT are under
research and development but none has yet undergone sufficient
evaluation and validation to be considered acceptable for widespread use
as a surveillance method.  The medical examination also must consist of
an evaluation of the individual’s medical and work history, with
emphasis on past, present, and anticipated future exposure to beryllium.
 The PLHCP must obtain any history of respiratory system dysfunction and
any past BeLPT results.

Employers are required to ensure the PLHCP has a copy of the beryllium
standard.  The PLHCP is also required to be given information on the
employee’s work duties, the levels of exposure, a description of all
PPE used including duration of use, and previous medical records that
are in control of the employer, so that the PLHCP will have appropriate
information for making necessary determinations to place any limitations
on the employee’s exposure to beryllium or the use of PPE, such as
respirators. 

The employer is required to obtain a signed, written medical opinion
from the PLHCP within 30 days for each medical examination performed.  
The written opinion must explain: (1) whether the employee has any
detected medical condition that would place the employee at increased
risk of developing health problems from occupational exposure to
beryllium; (2) any limitations that the employee may have in regard to
use of respirators and other PPE; and include (3) a statement that the
results of the medical examination were explained to the employee.  The
preliminary draft standard requires that the employer provide a copy of
the written medical opinion to the employee within two weeks after
receiving the written medical reports.

(l) Communication of Beryllium Hazards to Employees

This paragraph of the preliminary draft standard is a cross-reference to
OSHA’s Hazard Communication Standard (29 CFR 1910.1200) and requires
that employers include beryllium in their hazard communication program
covering labels, material safety data sheets, and information and
training.  The preliminary draft standard requires employers to display
legible and readily visible warning signs at all approaches to regulated
areas and labels on all bags or containers of contaminated clothing and
equipment as well as containers of waste, scrap, debris, and any other
materials contaminated with beryllium.  Warning signs and labels must
indicate that beryllium exposure is a respiratory and cancer hazard,
that only authorized personnel are permitted in the area, and that the
use of respiratory protection is required in the area.   

The preliminary draft standard requires that training be tailored
specifically to operations at the work site and is designed to provide
important information on operations that could result in exposures
exceeding the PEL.  Additionally, the training must be given in a way
that is understandable to the employee.  The employer must ensure that
the employee can demonstrate the knowledge of the health hazards
associated with beryllium exposure, including the signs and symptoms of
chronic beryllium disease, with an explanation of the role of
sensitization in developing chronic beryllium disease, principles of
safe handling of beryllium materials, the employer’s written exposure
control plan, and methods used to minimize exposure.   

The preliminary draft standard requires that training must be provided
prior to an initial assignment to a job involving potential exposure to
beryllium, unless the employer can demonstrate that the employee
received training within the last 12 months and the employee can
demonstrate knowledge of those elements.  Additional training is
required when (1) such training is necessary to ensure that each
employee maintains an understanding of the safe use and handling of
materials containing beryllium and (2) workplace changes result in an
increase in employee exposure to beryllium or where those exposures
exceed or are expected to exceed the action level or result in skin or
eye contact. 

 (m) Recordkeeping

The employer is responsible for maintaining an accurate record of
employee exposure measurements, historical monitoring data, objective
data, employee medical surveillance information, and employee training. 
Exposure and health records must be maintained and made available in
accordance with 29 CFR 1910.1020.

As for records of exposure measurements, the preliminary draft standard
requires that the records include: the date of measurement for each
sample taken; the operation involving exposure to beryllium that is
being monitored; the sampling and analytical methods used and evidence
of their accuracy; number, duration and results of the samples; type of
PPE used; and name, social security number and job classification of all
employees represented by the monitoring, indicating which employees were
actually monitored.

The preliminary draft standard requires that the employer keep accurate
records of historical monitoring data.  The information maintained
should: (1) come from data that were collected using methods that meet
the accuracy requirements of paragraph (d) of this section; (2) include
the processes and work practices that were used when the historical
monitoring data were obtained, which are essentially the same as those
to be used during the job for which initial monitoring will not be
performed; (3) include the characteristics of the beryllium-containing
material being handled when the historical monitoring data were
obtained, which are the same as those on the job for which initial
monitoring will not be performed; (4) include environmental conditions
prevailing when the historical monitoring data were obtained, which are
the same as those on the job for which initial monitoring will not be
performed; and (5) include other relevant data to the operations,
materials, processing, or employee exposures covered by the exception. 

The preliminary draft standard requires that employers maintain accurate
records of objective data.  The records maintained shall include
information on the beryllium – containing material in question; the
source of the objective data; the test protocol and results of testing,
or analysis of the material for the release of beryllium; a description
of the operation exempted from initial monitoring and how the data
support the exemption; and other relevant data to the operations,
materials, processing, or employee exposures covered by the exemption. 

The preliminary draft standard requires that employers establish and
maintain accurate records of each employee covered by medical
surveillance.  The information maintained shall include:  name, social
security number, job classification; a copy of the PLHCP’s written
opinions; and a copy of the information provided to the PLHCP as
required under section (k) of this preliminary draft standard.  

The preliminary draft standard requires that employers maintain training
records, including the identity of the individuals trained and the date
of the training.  Training records must be maintained for 3 years.

(n)  Dates

Employers are required to comply with effective dates and start-up dates
set forth in the standard for certain provisions.  The effective date is
typically set for 60 days after publication in the Federal Register. 
The start-up dates are typically set at 90 days after the effective
dates, except in provisions regarding change rooms required by paragraph
(i) of this section which should be implemented no later than 1 year
after the effective date and engineering controls required by paragraph
(f) of this section which should be implemented no later than 2 years
after the effective date.  

Reasons Why Action by the Agency is Being Considered

When establishing the need for an occupational safety and health
standard, OSHA must evaluate available health and safety data to
determine whether or not employees will suffer a material impairment of
their health or functional capacity as a result of being exposed to a
particular safety or health hazard.   The Occupational Safety and Health
Act (OSH Act) directs OSHA to set the standard “. . . which 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 even if such employee has regular exposure
to the hazard dealt with by such standard for the period of his working
life” (Section 6(b)(5) of the OSH Act).   The Supreme Court, in
reviewing previous OSHA standards, has also directed the Agency to make
a determination that significant risks are present and can be eliminated
or lessened by a change in practices before promulgating any health or
safety standard  (Industrial Union Dept. v. American Petroleum Inst.,
448 U.S. 607, 655 (1980)) [plurality opinion].  While observing that the
determination of what constitutes a “significant risk” must be
“based largely on policy considerations,” the Court also noted that
in non-binding dicta that a “reasonable person” standard could be an
appropriate measure of significant risk.  

OSHA makes its material impairment and risk determinations by first
evaluating the available data to identify and characterize hazards to
which employees are exposed in the workplace that are likely to induce
material impairments of their health or functional capacity.   The
Agency looks at a broad array of scientific data and assesses the
overall weight of evidence in making its health and safety
determinations.   In the next step, the Agency looks at the overall
quality of the data to identify studies or other data that are useful in
making quantitative estimates of the risk of those impairments of health
among exposed employees over their working life (as mandated by the OSH
Act).   While many studies may add to the overall weight of evidence in
making determinations about material impairments of health, often only
select data have suitable information for making quantitative estimates
of risk.    In the case of health risk analyses, the quantitative
estimation of risk often involves the use of dose-response mathematical
models.  This is a common approach used in the field of health risk
assessment that 

allows the Agency to extrapolate scientifically observable data, in
humans or animals, to a variety of exposure scenarios that may be of
concern to exposed employees.

Health Effects

 as its current permissible exposure limit (PEL) for beryllium and
beryllium compounds.  This enforceable workplace standard sets an eight-
hour TWA limit at 2 μg/m3, a 5 μg/m3 ceiling limit, and a maximum peak
of  25 μg/m3 above the ceiling that is never to be exceeded for up to
30 minutes in an eight- hour work shift.  Since the enactment of the
above OELs for beryllium, cases of ABD have been rare.

CBD is a hypersensitivity or allergic reaction to beryllium that leads
to a chronic inflammatory disease of the lungs.  Unlike ABD, it takes
months to years after initial beryllium exposure before signs and
symptoms of CBD occur.  Removing the employee from the beryllium source
does not always lead to recovery.  In some cases CBD continues to
progress following removal from beryllium exposure.  CBD is not a
chemical pneumonitis but an immune-mediated granulomatous lung disease. 
For CBD to occur, an employee must first become sensitized [i.e.,
allergic] to beryllium.  Once an employee is sensitized, inhaled
beryllium that deposits and persists in the lung may trigger a
cell-mediated immune response [i.e., hypersensitivity reaction] that
results in the formation of a type of lung scarring known as a
granuloma.  The granuloma consists of a localized mass of immune and
inflammatory cells that have formed around a beryllium particle lodged
in the interstitum of the lung.  With time, the granulomas spread and
can lead to chronic cough, shortness of breath especially upon exertion,
fatigue, abnormal pulmonary function, and lung fibrosis.  While CBD
primarily affects the lungs, it can also involve other organs such as
the liver, skin, spleen, and kidneys.  Some studies demonstrate that
sensitization and CBD cases have occurred in workplaces that use a wide
range of beryllium compounds, including several beryllium salts, the
refined beryllium metal, beryllium oxide, and the beryllium alloys. 
While water-soluble and insoluble beryllium compounds have the potential
to cause sensitization, it has been suggested that CBD is the result of
occupational exposure to beryllium oxide and other water-insoluble
berylliums rather than exposure to water-soluble beryllium or beryllium
ores.  However, there are inadequate data, at this time, on employees
selectively exposed to specific beryllium compounds to eliminate a
potential CBD concern for any particular form of this metal.
 Regardless of the type of beryllium compound, the inhaled beryllium
must contain particulates that are small enough to reach the
bronchoalveolar region of the lung where the disease takes place.

	Inhalation of respirable beryllium may be only one of several factors
that determine whether an employee becomes sensitized to beryllium.  It
has been shown by some studies that employees with genes that code for
specific protein molecules on the surface of their immune cells are at
greater risk of becoming sensitized to beryllium and developing CBD
(McCanlies et al., 2004).  Some recent research suggests that skin
exposure to small beryllium particles or beryllium-containing solutions
may also lead to sensitization (Tinkle et al., 2003).  These additional
risk factors may explain why some individuals with seemingly brief, low
level exposure to airborne beryllium become sensitized while others with
long-term high exposures do not.  Some studies suggest that even though
employees sensitized to beryllium do not exhibit clinical symptoms,
their immune function is altered such that inhalation to previously safe
levels of beryllium can now trigger serious lung disease (Kreiss et al.,
1996; Kreiss et al., 1997; Kelleher et al., 2001 and Rossman, 2001).

	In the 1980s, the laboratory blood test known as the BeLPT was
developed.  The test substantially improved identification of beryllium
sensitized individuals and may provide an opportunity to diagnose CBD at
an early stage.  The BeLPT measures the ability of immune cells (i.e.,
peripheral blood lymphocytes) to react with beryllium.  It has been
reported that the BeLPT can identify 70 to 90 percent of those
sensitized with a high (approximately one to three percent false
positives) specificity (Newman et al., 2001; Stange et al., 2004). An
employee with an abnormal (i.e., the individual is sensitized) BeLPT can
undergo fiberoptic bronchoscopy to obtain a lung biopsy sample from
which granulomatous lung inflammation can be pathologically observed
prior to the onset of symptoms.  The combination of a confirmed abnormal
BeLPT (that is, a second abnormal result from the BeLPT) and microscopic
evidence of granuloma formation is considered diagnostic for CBD.  The
BeLPT assists in differentiating CBD from other granulomatous lung
diseases (e.g., sarcoidosis) with similar lung pathology.  This
pre-clinical diagnostic tool provides opportunities for early
intervention that did not exist when diagnosis relied on clinical
symptoms, chest x-ray and abnormal pulmonary function.

The BeLPT/lung biopsy diagnostic approach has been utilized in several
occupational surveys and surveillance programs over the last fifteen
years.  The findings expanded scientific awareness of sensitization and
CBD prevalence among beryllium employees and provided a better
understanding of its work-related risk factors.  Some of the more
informative studies come from nuclear weapons facilities operated by the
Department of Energy (Viet et al., 2000; Strange et al., 2001; DOE/HSS
Report, 2006), a beryllium ceramics plant in Arizona (Kreiss et al.,
1996; Hennenberger et al., 2001; Cummings et al., 2007), a beryllium
production plant in Ohio (Kreiss et al., 1997; Kent et al., 2001), a
beryllium machining facility in Alabama (Kelleher et al., 2001; Madl et
al., 2007), a beryllium alloy plant (Shuler et al., 2005), and another
beryllium processing plant (Rosenman et al., 2005) both in Pennsylvania.
 The prevalence of beryllium sensitization from these surveyed
workforces generally ranged from 1 to 10 percent with a prevalence of
CBD from 0.6 to 8 percent.  

In most of the surveys discussed above, 36-100 percent of those workers
who

initially tested positive with the BeLPT were diagnosed with CBD upon
pathological evaluation.  Most of these CBD cases had worked four to
ten years on the job, although some were diagnosed within several months
of employment. Surveys that found a high proportion (e.g., >50 percent)
of CBD among the sensitized employees were from facilities with a large
number of employees who had been exposed to respirable beryllium for
many years.  It has been estimated from ongoing surveillance of
sensitized individuals with an average follow-up time of 4.5 years that
37 percent of beryllium-exposed employees were estimated to progress to
CBD (Newman, 2005).  Another study of nuclear weapons facility
employees enrolled in an ongoing medical surveillance program found that
only about 20 percent of sensitized individuals employed less than five
years eventually were diagnosed with CBD while 40 percent of sensitized
employees employed ten years or more developed CBD (Stange et al.,
2001). This observation along with the study findings that CBD
prevalence increases with cumulative exposure (described below) suggest
that sensitized employees who acquire a higher lung burden of beryllium
may be at greater risk of developing CBD than sensitized employees who
have lesser amounts of beryllium in their lungs.

There are some uncertainties to this approach.  For example, there are a
number of reasons to believe the occupational surveys above may
underrepresent the true rates of beryllium sensitization and CBD in
these work forces.   Case definition of sensitization required that an
abnormal BeLPT result be confirmed on a repeat test.  Some sensitized
employees did not receive clinical evaluation (e.g., 15% in Kelleher et
al. 2001; 7% in Henneberger et al. 2001; 6% in Schuler et al. 2005; 6%
in Kreiss et al. 1997).  The biopsy procedure does not always detect
granuloma formation unless repeated many times.  Therefore, it is
possible that some sensitization and CBD cases in the survey population
were missed during the medical evaluation.  The cross-sectional nature
of the surveys does not account for CBD cases that may have left the
workforce due to illness (i.e., healthy employee effect) or sensitized
employees who progress to CBD with the passage of time.  If significant,
these factors may also contribute to underestimation of disease rate.

The greatest prevalence of sensitization and CBD were reported for
production processes that involve heating beryllium metal (e.g., furnace
operations, hot wire pickling and annealing) or generating and handling
beryllium powder (e.g., machining, forming, firing).  For example,
nearly 15 percent of machinists at the Arizona beryllium ceramics plant
were sensitized compared to just 1 percent of workers who never worked
in machining (Kreiss et al., 1996).  A low prevalence of sensitization
and CBD was reported among current employees at the Department of Energy
(DOE) clean-up sites where beryllium was once used in the production of
nuclear weapons (DOE/OSS, 2006).  These sites have been subject to the
DOE CBD prevention programs since 1999.  While the prevalence of
sensitization and CBD in non-production jobs was less, cases of CBD were
found among secretaries, office employees, and security guards.  CBD
cases have also been reported in downstream uses of beryllium such as
dental laboratories and metal recycling.

Most, but not all, occupational studies suggest that the prevalence of
CBD increases with increasing exposure to airborne beryllium.  The
largest of these was a case-control study of employees at a nuclear
weapons facility that examined 50 CBD and 74 sensitization cases
paired-matched with disease-free employees (Viet et al., 2000).  There
was a statistically significant positive relationship between mean and
cumulative beryllium exposure and the prevalence of CBD, but not for the
prevalence of sensitization.  The exposure estimates were based on
general area sampling measured some distance away from the breathing
zone of the employee.  A smaller case-control study of 20 sensitized
employees from a machining plant with (13 CBD cases) and without CBD
reported that these employees had consistently higher mean and median
cumulative exposures compared to non-sensitized employees from the same
facility (Kelleher et al., 2001).  Employees with higher median
exposures were about twice as likely to be beryllium sensitized or have
CBD as employees with low median exposures.  A cross-sectional survey of
employees at a beryllium ceramics plant identified 15 sensitized
employees with (8 CBD cases) or without CBD (Hennenberger et al., 2001).
 The study found that employees with the highest airborne exposures
were, in general, more likely to be beryllium sensitized or have CBD. 
The exposure-related findings in the latter two studies did not meet
statistical significance perhaps due, in part, to the small number of
cases. A comprehensive preventive program at the ceramics plant that
included expanded respiratory protection, added emphasis on skin
exposure, and improved control of beryllium dust migration substantially
reduced the rate of beryllium sensitization among new hires (Cummings et
al., 2007).

 be retained in the bronchoalveolar region of the lung where CBD occurs.
 In addition, the very smallest ultrafine particles (<0.5 μm in
diameter) have high surface area per mass and, thus, may enhance the
opportunity for surface contact with pulmonary immune and inflammatory
cells that drive the CBD process.

ble (e.g., particles <10 μm in diameter) and alveolar-deposited (e.g.
particles <1 μm in diameter) beryllium mass and particle number
concentrations, as collected by a general area impactor device, were
measured at the beryllium metal production furnaces rather than the
beryllium alloy furnaces (Kent et al., 2001; McCawley et al., 2001).  A
statistically significant linear trend was reported between the above
alveolar deposited particle mass concentration and prevalence of CBD and
sensitization in the furnace production areas.  On the other hand, a
linear trend was not found for CBD and sensitization prevalence and
total beryllium mass concentration.  The authors concluded that these
findings suggest that alveolar deposited particles may be a more
relevant exposure metric for predicting the incidence of CBD or
sensitization than the total mass concentration of airborne beryllium.  
              

Several epidemiological cohort studies have reported excess lung cancer
mortality among employees employed in U.S. beryllium production and
processing plants during the 1930s to 1960s.  The largest and most
comprehensive study investigated the mortality experience of over 9,000
employees employed in seven different beryllium processing plants over a
30 year period (Ward et. al., 1992).  The employees at the two oldest
facilities (i.e., Lorain, OH and Reading, PA) were found to have
significant excess lung cancer mortality relative to the U.S.
population.  These two plants were believed to have the highest exposure
levels to beryllium.  A different analysis of the lung cancer mortality
in this cohort using various local reference populations and alternate
adjustments for smoking generally found smaller, non-significant, excess
mortality among the beryllium employees (Levy et al., 2002).  All the
cohort studies are limited by a lack of job history and air monitoring
data that would allow investigation of mortality trends with beryllium
exposure.  The majority of employees at the Lorain and Reading
facilities were employed for a relatively short period of less than one
year.

 lung cancer cases were estimated to be an order of magnitude above the
current 8 hour TWA PEL (2 μg/m3) and roughly two orders of magnitude
higher than the typical air levels in workplaces where sensitization and
pathological evidence of CBD have been observed.                    

Several studies in which large amounts of beryllium were inhaled or
instilled in the respiratory tracts of experimental animals have also
resulted in an increased incidence of lung tumors (IARC, v. 58 pp.
76-84; Gordon and Bowser, 2003).  A wide range of beryllium compounds
were administered in these studies including soluble beryllium sulfate,
insoluble beryllium hydroxide, beryllium metal, and certain beryllium
ores, oxides, and alloys.  The animal evidence supports the
epidemiological findings of excess lung cancer mortality in
beryllium-exposed employees.   However, the study protocols employed a
minimum number of dose groups for a less than full lifetime exposure
duration and, therefore, are not suitable for quantitative risk
assessment.  It is not clearly understood how beryllium may lead to
tumor formation at the cellular level.  Soluble beryllium salts are
generally not mutagenic in bacterial test systems, but have been shown
to morphologically transform normal mammalian cells into pre-malignant
cells in culture.  Other forms of beryllium have not been thoroughly
tested for genotoxicity.

OSHA has preliminarily determined that the weight of evidence indicates
that beryllium compounds should be regarded as potential occupational
lung carcinogens.  Other scientific organizations, such as the
International Agency for Research on Cancer (IARC), the National
Toxicology Program (NTP), the U.S. Environmental Protection Agency
(EPA), the National Institute for Occupational Safety and Health
(NIOSH), and the American Conference of Governmental Industrial
Hygienists (ACGIH) have reached similar conclusions with respect to the
carcinogenicity of beryllium.  Risk models that effectively analyze lung
cancer mortality from employee exposure to high airborne beryllium
concentrations that exceed the present OSHA PEL and reliably predict
lung cancer incidence at the much lower exposure limits under current
consideration are needed to evaluate the health risk at lower levels.   
            

Employee Risk of Lung Cancer from Exposure to Beryllium

As discussed in the previous section, a variety of epidemiological
studies and animal experiments suggest the potential carcinogenicity of
beryllium and beryllium compounds.  OSHA has reviewed the available
literature to determine which studies may provide a basis for
quantitative assessment of the cancer risk associated with occupational
beryllium exposure.  One study following a cohort of employees at a
Reading, Pennsylvania beryllium processing facility provided suitable
information to support quantitative risk assessment, and was used by
OSHA’s contractor to generate estimates of excess lung cancer risk
associated with a working lifetime of exposure at the current PEL of 2
g/m3 (8-hour TWA).  These risk estimates are presented and discussed
here, together with sources of uncertainty in the estimates and limited
evidence from a second study regarding the carcinogenicity of beryllium
at low exposure levels.  Additional information on analyses performed by
OSHA’s contractor is available in the attached contract report titled
“Preliminary Quantitative Risk Assessment for Beryllium” (docket #
for contractor report).

Study selection

Several epidemiological studies followed cohorts of employees at two
beryllium processing plants located in Reading, Pennsylvania (Mancuso
and El-Attar, 1969; Mancuso, 1970, 1979, 1980; Wagoner et al., 1980;
Sanderson et al., 2001) and Lorain, Ohio (Mancuso and El-Attar, 1969;
Mancuso, 1970, 1979, 1980), while others used employees enrolled in the
Beryllium Case Registry (Infante et al., 1980; Steenland and Ward, 1991)
or assembled cohorts from multiple companies and plant sites (Bayliss et
al 1971, Ward et al 1992).  Of these studies, only Sanderson et al.
(2001) provides a reconstruction of employees’ historical exposures on
which to base a quantitative exposure-response assessment.  The
Sanderson et al. study also benefits from a minimum of 20 years’
followup for all employees and careful consideration of occupational
exposures other than beryllium; its case-control study design does not
require an external reference population, which reduces the likelihood
that the results of the analysis will be seriously impacted by the lack
of information on smoking. Limitations of the study include a high
proportion of subjects with short durations of exposure (median < 1
year) or with exposures to very high concentrations of beryllium,
typical of the 1940s-50s working conditions experienced by most cohort
members.  

Although no other study provides sufficient exposure information to
support quantitative analysis, the human and animal studies described in
the previous section provide complementary information to the results of
the Sanderson et al. analysis.  In particular, the study conducted by
Ward et al. (1992) assesses lung cancer mortality across seven U.S.
beryllium production facilities, including the Reading, Pennsylvania
plant studied by Sanderson et al (2001), where over 3,500 of the Ward et
al. study’s 9,225 cohort members worked.  The Ward et al. study
presents a variety of analyses that describe the risk of lung cancer
among various cohort subgroups, including those who were exposed after
the institution of exposure controls in the 1950s.  Animal experiments
have generally not used enough animals or dose groups to adequately
describe a dose-response curve for beryllium-mediated lung cancer, but
yield additional support for the carcinogenicity of various beryllium
compounds.

Assessment of risk

Sanderson et al. examined the relationship between beryllium exposure
and lung cancer using data on exposures, work history, and mortality
among 3,569 men employed at a Reading, Pennsylvania beryllium alloy
plant between 1940 and 1969.  A case-control study design was
implemented, matching five controls with each case based on age and race
and truncating the controls’ exposures at the time of the case’s
death, where necessary.  Sanderson et al. fit conditional logistic
regression models with and without lags to account for lung cancer
latency, finding that lags of 10 and 20 years significantly improved the
fit of the exposure-response model compared to a model with no lag. 
Models with log-transformed exposure fit better than models with
untransformed exposure terms.  Estimates of employees’ cumulative
exposures, in units of (g/m3)-days,  average exposure concentration,
calculated as cumulative exposure divided by employment duration, and
estimated maximum exposure concentration had a statistically significant
relationship with lung cancer risk in the final models (p < .01).   

OSHA’s contractor used Sanderson et al.’s fitted exposure
coefficients from the 10-year- lagged and 20-year-lagged cumulative
exposure models to estimate the lifetime excess risk of lung cancer
associated with continuous occupational exposure to beryllium for a
45-year working lifetime.  The contractor implemented a lifetable with
year 2000 lung cancer and all-cause mortality rates to estimate excess
risks through age 100, adjusting for competing mortality.  Table 1 below
presents the excess lung cancer deaths expected to result from 45 years
of continuous occupational exposure to airborne beryllium concentrations
of 2, 1, .5, 0.2, and 0.1 ug/m3 in a group of 1000 employees, as
calculated by OSHA’s contractor based on the statistical analysis
performed by Sanderson et al.  It should be noted that OSHA estimates
risk from 45 years of exposure in order to gauge risk to employees for
a working lifetime, as specified in the OSH Act.  The predicted
excess risk to an employee exposed to beryllium for less than 45
years is expected to be lower than indicated in Table 1.

Table 1: Predicted Excess Lung Cancer Mortality

8-hr TWA beryllium exposure	Excess cancers

2 g/m3	28 - 47 per 1000

1 g/m3	26 - 43 per 1000

0.5 g/m3	24 - 39 per 1000

0.2 g/m3	21 - 34 per 1000

0.1 g/m3	18 - 30 per 1000



There are a number of uncertainties associated with the risk estimates
shown in Table 1.  While Sanderson et al. performed a careful and
thorough estimation of exposures based on a large number of industrial
hygiene samples, most of these samples were collected between 1971 and
1992.  Exposure estimates for the 1940s and 1950s, during which time
most cohort members were exposed, are based on limited information and
are subject to a variety of assumptions (Sanderson et al., 2001 (a) and
(b)).  Exposure misclassification due to possible error in Sanderson et
al.’s estimates of these early exposures could partly explain the
appearance of high excess cancer risks that increase gradually with
exposure in Table 1. 

g/m3, especially during the 1940s – 1950s, when Sanderson’s
estimated daily weighted average exposures reached higher than 20
g/m3 for many jobs, and higher than 80 g/m3 for a few (Sanderson
et al., 2001 (b), Table II, p. 154).  To produce the risk estimates
shown in Table 1, OSHA’s contractor assumed that the excess cancer
risk associated with a high concentration for a short period of
employment (e.g., one year at 45 g/m3) is similar to the excess risk
from exposure to a lower concentration for a longer period (e.g., 45
years at 1 g/m3).  Although this assumption is common in cancer risk
assessment, OSHA has not yet determined whether it is appropriate in the
case of beryllium, in part because of the extremely high concentrations
and short time periods of early employees’ exposures.  It is important
to note that employees in the Ward et al. study hired between 1960 and
1969, who were probably exposed to much lower concentrations of
beryllium than those hired in previous decades, did not appear to have
an excess risk of lung cancer (Ward et al. 1992, Table XI).  While the
limited exposure information and the small size of the population for
employees hired during the 1960s limit the conclusions that may be drawn
from this analysis, it suggests that the risk to employees exposed at or
below the current PEL of 2 g/m3 may be lower than the estimates
presented in Table 1.  

It should also be noted that OSHA preliminarily  assumes that the odds
ratios presented in Sanderson et al., which were derived from
conditional logistic regression models using exposure data that were
both lagged and log transformed, accurately reflects the relationship
between cumulative beryllium exposure and lung cancer mortality in the
study.   The risk assessment by OSHA’s contractor based on the
Sanderson et al. results also assumes that these odds ratios are
approximately similar to those that would result from an unconditional
regression model.  Other potential sources of uncertainty in the
statistical analysis performed by Sanderson et al., including possible
issues with matching techniques and transformations on exposure data,
have been raised in a recently published reanalysis of the data and are
being considered by OSHA (Levy et al. 2007).  NIOSH is currently
updating the Sanderson et al. study which may address some of the
uncertainties discussed.

Employee Risk of Beryllium Sensitization and Chronic Beryllium Disease

As previously discussed, beryllium sensitization and chronic beryllium
disease have been studied among workers employed in beryllium production
and processing (Kreiss et al. 1997; Rosenman et al., 2005; Madl et al,
2007), beryllium ceramics (Kreiss et al., 1996, Henneberger et al.,
2001; Cummings et al., 2007), nuclear weapons (Kreiss et al., 1993, Viet
et al., 2000, Stange et al., 1996, Stange et al., 2001; DOE/HSS, 2006),
beryllium machining (Newman et al., 2001, Kelleher et al., 2001, Newman
et al., 2005; Madl et al., 2007), copper-beryllium alloy (Schuler et
al., 2005), and beryllium mining and extraction (Deubner et al., 2001)
facilities.  The majority of studies found overall sensitization rates
of about 6-15% and CBD rates of about 4-8% in the populations or
facilities surveyed, with somewhat lower rates observed among nuclear
weapons employees (Stange et al., 1996, Kreiss et al., 1993; DOE/HSS,
2006), employees hired after implementing a comprehensive exposure
prevention program at a beryllium ceramics facility (Cummings et al.
2007), and workers in a beryllium mining and extraction facility
(Duebner et al., 2001).  Most epidemiological studies have reported
rates of sensitization and disease based on a single screening of a
working population (‘cross-sectional’ or ‘population prevalence’
rates).  Studies of workers in a beryllium machining plant and a nuclear
weapons facility have included followup of the population originally
screened, resulting in the detection of additional cases of beryllium
sensitization over several years (Newman et al., 2001, Stange et al.,
2001).  Table 2 presents the rates of beryllium sensitization and
chronic beryllium disease observed in several surveys that used the
blood beryllium lymphocyte proliferation test (BeLPT) to screen
beryllium-exposed workers.  It should be noted that in these studies,
persons who have evidence of subclinical disease are classified as
having CBD, regardless of whether they display overt symptoms of
disease.  

       

g/m3.  Evidence of an exposure-response relationship for beryllium
disease is reported in many of the studies cited in Table 2, based on
the association of increased risk with either quantitative measures of
exposure or specific work processes.  

OSHA is aware of important areas of uncertainty about the risk to
beryllium-exposed employees of sensitization and disease.  Some studies
have reported no statistically significant relationship between certain
measures of beryllium exposure and risk of sensitization and/or disease,
which could result from a small population size, lack of data on an
important measure of exposure, or inability to account for genetic
differences in individuals’ risk from beryllium exposure.  It is also
unclear whether all beryllium compounds cause equivalent risk.  Also,
while there is evidence that exposure to beryllium dust or fume composed
primarily of respirable particles (less than about 10 m in diameter)
may cause greater risk of CBD than exposure to larger particles, there
is currently insufficient information to quantitatively describe the
impact of particle size or compound on beryllium-related health risks. 
To address uncertainties such as these, and to create an improved basis
for quantitative risk estimates, the Agency is currently evaluating
additional data to determine whether it is appropriate for inclusion in
the quantitative risk analysis.  

As a part of this ongoing process, OSHA has conducted or contracted for
preliminary analyses based on two studies for which data are available. 
An analysis using data from the Henneberger et al. (2001) study of
employees at a beryllium ceramics plant appears in the attached
technical report, the results of which are presented in this document. 
Because the Henneberger et al. study is based on a cross-sectional
sample of a working population, this preliminary analysis presents
estimates of the population prevalence rate for a given exposure level
rather than the type of lifetime risk estimates that were presented for
lung cancer.  OSHA is in the process of determining whether it is
reasonable to make the assumptions necessary to derive lifetime risk
estimates from this cross-sectional dataset.  The initial estimates of
sensitization and CBD prevalence provided here, based on the Henneberger
et al. data, are likely to change as more datasets are analyzed, as the
Agency further develops its methodological approach, and as current
uncertainties about the exposure-response relationships are reduced or
resolved.      

Summary of analysis based on Henneberger et al. dataset

Henneberger et al. conducted a cross-sectional survey of 151 employees
employed at a beryllium ceramics plant in 1998 (Henneberger et al.,
2001).  Employees were eligible who either had not participated in a
similar survey conducted in 1992 (Kreiss et al., 1996), or who had
participated and were not found to have beryllium sensitization or
disease.  Fifteen of the 1998 survey participants were found to be
sensitized to beryllium, of whom eight were found to have chronic
beryllium disease and one declined clinical examination.  Study
participants’ cumulative, mean, and peak beryllium exposures were 

estimated from area and short-term task-specific breathing zone samples
collected since the start of full production at the plant in 1981.  
Henneberger et al., found a suggestive but not statistically significant
increase in the prevalence of sensitization among employees with
“high” (greater than median) mean and peak exposures compared to
employees with “low” exposures.  

The study authors permitted OSHA to analyze the same dataset used in the
Henneberger et al. (2001) publication, including participants’
cumulative, mean, and peak exposure estimates, health status, and
duration of employment, to calculate preliminary estimates of
sensitization and disease rates at various exposure levels.  The
prevalence of sensitization was analyzed with binomial regression models
using each of the available exposure terms (cumulative, average, and
peak exposure) and their log transforms.   Influence analysis was
performed on each of the resulting models to identify and adjust for a
small number of data points with disproportionate influence on the model
results, as discussed in the attached background document.  Some models
of sensitization prevalence also included an indicator variable to allow
separate analysis for the population of long-term employees, who had
participated in the 1992 survey, and short-term employees, who had not. 
As discussed in greater detail in the background document, employees’
peak and mean exposures were statistically associated with risk of
sensitization in some of the regression models that were fit to the
Henneberger data, while cumulative exposure showed no statistically
significant relationship with sensitization in any model.  This behavior
is consistent with other immune hypersensitivity reactions in which the
risk of sensitization shows a greater dependence on the concentration of
the sensitizing agent than on the duration of exposure.  

Although a relationship between beryllium exposure and CBD risk has been
observed in epidemiological studies comparing individuals with CBD to
healthy, unsensitized individuals, no study has yet examined the effect
of beryllium exposure on the risk of a sensitized person developing CBD.
 Thus, while there is evidence that duration or level of exposure
affects the risk of CBD, there is little quantitative information
available to develop a model of the effect on development of CBD
following sensitization.  For this preliminary analysis OSHA has assumed
that between 36% and 100% of sensitized employees develop pathological
evidence of CBD, regardless of their exposure history.  This range is
based on the lowest and highest proportion of sensitized employees found
to have CBD in the studies shown in Table 2 and on a study that
estimated 37% of beryllium-exposed employees are expected to progress to
CBD within six years of follow-up after a diagnosis of beryllium
sensitization (Newman et al., 2005).  It is important to recognize that
in the available studies, employees with CBD typically had experienced
occupational exposure to beryllium over the course of several years, and
it may not be reasonable to assume that CBD would occur with frequencies
as high as 36 – 100% for sensitized employees who have only brief,
low-level exposures.  OSHA is using this approach only as a simplifying
assumption for the PIRFA analysis, and is working to develop an
appropriate approach to account for the influence of exposure history on
the risk of CBD following sensitization.  The predicted prevalence rates
of beryllium sensitization and disease among employees exposed at a
variety of eight-hour TWA concentrations, based on the fitted logistic
log-mean and log-peak models and the observed prevalence of CBD among
sensitized individuals, appear in Table 3 below.

g/m3	12.8 - 27.4%	4.6 - 27.4%

1 g/m3	10.3 - 14.7%	3.7 - 14.7%

0.5 g/m3	7.3 - 9.4%	2.6 - 9.4%

0.2 g/m3	2.7 - 6.4%	1.0 - 6.4%

0.1 g/m3	1.3 - 5.1%	0.5 - 5.1%



     

g/m3.     

OSHA emphasizes the preliminary nature of the analysis. The model
results and predicted prevalence rates based on the Henneberger data
must be interpreted with considerable caution for several reasons.
Sources of uncertainty include the small size of the dataset, limited
knowledge about the relationship between exposure and the development of
CBD following sensitization, and the roles that skin exposure and
genetic susceptibility may play in causing beryllium sensitization and
disease. The small number of beryllium-sensitized individuals in the
dataset limits the value of regression analysis.  The models have high
statistical uncertainty, are difficult to assess for lack of fit, and
may be highly sensitive to error or unaccounted variability in exposure
and other points in exposure estimates.  In addition, the assumption
that a fixed proportion of sensitized employees will develop CBD does
not account for the probable importance of duration and intensity of
beryllium exposure to the risk of CBD following sensitization; in
particular, the high end of the range assumed (that up to 100% of
sensitized employees may develop CBD) is based on studies of employee
populations with long-term exposure to beryllium and may not accurately
reflect the risk to individuals with brief, low-level exposures. Other
important areas of uncertainty include the accuracy of the
daily-weighted-average exposure calculations on which individuals’
exposure estimates were based and the effect of additional risk factors
on the observed exposure-response relationship, including skin exposure
to beryllium, which may cause sensitization in some employees, and the
HLADPB1-Glu69 genetic polymorphism that has been associated with
increased risk of sensitization and CBD.  Finally, it should be noted
that risk estimates based on cross-sectional studies such as this often
underestimate risk because people who are found to be sensitized or to
have pathological evidence of disease in medical screening, or who
develop clinical manifestations of disease, may leave their place of
employment at a higher rate than healthy, unsensitized individuals. 
OSHA expects that the continued development of its risk assessment will
lead to improved estimates of sensitization and disease risk and will
help to clarify some of the most significant areas of uncertainty that
have been discussed here. 

Cases Prevented

OSHA has made a preliminary estimate of the possible benefits associated
with the alternative PELs in the preliminary draft beryllium standard. 
The Agency considers this analysis to be preliminary in nature and
expects to continue to develop and modify this analysis during
rulemaking in response to comments.  

Possible benefits are identified as preventing, or avoiding, instances
of occupationally caused disease or death, here including beryllium
sensitization (BeS), chronic beryllium disease (CBD), and lung cancer. 
Benefits are measured as the (estimated) reduction in the number of
instances of disease or deaths resulting from a new health standard. 
The calculation of benefits is made by estimating the number of cases of
disease and death occurring at current exposure levels and subtracting
the number that are estimated by the risk assessment to occur at the
lower airborne exposure levels of the alternative PELs.  Multiplying the
estimated risk at each exposure level times the number of employees
exposed at that level produces a prediction of the number of cases of
disease attributed to exposure.  

The Agency has developed two models for estimating risk that it has used
to assess benefits.  One model estimates the risk of disease based on a
working lifetime (45 years) of exposure.  This model has been used to
estimate the lifetime risk of lung cancer.  The second model estimates
risk based on the overall prevalence rate (the number or percentage of
current employees with BeS or CBD) associated with the various levels of
exposure to airborne beryllium.  The prevalence-based estimates of risk
are only for BeS and CBD.  The Agency is relying on the lifetime risk
model to estimate the reductions in lung cancer at lower exposure levels
and relying on the estimates of prevalence rates for BeS and CBD, as the
underlying data for BeS and CBD come from cross sectional data sets. The
Agency has also used two variations of the prevalence data to model and
estimate the mean risk at each of the PEL alternatives (discussed above
in the risk assessment).  

Given the prevalence rates from the risk assessment for each exposure
level of the alternative PELs, the Agency estimated annual incidence
rates.  The annual incidence rates are estimated as the overall
prevalence rate multiplied by the estimated annual turnover rate of
persons with BeS or CBD (10%).  This approach will be accurate if the
measured prevalence rates are constant, or are from a steady-state
condition.  The annual incidence rate then is the expected rate, for a
given level of exposure, of annual new cases of disease.  For each
exposure level, multiplying the annual expected incidence rate and the
number of employees exposed at that level predicts the number of
employees that will develop disease.  

OSHA estimated current employee exposure levels (the baseline) by
developing an exposure profile for industries with beryllium in its
analysis of technological feasibility.  Data that were used to assess
current exposures came from industry sources, OSHA and its
contractor’s site visits, NIOSH reports, published sources, IMIS data,
and information in the docket from the rulemaking submitted in response
to the Agency’s request for information.  

g/m3 .

The effect of these shifts, in terms of the exposure profile and
estimated benefits, is that employee exposures in those operations would
drop from current levels of exposure to no exposure to beryllium as the
result of any standard.  Because there are relatively large numbers of
employees in both groups, this strongly affects the overall structure of
benefits.  For this reason, the Agency is presenting estimated benefits
in four tables (Tables 4 through 7):  1) for all affected employees
(Table 4); for only abrasive blasters (Table 5); for only dental lab
technicians (Table 6); and for all other employees (i.e., not including
both abrasive blasters and dental lab technicians)(Table 7).

	The benefits are presented in the tables below as ranges (of cases of
diseases avoided).  The range of values for cancers prevented represent
estimates based on two different exposure lags.  In the case of BeS and
CBD, the ranges reflect the values from the risk models of prevalence
rates.  Because annual incidence rates are determined by the disease
prevalence rate and the annual turnover rate, benefit estimates are
directly affected by estimates of employee turnover rates. The Agency is
preliminarily estimating turnover at 10 percent, which is a conservative
(low) estimate.  A number of the industries that are most affected by
proposed standard, because they routinely work with beryllium materials,
have mainly highly skilled production jobs which may have lower than
average turnover rates.  Most of the other affected sectors would have
higher rates more typical of manufacturing.  A more typical turnover
rate of 20 percent, for example, would double estimated benefits.  The
Agency will continue to study the issue as it further develops its
standard on beryllium through the rulemaking stages.  All benefits are
attributed to reductions in airborne levels of beryllium (total mass)
and not from other provisions in the standard or from reduced skin
exposure.  

g/m3 , Table 4 shows that for all occupations an estimated 159  to
813 cases of CBD and 16 to 34 cases of lung cancer would be avoided
annually.  Similarly, when only abrasive blasters’ benefits are
estimated in Table 5, 121 to 613 cases of CBD and 11-24 cases of lung
cancer are avoided if beryllium-free blasting media are used.  Table 6
shows that for the dental lab technicians, an estimated 29 to 157 cases
of CBD and 4 to 8 cases of lung cancer are avoided when non-beryllium
alloys are substituted for beryllium ones.  Finally, based on current
exposures, the Agency estimates (Table 7) a PEL of  0.1 would annually
prevent, for all occupations but abrasive blasters and dental
technicians,  1 to 2 cases of cancer, 9-43 cases of CBD, and 28-43 cases
of beryllium sensitization.  The exposure profile, developed in the
technological feasibility analysis, represents current exposures within
ranges--0.2 to 0.5 g/m3 per cubic meter, as an 8-hour, time-weighted
average, for example--while the risk estimates provide point
estimates--at 0.2, for example.   To estimate benefits, the Agency
assumed the exposures were at the

midpoint of the range in the exposure profile and that the risk at that
midpoint was the average of the risk at the upper and lower bounds of
the same range.   

Table 4.  Ranges of Estimated Benefit for all Occupations and Employees

Diseases Prevented by the Preliminary Draft Standard, Annually*

                                       BeS		   CBD		Cancer

PEL = 1		 415-779	 150-779  	 15-32

PEL= 0.5		421-786	153-786	15-32

PEL = 0.2		432-794	156-794	15-34

PEL = 0.1		440-813	159-813	16-34

*The Agency estimates that based on current exposures there are 17-37
cases of lung cancer, 447-842 cases of beryllium sensitization, and
163-842 cases of CBD annually.

Table 5.  Ranges of Estimated Benefits for all Abrasive Blasters

Diseases Prevented by the Preliminary Draft Standard, Annually

    			                            BeS		   CBD		Cancer

All PELs		336-613	121-613	11-24

Table 6.  Ranges of Estimated Benefits for all Dental Technicians

Diseases Prevented by the Preliminary Draft Standard, Annually

    			                             BeS		    CBD	Cancer

All PELs		76-157		29-157		4-8

Table 7.  Ranges of Benefits for other Occupations (not blasters or
dental techs)

Diseases Prevented by the Preliminary Draft Standard, Annually

    			                            BeS		   CBD		Cancer

PEL = 1		3-9		1-9		0-0

PEL= 0.5		9-16		3-16		0-1

PEL = 0.2		20-24		7-24		0-1

PEL = 0.1		28-43		9-43		1-2

	Note that the Agency’s baseline estimates of lung cancer (and the
number of cases avoided at various PELs) are based on an employee who is
employed in a beryllium-exposed occupation for his entire working life,
approximately from age 20 to 65.  The calculation is as if employees do
not enter or exit beryllium-related jobs, or switch to other exposure
groups during their working lives.  While the assumption of no mobility
among exposure groups over a working lifetime may seem restrictive, the
assumption actually yields somewhat conservative (lower) estimates of
the number of cases of avoided disease, given the nature of the risk
assessment model.  For example, if a job is performed by five employees,
each working nine years, rather than one employee for 45 years, the risk
model predicts a higher rate of lung cancer for the five employees.

Monetized Benefits

For informational purposes only, OSHA has estimated the monetary value
of the benefits associated with the draft proposed rule—as required by
OMB. These estimates are informational only because OSHA cannot use
benefit-cost analysis as a basis to determine the PEL for a health
standard under Section 6(b)(5) of the OSH Act and pertinent Supreme
Court decisions. Estimates of monetary values for the benefits
associated with the draft proposed rule—preventing BeS, CBD, and
cancers—are based on methodologies OSHA used in its recent rulemaking
for hexavalent Chromium (docket H-054A).  For illnesses that cause
premature death, such as for cancer, the Agency applies a value of
statistical life (VSL) approach.  OSHA estimates that an appropriate VSL
today is about $6.6 million.   

The $6.6 million value represents individuals’ willingness-to-pay to
reduce the risk of premature death.  It does not take into account the
medical costs associated with the period of illness preceding death (the
“morbidity increment”).  The morbidity increment can be valued using
a cost of illness approach.  The monetary value assigned to the
morbidity component of a premature fatality is simply the medical costs
associated with a case of lung cancer.  The Agency estimates that the
average medical costs for lung cancer are about $69,000, based on
OSHA’s analysis for its recent hexavalent Chromium proposal.  To
estimate monetary benefits due to the number of lung cancers avoided by
lower PELs for beryllium, the Agency is using a total of $6.7 million
per case, the VSL plus the morbidity increment.  

Monetizing the benefits of preventing CBD is more complex and uncertain.
 In one study, average life expectancy following CBD diagnosis was about
9 years.  In terms of monetizing the benefit of avoiding a case of CBD,
the benefit of avoiding a fatal disease, which may linger for several
years with several final years of rapidly deteriorating health, argues
for assessing the same value as for a premature death due to cancer, at
a minimum.  In contrast, a premature death from cancer is generally
thought to have a long latency period (often about 20 years) of normal
health preceding a premature death.

Today, employees are more likely to have beryllium sensitization
detected before CBD develops.  CBD may also be detected before obvious
symptoms are manifest (subclinical CBD).   However, even in cases with
subclinical CBD there can be a measurable loss of pulmonary function.  
When sensitized individuals are placed under medical surveillance and
removed to jobs with little or no exposure, the risk of developing CBD
may depend on the amount of beryllium inhaled and retained in the lung. 
Beryllium may reside in the lung for many years, and individuals have
become sensitized and developed CBD even after removal from exposure.
Research suggests that a majority of sensitized employees with a
substantial lung burden of beryllium progress to CBD eventually.  Once
diagnosed with CBD, steroid or other immune-suppressing therapies are
reportedly lengthening the life span of individuals with CBD.  However,
steroid therapy often has many serious side effects and health
complications, and some individuals still succumb to CBD with little
respite from medical intervention.  At this time there remains no cure
for CBD.  The immune-suppressant therapies have not been used long
enough to accurately predict their ultimate impact on extending the life
span of individuals with CBD.  Since CBD results in premature death,
like cancer, and may strike individuals at an earlier age, the Agency
has preliminarily concluded that the appropriate measure to monetize the
benefit of avoiding a case of CBD is the VSL.  If CBD did not result in
premature death, but was still a life-long debilitating disease, the
Agency would estimate the monetized benefit with some other
method—such as either a cost of illness model (medical costs, lost
income, etc.) or as a non-fatal cancer (approximately 60 percent of
VSL).  The Agency will also estimate the medical costs associated with
CBD for a proposed standard, but has not included an estimate here.  

For the PIRFA, the Agency has estimated the monetized benefits as if all
individuals suffer CBD as a debilitating disease resulting in premature
death.  (This is a conservative assumption that can overstate benefits
to the extent that individuals with subclinical CBD do not progress to
clinical CBD.)  Earlier studies of a cohort of 124 CBD cases in the
Beryllium Registry found that the mean length of the course of disease
was about nine years after onset (Freiman and Hardy).  The life
expectancy of  CBD-diagnosed employees in a study of employees at a
beryllium facility in Kazakhstan in the former U.S.S.R. was eight years
after diagnosis, which was on average at age 44 (Creek et al.). 
Mortality studies of CBD cases of the Beryllium Registry have also
reported extremely high standardized mortality ratios (SMRs) for CBD
cases of 211 and 321, for all causes of death, and mortality as high as
35 and 38 percent (Infante et al., Steenland and Ward, Hardy et al., and
Peyton).  The Agency has not found a more recent authoritative study of
CBD mortality and will continue to develop a model that more
realistically estimates the outcomes of individuals with CBD for
estimating monetized benefits than the simplifying assumption made for
this analysis. 

The monetized benefits associated with an avoided case of lung cancer
reported above are undiscounted.  (Since there is no real latency period
for the onset of CBD, the Agency is not discounting the monetized value
for CBD.  Extending the life span of individuals with CBD does argue for
this step, however.)  Due to the lung cancer latency period, the
benefits of any change in OSHA’s beryllium PEL will be delivered at a
point later in time than when costs will be incurred.  Consequently,
discounting is appropriate.  The Office of Management and Budget (OMB)
recommends discounting using the rates of 3 and 7 percent.  Discounted
VSLs would be lower than these unadjusted values.

	For purposes of monetizing the benefit of avoiding BeS, the Agency is
relying on a simple cost of injury model.  Beryllium-sensitized
individuals must continue to receive annual medical examinations and may
be moved to jobs with lower exposures or removed from any exposure.  The
Agency estimates that the cost of medical services, supervisor and
employee time for these services, administrative costs, and work
accommodation to eliminate or avoid BeS exposure is about $2,000 per
year for, on average, five years. If an individual was moved to a job
with no exposure and paid at the same wage as a skilled production
employee, the Agency estimates that would cost about $5,000 additional
per year. Alternatively, providing an employee with a purified air
respirator (PAPR) would alone cost an additional estimated $1,500 per
year.   These alternatives are discussed further in the estimated costs
of regulatory alternatives below (medical removal protection). 
Discounting the estimated $2,000 annual costs in future years at 7
percent, the Agency preliminarily estimates the annualized present value
of preventing a case of BeS is $8,774 per case.  Since sensitization
creates a risk of a potentially fatal disease, CBD, it is likely that a
willingness to pay or other utility based approach of valuing BeS would
result in a far higher valuation.  The Agency will be considering such
alternatives in developing the rulemaking.

g/m3 are estimated to be between $1.2 and $5.6 billion annually
(Table 8).

Table 8.  Summary of Monetized Benefits of Preliminary Draft Beryllium 

Proposal for All Occupations

(Based on values from Table 5, in $millions, annually)

 CBD		Cancer		Total of	               Total of

PEL	    BeS 		(VSL)		 (VSL)		low range	upper range

1.0	   4-7		990-5,14	1	98-211		1,092		5,359

0.5	   4-7		1,010-5,188	99-211		1,113		5,406

0.2	   4-7		1,030-5,240	100-226		1,134		5,473

0.1	   4-7		1,049-5,366	106-224		1,159		5,597

Table 9.  Summary of Monetized Benefits of Preliminary Draft Beryllium
Proposal, 

Excluding Abrasive Blasting and Dental Technicians

(Based on values in Table 7, in $millions, annually)

CBD		Cancer		Total of	             Total of

PEL	    BeS 		(VSL)		 (VSL)		low range	upper range

1.0	     0		7-59      		0-0		7   		59	

0.5	     0		20-106     	0-0   		20   		106			

0.2	     0		46-158		0-13		46		172

0.1	   0-1		59-284		7-13		66  		297

g/m3, would range between 0-$1 million for cases of avoiding BeS,
$59-284 million for avoiding cases of CBD, and $7-13 million annually
for avoiding cases of cancer.  Total estimated monetized benefits at
that lowest PEL would be $66 million annually as a lower bound (sum of
low ranges) to $297 million (sum of upper ranges) for the same
occupations.  Note that the monetized benefits, just like the benefit
estimates in Tables 4 through 7, are sensitive to the estimated turnover
rate.  A conservative (low) turnover rate of 10 percent has been used to
estimate benefits here.  A turnover rate of 20 percent would double the
estimates of monetized benefits.  

Objective of and Legal Basis for the Preliminary Draft Standard

The objective of the preliminary draft standard is to reduce the number
of illnesses occurring among employees exposed to beryllium in general
industry, construction, and maritime.  This objective will be achieved
by requiring employers to install engineering controls where appropriate
and to provide employees with the equipment, respirators, training,
medical surveillance, and other protective measures to perform their
jobs safely.

The legal basis for the standard is the responsibility given the U.S.
Department of Labor through the Occupational Safety and Health Act of
1970 (OSH Act).  The OSH Act authorizes the Secretary of Labor to
promulgate occupational safety and health standards as necessary “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.”
 

29 USC 651(b); 655 (b). 

Description and Estimate of Affected Small Entities

	The Small Business Regulatory Enforcement Fairness Act (SBREFA)
requires that OSHA estimate the number of small businesses (“small
entities”) affected by the preliminary draft beryllium standard. 
“Entity” describes a legal business entity or firm; an
“establishment” describes a particular site of economic activity.  
SBA size standards were collected from the table of Small Business Size
Standards Matched to the North American Industry Classification System
(2002) from U.S. Small Business Administration’s website. SBA size
standards for the affected beryllium-using industries are expressed most
often in terms of employment (U.S. SBA 2002).   For the NAICS industries
affected by the preliminary draft standard, there were three different
size standards of “small entity” based on number of employees: 
entities with 500 or fewer employees, up to 750 employees, and fewer
than 1,000 employees. For three industries, size standards are expressed
in terms of annual average revenue. For NAICS 238320 (Painting and Wall
Covering Contractors) and NAICS 238990 (All other Specialty Trade
Contractors) entities are considered small if their annual revenues are
less than $12 million. For NAICS 811310 (Commercial and Industrial
Machinery and Equipment), entities are defined as small if their annual
revenues are less than $6 million.  The Agency relied on the most recent
census data for its description of small entities in affected
industries.

Criteria for small entities are presented in Table 10 for each affected
industry.  Identification of affected industries and the number of
affected small entities within each industry were developed in the
analysis of technological feasibility.  The Agency developed data on the
total number of small entities in each affected industry in order to
estimate the revenues and profits for affected firms, permitting
estimates of the impact of estimated costs on revenues and before-tax
profits.  For industries in which SBA has defined small entities as ones
having fewer than 500 employees, the total number of entities was taken
from the 2001 Statistics of U.S. Business (SUSB) from the U.S. 

Census Bureau.  The SUSB does not provide data in a way that permits
estimating the number of small entities with either 750 or 1,000 or
fewer employees.  For those industries with these SBA size criteria,
OSHA estimated small entities from 1997 SBA firm size employment data. 
The Agency used the ratio between total number of entities in 2001 and
1997 to extrapolate the total number of entities defined as small from
SUSB data.  For NAICS 238320, 238990, and 811310 (where small entity is
defined by revenues rather than number of employees), OSHA estimated the
number of small entities by calculating average revenues per entity for
all employment size groups using 1997 SBA firm size employment database.
 After adjusting the 1997 revenues up by the ratio of 1997 to 2001
annual payroll ratio, OSHA identified the largest employment size
category with revenues less than the SBA standard of revenue for each
NAICS.  This determined the number of small entities for those three
industries 

Revenues for small entities were calculated as the sum of revenues for
all size groups below the maximum SBA classification. Since revenue data
by employment-size 



Table    10   page 1

Table 10 page  2category for entities are only available in the 1997
SBA firm size employment database, OSHA extrapolated total revenues for
all employment size groups using 1997 SBA revenue data and 1997 and 2001
annual payroll ratio, taken from 2001 SUSB data. OSHA estimated total
revenues for SBA-defined small entities by multiplying the total 

revenues with the percentage of annual payroll for this size group of
SBA standards. The percentage of annual payroll for SBA-defined small
entity size was computed from 1997 SBA firm size employment data. 
Revenues of small entities presented here represent 2001 information and
presumably, for most affected industries, are somewhat lower in most
cases than today, if only due to general price increases from inflation.
 Economic impacts will then be somewhat overstated when costs, which are
current estimates, are compared to revenues and profits (which are
derived from revenues). 

Before-tax profit rates were collected from Risk Management
Association’s (RMA) Annual Statement Studies 2004-2005.   OSHA
estimated the profit rate as the five-year average of before-tax profit
rates for 2000 through 2004. For six-digit NAICS industries for which
data are not available in RMA, OSHA estimated the profit rate based on
the broader four-digit NAICS industry. Profits per entity were
calculated by multiplying revenues per entity with the profit rate. 

The Agency believes that coal-fired electric utilities and primary
aluminum manufacture are two industries that may be affected by the
standard but have not been analyzed for the PIRFA.  The Agency believes
that beryllium exposure in coal-fired utilities (most of whom are large
employers) is confined to maintenance and cleaning activities of
boilers, furnaces, and bag houses in activities related to the handling
of fly ash or residue. Some industrial power-generating facilities could
also be affected. Exposures in aluminum smelting are very low but
persistent, possibly due to traces of beryllium in ore.  Most aluminum
manufacturers are also large employers.  

The cost estimates presented in the PIRFA are preliminary.  In most
cases, where the Agency (or its contractor) had uncertainty about the
appropriateness or effectiveness of engineering controls, it opted for
the most extensive—and therefore most costly--control.  The Agency
will revise its cost estimates as it collects information and feedback
from the SBREFA panel and later submissions to the rulemaking. 
Similarly, where there was uncertainty about the number of small entity
employers who had already installed needed engineering controls, or
program elements such as exposure monitoring or medical surveillance,
the Agency choose to assume in most cases that few or no employers had
done so. (Details of the assumptions and estimated costs for each
affected sector can be found in individual sector cost reports.)  As a
result, the Agency believes that the estimated impacts in Tables 11 and
12 as well as aggregated costs (Tables 14 and 16 below) are likely to be
overestimates of final costs and economic impacts.  

As a preliminary method of estimating the significance of the economic
impacts on affected entities, OSHA compared, for each industry, the
average costs of compliance for affected small entities with average
small entity revenue and profits. This analysis, which OSHA terms a
screening analysis, is a simple calculation of the costs as percentage
of profits and as a percentage of revenues.  It is not a prediction that
either revenues will increase by this percentage or that profits will
fall by this percentage. Instead, this is a screening analysis for the
potential significance of the economic impacts.  OSHA has not yet done a
full economic feasibility analysis, which would analyze the likelihood
of severe impacts on profits, and thus the likelihood that the standard
is economically infeasible.  In general, the issue of whether a fall in
profits will actually occur as a result of incurring these costs is
dependant on whether prices can be increased without such major losses
in revenue that few if any firms in a class remain viable.

 The preliminary screening analysis is presented in Table 11.  The
Agency also estimated costs and conducted a screening analysis for very
small employers, those with fewer than 20 employees.  Estimating the
number of these smallest employers and their revenues and profits was
performed in the same manner as for SBA-defined small 

entities.  Compliance costs as percentage of revenue and profits for
these smaller size employers are presented in Table 12.

Average, unweighted compliance costs as a percent of revenues for small
entities range from 0.3 percent at a PEL of 2 g/m3  (but units will
be dropped for convenience in the next paragraphs) to 0.5 percent at a
PEL of 0.1 (Table 11).    These summary statistics are simple
averages--unweighted by the number of small entities in different
industries or their relative size. This suggests that the typical
affected small firm could remain viable 

with price increases of 0.3 to 0.5 percent if these price increases did
not significantly change total sales. The unweighted average costs per
small entity are about $20,000 for the PEL of 1.0 and about $35,000 for
the lowest PEL alternative.  Small entities in 5 of 57 affected
industries are shown to have costs as a percentage of revenues, at a PEL
of 1, of over 1 percent of revenue based on the preliminary estimates of
cost.   Small entities in 9 of 57 affected industries are shown to have
cost as percentage of revenue in excess of 1 percent at the 0.1 PEL
alternative. 

The unweighted costs as a percentage of before-tax profits range from
10.7 percent at a PEL of 2 to 19.4 percent at a PEL of 0.1. Small
entities in 24 of 57 affected industries are shown to have costs as a
percentage of before tax profits, at a PEL of 1, of 

over 5 percent of profits based on the preliminary estimates of cost.   
Small entities in 28 of 57 affected industries are shown to have costs
as a percentage of profits before taxes, at the lowest PEL alternatives,
of over 5 percent of profits based on the preliminary estimates of cost.
  

When the impact data is weighted by the number of small entities in each
sector, which more appropriately averages the data, at the lowest PEL of
0.1, the average weighted cost is about $8,000 per small entity, and the
impact measured as a percent of revenue is 0.4 percent and impact on
profits 10.2 percent, as shown in Table 11.

g/m3.  

When the data are weighted by the number of small entities in each
sector, at the lowest PEL alternative, the average cost per affected
smallest entity is about $2,587; the average value of costs as
percentage of revenues is 0.5 percent and the average value of costs as
percentage of profits is 11.9 percent.  

The Agency also estimated the costs of adding the preliminary draft
standard’s comprehensive program without changing the current PEL. 
This is discussed in the 

alternatives near the end of this PIRFA, and the employer costs and
screening analysis results for this alternative are also shown in Tables
11 and 12.  

	The Agency has not performed any additional analysis of potential
impacts that 

Table 11 page 1table 11  page 2table 12 page 1table 12 page 2would
include secondary effects.  For example, for many machine shops, work on
copper-beryllium alloy is a small part of the volume of business.  For
many of these shops, the cost of meeting the requirements in the
preliminary draft standard may result in some employers investing in the
program and controls to meet the standard and others not doing so.  In
this case, the impact of the standard would be less since not all
employers would absorb the costs.

Summary of Reporting, Recordkeeping, and Other Compliance Requirements
tc \l1 "Summary of Reporting, Record Keeping, and Other Compliance
Requirements 

OSHA has estimated costs resulting from the preliminary draft standard
in the areas of engineering controls, exposure assessment, written
control plans, health screening, regulated areas, hygiene facilities,
training, housekeeping and respirators in all affected industries.  A
summary of the employee time unit cost assumptions used in the analysis
is presented in Table 13.



Table 13. Time Requirements of the Beryllium Proposal

Section	Requirement	Time	Employee Type

Exposure assessment	Employee productivity loss while pump is attached to
employee	30 min. per assessment	Employee

	Recordkeeping performed by manager	5 min. per assessment	Manager

Medical Surveillance 	Travel to and from examinations	60 min. per exam
Employee

	Complete occupational and health history survey	60 min. per survey
Employee

	Health history review and update	30 min. per review	Employee

	Skin and respiratory exam	30 min. per exam	Employee

	BeLPT	5 min. per test	Employee

	Recordkeeping (initial and periodic screenings)	2 min. per screening
Manager

	Recordkeeping (referrals)	45 min. per referral	Manager

Regulated areas	Time to set up a regulated area	8 hours per area
Supervisor

Hygiene facilities	Vacuum clothes	2 min. per day	Employee

	Shower	15 min. per day	Employee

Training	Time spent in class	8 hours	At-risk employees for PEL options
of  0.2 or 0.1 µg/m3



4 hours	At-risk employees for PEL options of  1.0 or 0.5 µg/m3



2 hours	Production employees exposed above any PEL



1 hour	Administrative employees exposed above any PEL

Housekeeping	Vacuum work area	15 min./day/employee	Employee

Written exposure control plan	Time spent producing the plan	24 hours
Manager, initially in high exposure sectors



8 hours	Manager, initially in low exposure sectors



4 hours	Manager, quarterly in high exposure sectors



2 hours	Manager, quarterly in low exposure sectors



OSHA estimated the cost of complying with the provisions in the
preliminary draft beryllium standard for small entities in each
application group and for each affected industry.  These were based on
detailed application group reports that were translated to affected
industries in the economic impacts chapter of the contractor report. 
The total costs of an industry resulted from the number of affected
entities in the industry and the level of current compliance with
provisions in the draft standard.  The following sections describe the
cost methodology for the provisions in the standard.  (Details of the
program cost methodology can be found in Section 2 Compliance Costs
among the specific reports available with the PIRFA, and there is as
well a detailed report on each application group’s estimated costs.)

Engineering Controls

The Agency’s preliminary analysis of technological feasibility
identified, for each process in each affected industry, the engineering
controls that were necessary to achieve employee exposure at the
alternative PELs.  Work practices are also part of the controls that can
be considered in meeting a permissible exposure limit.  The Agency then
estimated the costs of engineering controls, based on support provided
by its contractor and engineering experts.  The engineering controls
necessary for entities in each affected industry are included in both
the industry technological feasibility as well as detailed industry cost
reports.  

Many of the engineering controls take the form of enclosing open
processes and providing exhaust ventilation to remove airborne particles
or fume away from employees.  The costs for ventilation controls
included the cost of design and installation as well as operating and
maintenance costs (which were estimated to be 15 percent of capital
costs).  OSHA estimated the cost of ventilation controls by calculating
an average annual capital cost that would include ductwork and other
equipment, and in some cases, the cost of increasing system-wide
capacity and enhancement for heating or air conditioning make-up air. 
Once the amount of additional ventilation to meet a PEL was estimated,
the cost was determined by multiplying the additional volume needed by
the average annualized cost per cubic foot per minute (cfm) for that
industry (which varies from $12-25, depending on the industry).  Costs
for hoods, covers, or other types of enclosures were estimated
separately.  The costs of capital and other first-year cost were
capitalized, over a ten year period for most engineering controls, with
a 7 percent discount rate.  Costs for the use of respirators for
operations that could not be brought to an alternate PEL level were
included with the cost for engineering controls.  A table listing
engineering controls for each industry is attached as an appendix to
this PIRFA.  The analysis of engineering controls is preliminary, and
the Agency is continuing to evaluate the cost and appropriateness of
controls for affected industries.

Program Costs

A number of the provisions in the preliminary draft standard are
triggered by the PEL, the action level, or where there is anticipated
skin exposure from routine handling of beryllium powders and dusts or
contact with contaminated surfaces.  The most important of these
ancillary provisions, for estimating costs, was the need for protective
clothing and equipment, hygiene areas, housekeeping, and medical
surveillance.  The Agency preliminarily determined that most of the
affected industry sectors had production employees that regularly worked
with beryllium materials and would therefore need the protections
provided by those provisions (precision machining, copper rolling and
drawing mills, smelters, foundries, welding operations, beryllium
production, and beryllium oxide products—see Section 2 Compliance
Costs of the cost reports for details).   Dental laboratories and
abrasive blasting operations were judged to have little or no program
costs because it is anticipated that they will substitute for their
beryllium-containing substances.  Low-content beryllium machining and
spring manufacturing and metal stamping operations were also judged to
have production processes and exposures that would not entail protective
clothing, showers, etc. because only a few affected employees had
routine--and low--exposure.  In the high-potential exposure sectors, the
Agency concluded that the ancillary provisions for clothing, showers,
medical surveillance and so forth would be needed for all production
employees.  Estimated costs of the preliminary draft standard’s
various provisions are presented in Table 14, according to the various
PEL options.  OSHA estimates most of the production employees in the
high exposure industries will need protective clothing and hygiene areas
(changing rooms and showers).  Medical surveillance will also be
required as a result of dermal exposure which occurs at all PELs.  This
makes the costs of the ancillary provisions somewhat constant across the
PEL options.  Table 15 lists many of the unit costs that appear
throughout the analysis.  Table 16 shows the costs of engineering
controls and programs costs across the affected application groups.   

Table 14table 15table 16

The Agency does not attribute costs to the standard for activities that
affected entities are already doing. To determine current compliance
with program provisions in 

the standard, the Agency relied on information from site visits and
information supplied by employers in the industry.  When there was no
information available from an industry source, the Agency relied on
estimates from the National Institute of Occupational Safety and
Health’s (NIOSH) National Occupational Exposure Survey (1982).

In abrasive blasting operations and in dental labs, OSHA estimated the
cost of using substitutes for beryllium-containing materials, which is
an engineering control.   In each case, much of the application group is
already using substitute materials.  As a result, these industries have
very low if any costs for ancillary provisions.

Exposure Control Plan

A written exposure control plan is required by paragraph (f) of the
preliminary draft standard for employers within the scope of the
standard.  OSHA believes that the nature of the risk of beryllium--with
risk of airborne exposure in microgram levels, possible beryllium
sensitization via skin, and issues of surface contamination and cross
contamination—means that employers would require considerable
resources to address the risk throughout facilities.  OSHA has estimated
that it would require managers in high

 exposure industries as much as 24 hours to develop such a plan and
eight hours in industries with few operations and lower current
exposures.  Updating these plans is expected to take 2 to 4 hours each
quarter.   

The Agency believes that employers should already have a respiratory
protection plan, as required by current standards [29 CFR 1910.134], but
the Agency has included additional costs for respirator use when PELs
cannot be met, for maintenance operations, and for entering regulated
areas.  This cost model includes providing one purified-air powered
respirator (PAPR) for every 10 employees for high exposure maintenance,
cleaning up spills, emergencies, and so forth.

Exposure Assessment and Monitoring

	The preliminary draft standard requires affected employers to perform
exposure monitoring to assess employee exposures to beryllium.  If
available, historical data or objective data may serve as a substitute.
In addition, periodic monitoring is required if 

some employees or operations are above the PEL or action level.  The
Agency estimates that air monitoring samples will cost $187 per sample,
including time to take the sample, lab analysis costs, notification and
recordkeeping, and 30 minutes of lost employee

production time, on average. The Agency estimates that four samples will
be taken, where necessary, for each affected work area.  The number of
work areas was derived from process identification and description in
the technological feasibility analysis.   The first-year costs of
exposure assessment/monitoring is about $3.5 million, which when 

annualized is about $0.5 million.  This corresponds to about 19,000
samples for initial assessments.

The number of periodic samples necessary is both a function of the
alternate PEL as well as current exposure levels. The Agency estimates
that for the range of PELs employers will take the following numbers of
samples, annually:  2,933 (PEL of 1.0); 3,505 (PEL 0.5); 4,081 (PEL
0.2); and 6,404 (PEL 0.1).  Consequently, while the costs for initial
sampling are constant across the alternative PELs (Table 14), the costs
for on-going sampling range from $0 at the current PEL to about $0.8
million at the lowest PEL alternative of 0.1 g/m3.  

Protective Work Clothing and Equipment

Paragraph (h) of the preliminary draft standard requires protective
clothing for employees whose exposure exceeds the PEL or are at risk of
skin or eye exposure from routine handling of beryllium powders or dusts
or contact with contaminated surfaces.  As noted above, OSHA estimates
that this requirement will apply to all production employees in high
exposure industries.  The Agency estimated that reusable (washable)
clothing would be the least costly alternative ($480 per employee per
year) versus disposable Tyvek suits ($900 per year).  The cost for each
employer was estimated by the number of production employees requiring
clothing, gloves, etc. times the costs of these items. The total costs
of this provision were estimated to be $4 million per year.  The
Agency’s estimate of the unit cost of reusable clothing was based on a
similar cost analysis for crystalline silica.  However, the Agency
believes that this cost will be revised upwards somewhat as handling of
contaminated clothing may be more problematic with beryllium.

Regulated Areas

	Regulated areas are needed under the preliminary draft standard to
demarcate where expected exposures are above the PEL as a warning to
those entering to wear respiratory and other needed protection. 
However, because there are very few processes that cannot meet even the
lowest alternative PEL, the total costs for this provision are low.  The
largest cost item would be providing respiratory protection to
individuals who must enter the area.  

Hygiene Areas

	The Agency estimated that employers in industries with high exposures
would  require change rooms, showers, and locker areas (the same cohort
of production employees requiring protective clothing, medical
surveillance, etc.).  The Agency estimated costs for these based on the
cost of long-term rentals of modular units, even though some employers
could opt to construct permanent facilities (see Table 13 for some of
the unit costs).  For example, the Agency estimates that it would cost
$2,700 to install a change room and $396 for lockers, or about $279 for
each affected employee.  Showers would cost $32,960, or $412 per
employee.  By far the largest estimated costs of hygiene facilities is
the time employees will need for showering. The Agency included the cost
of employee time and cost to shower (15 minutes) and to vacuum debris
from clothing and shoes.  The cost in wages for a 15-minute shower for
affected production employees is estimated at about $14.5 million per
year--far exceeding the estimated $0.6 million in annualized costs of
the shower facilities themselves.  Costs were also estimated for shower
water and supplies (towels) and HEPA vacuums for cleaning clothing in
the change areas.  An estimated 2 minutes per employee per day to vacuum
clothes with a HEPA-filter vacuum, produces an annual wage cost of
almost $2 million.  

Housekeeping

	The preliminary draft standard requires employers to keep surfaces as
free as practicable of beryllium dust.  The Agency has estimated that,
on average, this will require 15 minutes of time from every employee per
day, as well as the cost of HEPA vacuums for this chore.  As can be seen
in Table 14, the Agency has estimated that this will cost about $13
million dollars annually, regardless of the PEL.  This is about the same
cost as the time estimated for employees to shower each day.  The Agency
estimates that employers would have to purchase a HEPA-filter vacuum for
every 10 employees.  

Medical Surveillance

	Paragraph (k) of the preliminary draft standard requires employers to
provide annual medical exams as well as exams before being assigned work
and at termination.  Annual exams are required for employees exposed
above the action level or having routine potential of skin exposure. 
Annual exams would include a medical and work history (including
beryllium exposure), a physical exam of the skin and respiratory system,
and a BeLPT.  The Agency estimates the cost of exams, including
pulmonary tests and the BeLPT, at $377, which includes an hour of travel
time to a clinic, about two hours for the complete exam and tests, and
recordkeeping costs.  The Agency has preliminarily estimated that
employees in high exposure industries will have to provide medical
surveillance for all production employees.  

Training

	The Agency estimated that employee training would be necessary for all
employees, administrative personnel as well as production employees. At
the lowest PEL alternatives, the Agency estimated that 8 hours of
training would be necessary, but less at higher PELs (4 hours).  Such
high estimates of time for training is due to the risk of skin exposure,
the importance of work practices, extensive practices in hygiene areas,
housekeeping activities, as well as more typical production work
practices and engineering control effectiveness to reduce source
exposures. Administrative personnel would only receive one hour of
training.  After initial training of all current employees, future
training would only be for new hires and when changes in controls,
processes, or other factors indicated additional training in altered
hazards.  Training was estimated to have total costs of $4.3 million per
year to $8.3 million per year, depending on the PEL.

Federal Rules That May Duplicate, Overlap or Conflict with the Proposed
Rules 

The Department of Energy (DOE) has issued a regulation entitled Chronic
Beryllium Disease Prevention Program (CBDPP) (10 CFR 850, Federal
Register Vol. 71, No. 27 Feb. 9, 2006).  The CBDPP establishes a
beryllium program for DOE employees and DOE contractor employees at DOE
sites which are subject to Atomic Energy Act (AEA).  OSHA does not cover
those sites.   Therefore, the Agency’s beryllium standard would not
duplicate or conflict with the DOE regulation.  Section 4(b)(1) of the
OSH Act exempts working conditions of certain Federal and non-Federal
employees from the provisions of the OSH Act to the extent that other
Federal agencies exercise statutory authority to prescribe and enforce
occupational safety and health standards.  The Department of Energy is
one of these agencies. The CBDPP does not apply to any non-DOE sites or
employees, which OSHA has coverage over.  There are some DOE sites that
are not covered by the AEA where the CBDPP does not apply and OSHA has
coverage.  The DOE has indicated that it will adopt a beryllium PEL
which OSHA issues through rulemaking as part of its CBDPP.

	There is also a Federal statute addressing the compensation of some
employees with beryllium related illnesses--The Energy Employees
Occupational Illness Compensation Program Act (EEOICPA) of 2000 and its
subsequent amendments.  This Act creates a Federal employees’
compensation program to cover beryllium-related disease for DOE
employees and its contractor employees, including many private companies
that work away from DOE sites.  Several of the private companies whose
employees are covered by the Act, either directly in amendments to the
Act or identified in subsequent Department of Labor regulations on the
Act, would be covered by an OSHA occupational health standard for
beryllium.  There would be no conflict or duplication between an OSHA
standard and the EEOICPA.  The OSHA standard would have requirements to
protect employee health in the future, and the EEOICPA provides
compensation for employees who have developed beryllium-related illness.
 

Alternatives

OSHA is considering various options to some of the provisions of the
preliminary draft standard as well as some broader regulatory options. 
The options to the provisions are discussed below, followed by the
regulatory options.

Do not promulgate a beryllium standard

	If the Agency determines that there is a significant risk of material
impairment at its current PEL that can be feasibly reduced, then the
Agency will consider promulgation of a beryllium standard.  If a
standard is promulgated, then a comprehensive program will be necessary
to address the occupational risks of beryllium.  If evidence suggests
that there is no significant risk below the current exposure limit, then
action would not be justified.  The OSH Act indicates that development
of a new standard is warranted when research, demonstrations, and
experiments, and such other information as may be appropriate, such as
feasibility of the standard, indicates that an employee will suffer
material impairment of health or functional capacity.  Moreover, the
priority of establishing standards is under the discretion of the
Secretary depending on the urgency of the need for mandatory safety and
health standards for particular industries and work environments.  A
more detailed explanation of the Agency’s approach and justification
for developing a health standard for beryllium can be found above in the
PIRFA in “Reasons Why Action by the Agency is Being Considered.”

Scope

 	The preliminary draft proposed standard covers exposure to all types
of beryllium and beryllium compounds in general industry, construction,
and maritime. As discussed above in the health effects section, OSHA has
made a preliminary determination that there is inadequate data, at this
time, on employees selectively exposed to a specific beryllium compound
to eliminate a potential CDB concern for any particular form of
beryllium.  Therefore, the Agency has made a preliminary determination
to cover all beryllium compounds under the standard.   Preliminary
evidence also indicates that in addition to general industry settings,
certain construction and maritime work settings (e.g., abrasive blasting
operations using copper or coal slag) could generate significant
exposures of beryllium to employees involved in these operations.
Therefore, all three industries are included in the preliminary draft
proposed standard.    

	In some situations OSHA has not covered construction and/or maritime
operations in a health standard because there were no exposures or
significant risk, or there were unique problems which the Agency was not
in a position to address.  OSHA is not currently aware of these factors
for a beryllium standard.  If during the course of the rulemaking
evidence of these factors were brought to OSHA’s attention, the Agency
would consider exempting the construction industry and maritime
industries from the scope of this rulemaking.  OSHA of course will
consider suggestions from the SBREFA panel, its advisory committees, and
comments to modify the standard for construction and maritime to make
the standard more appropriate for those industries.   

In the construction industry, abrasive blasting operations that employ
beryllium-containing coal and copper slags used as blasting media have
been identified by the Agency as principally the operations with high
exposures to beryllium.  (DOE has reported considerable beryllium
exposure from construction activities, but those DOE
facilities/operations are not generally within OSHA’s jurisdiction.) 
In the shipbuilding industry, likewise, only abrasive blasting
operations have been identified as the principal sources of high
exposure.  The Agency has preliminarily concluded that abrasive blasting
operations will probably switch to blasting media that do not contain
beryllium to meet the requirements of a standard.

	There are substantial exposures in these industries to beryllium. 
Substitution of a blasting medium without beryllium contamination would
greatly reduce the risk of disease to employees at an estimated annual
cost of $36 million.  If the beryllium standard did not apply to these
industries, the substitution of alternative blasting media might not
occur and the benefits of a standard would not be realized (though the
costs would not be expended).  Most construction operations have no
exposure to beryllium, and the Agency is considering ways to make it
clearer that in those sectors the standard’s obligations do not apply,
and, of course, those employers would have no costs due to a standard. 
Alternatively, the Agency is considering whether or not the protection
required under its standards for abrasive blasting [29 CFR 1910.94(a)]
is sufficiently protective when beryllium-containing blasting media are
used.

There may also be exposure in construction and maritime operations from
welding.    In addition major repairs, rebuilding and renovation of
facilities where beryllium is present could lead to exposure of
construction employees.  In those circumstances, it might be possible to
tailor the standard’s provisions to protect employees while reflecting
the reality of jobs of short duration in the construction industry where
it may be difficult to provide hygiene facilities and monitoring data
may arrive after the job is complete.  Another problem with any
exclusion of the construction industry is the similarity between
construction and maintenance in renovation activities and the need to
cover all employees exposed to beryllium during such “dual work.” 
The Agency is seeking alternatives that can reduce the impact of a
beryllium standard on these industries while protecting employees and
making enforcement clear.

Limit Coverage to High-Content Substances

Another option to the scope of the preliminary draft proposal is to
limit the application to operations and industries that work with
materials with 0.1 percent or greater beryllium.  This option is similar
to the approach taken by DOE in its CBDPP rule where coverage is limited
to elemental beryllium and any insoluble compound or alloy containing
0.1 percent or more beryllium.  The advantage to this option is that
employers using compounds with only small amounts of beryllium, which
might present less health risk, would be excluded from the requirements
of the OSHA standard. The disadvantage to this option is that the data
may not be sufficient to conclude that compounds with less than 0.1% do
not present certain health risks such as beryllium sensitization.

The Agency expects that this option would exclude abrasive blasting and
most welding operations from the standard.  Both abrasive blasting media
(even coal and copper slags) and welding materials contain less than 0.1
percent beryllium. The cost reduction of this exclusion is shown in
Table 17.  It is also possible that some work on aluminum alloys would
fall below the 0.1 percent threshold as well as some smelting and
recycling operations.  Some of these operations have “high”
exposures (relative to the current PEL or the preliminary draft
standard’s alternative ones) even though the percentage content of
beryllium is very low. 

Exposure Metric  

The Agency is considering other possible ways to measure exposure and
establish

Table 17 health-based permissible exposure limits.  One alternative is
a PEL based on respirable mass concentration rather than total mass
concentration.  Respirable particle mass would 

conform with the International Organization for Standardization (ISO)
and the European

Committee for Standardization (CEN) recommendation for the aerodynamic
particle size range that penetrates the alveolar region of the lung
where CBD predominantly occurs.  Thus, a respirable particulate mass
concentration exposure metric would focus on beryllium particles most
relevant to impairment.  Respirable mass concentration can be measured
using an OSHA- or NIOSH-approved sampling method.  Samples are collected
with the use of a size-selective air sampling instrument, such as an
impactor or cyclone, and measured as an 8-hour time-weighted average
(TWA) concentration of respirable beryllium particulate.  

There is a substantial body of total mass measurements that have been
utilized in the health and feasibility analyses.   There is little
existing or historical air monitoring based on respirable beryllium. 
This poses a challenge in estimating the risk of CBD and determining
technological feasibility.  The Agency does not have an analysis of the
technological feasibility of these alternative approaches at this time. 
The Agency believes that the costs presented in this report are a
reasonable approximation to the magnitude of costs for other exposure
metrics without being able to make a correlation 

between the costs of any particular level of respirable mass and total
airborne mass (which is the exposure metric upon which the preliminary
draft standard is based).	

A second exposure metric alternative relies on a total mass PEL, but
tailoring the standard’s provisions for beryllium operations that
primarily release large particles.   

There is some information that inhalation of larger particles is less
likely to cause CBD.  

If such information ultimately shows larger particles cause
substantially less risk to employees, less restrictive regulatory
requirements might be developed for operations where air sampling
indicates that employees would be exposed to a minimal amount of smaller
particles.  Like the respirable beryllium metric, this alternative would
reduce the burden of complying with a low total mass PEL for those jobs
that entail little risk of CBD.  Unlike the respirable beryllium
alternative, it may allow OSHA to rely on the extensive total mass
exposure monitoring data for its risk and feasibility findings.  While
there are sampling and analytical methods for collecting small particles
(ultrafines) that have been developed and used experimentally, these
have not been fully characterized or routinely used by industry to
analyze beryllium particulate.  This alternate would require OSHA to
characterize new sampling methods as well as define the appropriate
particle size cut points and mass limits and have confidence that this
new approach would not put employees at significant risk.  The Agency
has not estimated costs for this alternative.

Require a Short-Term Exposure Limit (STEL) in Addition to a New PEL

Industries with operations that had exposures below a PEL could be
affected by a STEL, especially if there were episodic exposures as a
typical part of the production process.  The second exposure metric
option above (relying on total airborne mass and exempting
large-particle-generating operations) would also require a STEL.  
However, since most of the operations are not episodic but more
continuous throughout the production processes, OSHA believes most
operations could achieve a STEL.  The Agency has not yet extended its
current technological feasibility analysis to make a determination in
regard to a STEL, but for the reasons given above, the Agency does not
believe that a STEL would have a great impact on costs, except in a few
processes.  The costs for performing STEL exposure monitoring (as
distinct from controlling exposures to a STEL as well as a PEL) are
estimated below under exposure monitoring options.

Require a PEL for Surface Contamination of 3 Micrograms per 100 Square
Centimeters  

This option would require provisions for a surface contamination limit
of 3 µg/100 cm2 for non-production times.   This level was derived from
DOE’s housekeeping provision in the CBDPP rule (see Section 850.30).
The purpose of having this level would be to provide a quantitative
(rather than qualitative) limit to prevent the accumulation of beryllium
contamination on surfaces throughout the workplace.  Such accumulations,
if not controlled, pose a risk to skin (and perhaps beryllium
sensitization) and the re-suspension of beryllium particles into the
air. 

The Agency has collected only a little information on beryllium surface
contamination.  Surface exposure is a major part of DOE’s CBDPPP
regulation where contamination and decommissioning of shops and
equipment are important issues.  There are two reasons for limiting
surface contamination:  one is to check the effectiveness of engineering
controls and housekeeping practices; the other to reduce skin exposures.
 Limiting surface contamination could potentially raise costs
significantly; however, as explained above in the costs for housekeeping
provisions, the Agency has provided considerable estimated additional
costs to the draft standard to cover additional clean up—15 minutes
per day for each affected production employee.  At this time the Agency
does not have enough information about the extent of surface
contamination levels or exposure to beryllium in affected industries to
provide a further estimate for this cost.  The cost of surface sampling
alone (without a requirement as a PEL) is addressed below.  The Agency
is not estimating the costs of maintaining a surface sampling PEL beyond
the costs already estimated under housekeeping provisions.

No Action Level  

The written control program, medical surveillance, and exposure
monitoring are triggered by an action level (one-half the PEL).  An
option of no action level would trigger these provisions on the PEL
instead.  Generally, in its cost analysis, OSHA has assigned costs to
most of these provisions whether or not airborne exposures were above an
action level because of potential skin or eye exposure--which also
triggers these provisions.  The cost-savings of no action level are
estimated to range from about $0.2 million annually to about $1 million
(Table 17).  The disadvantage is that requiring provisions to be
triggered by the PEL might be less protective, especially if there is
remaining significant risk at the final PEL selected for the standard.

Exposure Monitoring

The preliminary draft standard contains provisions that require
employers to repeat personal monitoring at least every six months where
exposures are above the action level and every three months where
exposures are above the PEL.  Periodic monitoring can be stopped when
the monitoring indicates that the exposure is below the action level and
a second sample has confirmed that finding. 

	The Agency is considering two options.  The first is to require
monitoring less frequently, shifting to every 6 months if above the PEL
and every year if above the action level. This option would relieve
employers of approximately one-half of the monitoring costs.  The cost
savings of less frequent monitoring are shown in Table 17 and range from
no additional cost at the current PEL level to $0.4 million across the
PEL alternatives. 

However OSHA recently pointed out in its final rule, Standards
Improvement Project - Phase II (70 FR 1111, 1118 (Jan. 5, 2005)) that
"OSHA concludes, after reviewing the comments, that uniformity of
monitoring frequency [across occupational health standards] is
beneficial for employers and employees (unless there are specific
reasons for different frequency) because uniformity permits an employer
to develop a more efficient and thus better, industrial hygiene
program...."  In that rulemaking, OSHA standardized for most regulated
substances the 3 month/6 month frequency.  The advantages of uniformity
of monitoring frequency for small businesses needs to be considered.

The second option is to require periodic surface sampling. Surface wipe
samples would help employers determine the type of PPE and engineering
controls needed to effectively reduce or minimize exposures to
beryllium.  It can also permit evaluation of current housekeeping
efforts.  The disadvantage to this option is that, depending on the
frequency of the surface sampling, it would be costly for the employer,
especially small businesses, to collect samples, order necessary
equipment, and ship samples to a laboratory for analysis.   The Agency
has estimated the annual costs of this alternative as taking one sample
per year for every employee, including production employees in high
exposure industries and affected employees in the remaining industries. 
Lab analysis of surface samples is expected to cost about $45 and the
Agency estimates that the time to take the sample, notify employees, and
perform recordkeeping is about 15 minutes.  Total unit costs of the
samples are about $55 and total annual costs are estimated at $605,000.

        The Agency is also interested in, but has not estimated
the costs of, a performance standard for exposure monitoring.  A
performance standard would place the burden on employers to determine
the appropriate number of air (or surface) samples to take, presumably
sufficient to meet some criterion for reliability that would be set by
the Agency.  In general, a performance based approach, in contrast to
simpler specification one, could be a much more complex and potentially
more burdensome, especially for small employers, although it might
reduce monitoring for some larger employers with an extensive database
of measurements and professional industrial hygiene staff.  The Agency
is requesting comment on what such an approach might look like and how
it could be effectively enforced.

Regulated Areas

The primary cost of regulated areas is providing respiratory protection
to non-production employees who enter.  The cost of setting up a
regulated area is modest.  One option the Agency is considering is to
expand the definition of regulated area to include the potential for
skin contact with beryllium.  This would essentially make production
areas into regulated areas due to the potential for skin exposure and
concerns about surface contamination.  This would reduce to some extent
the number of employees or others coming into contact with beryllium and
could also reduce cross contamination.  Under this definition, since
employees would actually already be under the airborne PEL, in most
cases, it would not affect employee respirator use, but it could affect
employees from other work areas who wish to enter the space.   If they
would have to don a respirator, this would increase those costs. 
However, since exposures would not be above an airborne PEL, it is not
clear that for most of these regulated areas other employees would have
to wear a respirator to enter.  Perhaps a greater cost and impact on
exposure would be to further isolate production areas from other
employees.  The Agency is not providing an estimate of costs for this
option.  Clearly, the costs would be more than the simple PEL trigger,
but they may not be much greater if those entering do not have to don a
respirator.  The Agency has already concluded that most employees in
beryllium production areas are now wearing respirators, protective
clothing, and using change rooms and showers.  So these costs for those
employers would not be increased by expanding the scope or definition of
the regulated areas.

Methods of Compliance

The Agency is considering as an option to drop the requirement for
engineering controls to meet a PEL if beryllium is used fewer than 30
days per year.  Instead, employers could use respirators to protect
employees from exposure above a PEL. Engineering controls are part of
OSHA’s policy of following the hierarchy of controls to protect
employees from occupational hazards.  Removing the cause, or source, of
exposure with engineering and work practice controls is superior to
creating exposures/risk and then depending on PPE for safety.
Respiratory protection is dependent on the respirator and employee
behavior to succeed.  There is particular concern here because there is
evidence in the literature which suggests that exposure of less than 30
days can lead to beryllium sensitization.  This option would not affect
many of the industries in this analysis, but it could impact some
machine shops and abrasive blasting operations.  Machine shops currently
have fairly low exposure levels, and for many of these,
beryllium-related work is a small part of their business volume (see
attached reports on the Precision Turned Products industries).  Abrasive
blasters could be more greatly affected.  The Agency at this time does
not have any data that would provide an estimate of the frequency or
occasional use of coal and copper slags by blasters to meet the
fewer-than-30-day criterion.  The Agency has estimated that it would
cost about $36 million annually to substitute away from these
beryllium-containing slags.  If as much as one-quarter of the blasting
operations that would require a substitute (as currently under the draft
standard) would be allowed a 30-day limit, this would result in cost
savings of about $9 million annually.

Respiratory Protection  

The preliminary draft standard directs employers to follow OSHA’s
Respiratory Protection Standard (29 CFR 1910.134). The Respiratory
Protection Standard includes provisions for written procedures for the
proper selection, use, cleaning, storage, and maintenance of
respirators.   OSHA is considering an alternative to require employers
to provide a higher level of respiratory protection (such as powered air
purifying respirators (PAPRs)) for its workers when performing
activities that generate sub-micron particles of beryllium.  PAPRs with
an assigned protection factor (APF) of 1,000 would be expected to
protect the worker more effectively against high levels of airborne
exposure to beryllium than other types of respirators such as air
purifying half mask (APF of  ten) or full facepiece (APF of fifty). 
Requiring PAPRs could be more costly for the employer, especially small
businesses, not only to order them, but to maintain them, clean them,
and store them. The Agency’s cost model has many employees wearing
PAPRs now—one PAPR is made available for every 10 negative pressure
respirators.  The annual cost for wearing a PAPR is about $1,500
compared to about $500 for a negative-pressure respirator.  Annual costs
of providing PAPRs for every 1 in 5 employees are about $1.5 million and
for every employee about $13.4 million.  

Personal Protective Clothing and Equipment

The Agency is considering two options in this area.  First, whether or
not to specify the type of clothing, gloves, etc. for certain jobs. 
Secondly, to specify procedures for putting on and taking off PPE where
there is concern about ultrafine particles.  The Agency has
preliminarily concluded that neither of these provisions would pose a
significant addition to its current estimates of the costs of
compliance.  Specifications for clothing and equipment could potentially
raise employer costs by constraining employer flexibility or restricting
the choice as new products are developed.  The advantage to this option
would be that small business employers who would not normally employ
industrial hygiene expertise would know what PPE would be necessary for
a given operation without the expense of hiring an expert.  In regard to
methods of donning and doffing PPE, the Agency has already provided
considerable time for training and believes that such requirements could
be easily accommodated in the time given to that provision.  

Hygiene Facilities

The preliminary draft standard requires that employers provide clean
change areas and hand washing and showering facilities when employee
exposures to  beryllium exceed the PEL or where there is anticipated
skin exposure from routine handling of beryllium powders and dusts or
contact with contaminated surfaces.  This provision is similar to other
OSHA standards and is designed to prevent spreading beryllium to other
non-exposed areas of the work site, and to employees’ homes, and to
provide facilities for removing any beryllium that may be on the
employee’s skin.   

	The Agency is considering an option to drop the requirement for
employees to shower as part of the hygiene facility provisions. 
Instead, only hand washing facilities would be required for removing
beryllium from employees’ skin.  Removing this provision would reduce
costs by about $12 million annually across all PEL levels as the
Agency’s cost model and assumptions had put most production employees
in protective clothing and provided change rooms and showers as well. 
However, there are reported incidents of families members being
sensitized to beryllium and skin exposure may be a substantial source of
sensitization (Newman 1992).  

Housekeeping

The preliminary draft standard requires that employers ensure that all
surfaces are free of accumulated beryllium, and they are cleaned by a
method that minimizes the likelihood of redistributing beryllium, such
as by the use of a HEPA-filter vacuum.   In addition, shoveling,
sweeping, and brushing may be used only where HEPA-filter vacuuming or
other methods that minimize the likelihood of beryllium exposure have
been tried and found not to be effective.  Employers are required to
eliminate the use of compressed air without ventilation to remove
beryllium from surfaces. 

OSHA is considering one alternative in which OSHA would specify cleaning
procedures for removing, replacing, and cleaning air filters
contaminated with beryllium.  Employers would have specific procedures
to follow and would not have to consult outside expertise to know what
to do. The Agency does not believe that this would be a significant cost
item given that it has made substantial estimates of the time for
training under the standard.

Medical Surveillance

The Agency is considering an option to require a chest x-ray in the
initial medical exam to be evaluated by a B reader, and, as a second
option, to require x-rays every five years.  In the first option, as
part of the medical evaluation, the posteroanterior chest radiographs
are scored independently by a B reader.  The chest x-ray, along with
spirometry, may be used to establish a baseline for possible comparison
with future testing or evaluations.  This provision is consistent with
DOE’s provision on periodic x-rays (see 10 CFR 850.34).   The Agency
estimates that requiring an x-ray and B-reader in the initial exam would
add about $111,000 to costs when annualized over five years.  Requiring
an x-ray every five years would add about $215,000 annually to costs. 
These costs would be constant across the PEL alternatives.

	The Agency is also considering an option to require that previously
exposed employees be included in medical surveillance—with a medical
exam every three years.  The Agency does not believe that this would be
a significant additional cost for most affected industries.  The Agency
believes that internal turnover out of production work (but staying with
the employer) is not high in most of the affected industries in the
analysis.  The unit costs of providing surveillance have been presented
(above) in this analysis ($377); providing medical surveillance every
three years for previously exposed employees would roughly cut that
annual cost by one-third (on an annual basis) for previously exposed
employees.  But the Agency has no data with which to estimate how many
employees would be affected.

	The preliminary draft standard requires that all medical examinations
be performed by or under the supervision of a physician or licensed
health care professional (PLHCP).  However, in the case of beryllium,
there are certain medical complexities in diagnosing, treating, and
providing effective medical follow-up of beryllium-exposed employees,
particularly those who have had an abnormal (i.e. positive) response on
the BeLPT.   In order to address these medical complexities, OSHA is
considering three alternatives to the PLHCP.  The first alternative is
to require that all medical examinations be performed by or under the
supervision of a physician only.  OSHA is also considering a second
alternative to address the added concern that not all physicians may
have experience in this area due to the fact that cases of CBD are not
as common as other medical conditions.   Under this alternative, all
medical examinations would be required to be performed by or under the
supervision of a physician with knowledge and expertise in the diagnosis
and treatment of CBD.  However, since many of the medical complexities
may arise in the management of employees who have been determined to
have an abnormal or positive BeLPT test (i.e. they are sensitized to
beryllium), OSHA is also considering a third alternative.  This
alternative would require that if an employee is determined to be
sensitized to beryllium, then all medical examinations would be required
to be performed by a physician with knowledge and expertise in the
diagnosis and treatment of CBD.  This would limit the occasions when
beryllium-specific expertise would be required and would help to assist
the employer in deciding when special medical expertise would be
necessary.  The Agency has not estimated the costs of these
alternatives.

Training

The Agency is considering a requirement to provide additional training,
or retraining, every three years.  The Agency has estimated the annual
cost of such an optional provision as roughly one-third of the cost of
initial training, or about $1.9 million to $3.7 million annually,
depending on the PEL (Table 17).  The Agency’s view of training is
that it is one of the most effective ways to educate employees about
occupational risks and provide protection from these hazards, and that
it re-inforces employers’ policies and commitment.

	

Adopt the DOE Regulation

One major alternative to the preliminary draft standard would be to
adopt, in its entirety, the DOE Chronic Beryllium Disease Prevention
Program regulation--to the extent that the OSH Act would allow.  DOE
currently has provisions to prevent CBD from occupational exposures to
beryllium for DOE employees and its contractors at all DOE facilities
[see DOE’s (CBDPP) final rule, 10 CFR Part 850, December 8, 1999, Part
III, pp. 68854-68914].  In its regulation DOE specifically delegated to
OSHA responsibility for setting an airborne permissible exposure limit
(10 CFR 850.22).    The DOE Chronic Beryllium Disease Prevention Program
has an action level of 0.2 g/m3, and it and the preliminary draft
standard share most of the same provisions.  The DOE regulation has more
requirements for surface sampling than OSHA’s preliminary draft
standard.  The Agency believes that the costs estimated in this analysis
for a PEL of 0.2 g/m3 would reasonably approximate the costs of
adopting the essentials of the DOE standard as an OSHA health standard. 


PEL Only Standard

This regulatory option would eliminate all costs except for engineering
controls (and respiratory protection where still necessary) and
protective clothing, assuming that the PEL would have a designation for
skin protection, for which employers would provide protective clothing,
gloves, equipment, etc.  The cost savings for this alternative range
from $30 million to $50.8 million annually, which basically reflect
dropping remaining program or ancillary provision costs.   However, the
Agency has preliminarily determined that the risk of beryllium,
especially, argues for the necessity of a comprehensive program.  There
is evidence that employees may become sensitized via the skin.  This
argues strongly not only for stringent housekeeping but also protective
clothing, changing rooms, and showers.  The current theory of disease is
that it a two-step process, with CBD following beryllium sensitization. 
Medical surveillance is therefore needed to protect employees, to alert
employers of weakness in exposure control efforts, and to continue to
collect information about this not-fully-understood disease process. 
Provisions for exposure assessment are critical for planning engineering
controls.  In addition, Section 6(b)(7) of the OSH Act requires a 6(b)
standard to include provisions for labeling and warning signs, exposure
monitoring, and medical surveillance.  Such provisions seem especially
appropriate for a beryllium standard.

Medical Removal Protection (MRP)

␱崁Ꚅ廿㢄怄预懾̤

øòøòøòøòìæßæÙæßæòøòøòøòøÙòÙòÙòÙòÙòÒË
Â¹Â°©¢òÙòÙò¢ò¢ò¢òÙòÙòÙòÙòÙò¢

h

h

hr

h

ñÿP

'_

ñÿP

'_

葠ːሀ*

¦

©

@

	̤̀搒Ũ␱愁̤

̤̀옍)

	̤̀搒Ũ␱愁̤

葠ː̤̀옍)

昀ĴԀ

4

=

ÈÎÈãÈ–¿Ž†Ž†ŽÈ¿£È£È€ÈxÈrÈkÈ

ሎ฀䲄ᇿ킄ሂ桤ā㄀$葝ｌ葠ːሊᄀ킄ሂ桤ā㄀$葠ːሆሀ
桤ā㄀$ᔀconsidering whether to remove an employee from the area(s)
with beryllium exposure, the MRP provisions normally would specify that
the employee may select a second PLHCP for a second opinion, after first
visiting a PLHCP designated by the employer.  The second PLHCP normally
would review any previous findings, determinations or recommendations of
the initial PLHCP and conduct any appropriate medical examinations,
consultations, and laboratory tests deemed necessary.  

	If an employee tested positive for beryllium sensitization or was
diagnosed with CBD, the MRP option would require employers to provide a
PAPR or to offer an available job with exposures below the action level.
 The employer would have to maintain the employee’s original earnings
and other benefits. This option would also provide for additional
physician’s review and evaluation of the employee’s health status if
necessary.  Table 18 presents a model for estimating the costs of MRP.  

	OSHA estimated the number of employees subject to MRP as the total
number of employees in the exposure profile less the number of employees
in abrasive blasting and 

Table 18dental technicians.  The Agency then estimated the number of
employees with median exposures above 0.1 g/m3 and who did not work
for an employer who already has a 

removal program, which left 7,000 employees.  To estimate the number who
might be subject to MRP, the Agency multiplied that total by 3.7
percent, which was the prevalence rate that DOE found across a large
number of its employees and contractors’ employees (this also roughly
corresponded to prevalence rates from other cohorts).  An estimated 264
employees are expected to be sensitized to beryllium or have CBD in the
first year, and there will be 26 new employees each year who are
expected to qualify for 

the benefit. (Note:  In the benefits analysis above, the Agency
estimated 59-144 new cases of BeS and CBD at a PEL of 0.1 (Table 7); but
for purposes of estimating costs for MRP the Agency simplified the
counting of cases and also subtracted employers with existing removal
programs.)  Assuming employers comply with the medical removal
protection requirements by assigning employees with beryllium-related
health conditions to alternative jobs, the cost of MRP would depend on
the difference between the employee’s customary wage and the wage that
would normally be paid for the alternative work.  OSHA estimated costs
for medical removal protection by using the difference between the
average annual wages, including benefits, for production employees in
manufacturing and the average wages for a basic clerical job, or file
clerk. The Agency used a similar method of estimating the cost of MRP in
its Crystalline Silica PIRFA. This differential amounted to about $4,600
for a year, the total length of the MRP protection.   The Agency
estimates that one-half of employees entitled to MRP will be placed in
an alternative available job.  The remaining 50 percent would be divided
between individuals who would receive employees’ compensation for
beryllium-related disability and employees who would work in PAPRs since
(the model assumes) there would not be an available alternative job with
exposures below the action level.  The annual cost of a PAPR is
estimated at $1,416 and the cost of employees’ compensation two-thirds
of normal gross wages.  The first year costs of MRP are then estimated
to be about $1.5 million and  about $150,000 for new cases in future
years (Table 18).  Annualizing the first year costs over 10 years with a
7 percent discount rate yields about $225,000 (annually).   Total annual
costs are then estimated at about $375,000 for the MRP option. 

This cost impact may not be borne evenly by all firms.  Only a small
number of firms will be affected in a given year, but for those firms
the impact could be considerable—about $5,000 per case.  In addition,
this analysis assumes that businesses will often be able to find an
alternate job for the employee on medical removal, albeit at a reduced
level of productivity.  While this provides a useful means of estimating
the average case, this may not always be the case for individual firms,
particularly for small firms. While some firms might be able to offset
such a cost by placing the employee 

on employees compensation, for those firms that are not able to, or are
largely experience rated, the cost per case could be greater than the
average.

References

Creek, Kathyrn, Jacobson, L, Kovyazin, Al, Urikh, A., Medical Aspects of
Beryllium Production at the Ulba Metallurgical Plant,   HYPERLINK
"http://www.aha.org/aihce03/handouts/po117creek.pdf" 
http://www.aha.org/aihce03/handouts/po117creek.pdf 

Cummings, K. J., Duebner, D. C., Day, G. A., Henneberger, P. K., Kitt,
M. M., Kent, M. S., Kreiss, K., Schuler, C. R., Enhanced preventive
programme at a beryllium oxide ceramics facility reduces beryllium
sensitisation among new workers, Occupational and Environmental
Medicine, Volume 64(2): 134-140 (2007).

DOE/HSS, Beryllium Current Worker Health Surveillance Through 2005 ,
Publication ORISE 05-1711,   HYPERLINK
"http://www.hss.energy.gov/HealthSafety/IIPP/hservices/beregistry.pdf" 
http://www.hss.energy.gov/HealthSafety/IIPP/hservices/beregistry.pdf 
(2006).

Deubner, D., Kelsh, M., Shum, M., Maier, L., Kent, M. and Lau, E.,
Beryllium Sensitization, Chronic Beryllium Disease, and Exposures at a
Beryllium Mining and Extraction Facility, Volume 16(5):579-592, (2001).

Freiman, D. G. and Hardy, H. L., Beryllium Disease:  The Relation of
Pulmonary Pathology to Clinical Course and Prognosis Based on a Study of
130 Cases from the U.S. Beryllium Case Registry, Human Pathology, Volume
1, No. 1, 25-44, March 1970.

Hardy, H. L., Rabe, E. W., and Lorch, S., United States Beryllium Case
Registry (1952-1966) Review of Its Methods and Utility, Jornal of
Occupational Medicine, Volume 9, No. 6, 271-276, June 1967.

Henneberger, P. K., Cumro, D., Deubner, D.D., Kent, M.S., McCawley, M.,
and Kreiss, K., Beryllium sensitization and disease among long-term and
short-term workers in a beryllium ceramics plant, Int. Arch Occup
Environ Health, 74:167-176, (2001).

Infante, P.F., Wagoner, J.K., and Sprince, N.L., “Mortality Patterns
from Lung Cancer and Nonneoplastic Respiratory Disease among White Males
in the Beryllium Case Registry”, Environmental Research, 21 (1980),
35-43.

Kelleher, P.C., Martyny, J.W., Mroz, M.M., Maier, L.A., Ruttenber, A.J.,
Young, D.A., Newman, L.S., Beryllium Particulate Exposure and Disease
Relations in a Beryllium Machining Plant, Journal Occupational
Environmental Medicine, 43:238-249, (2001).

Kent, M. S., Robins, T. G., and Madl, A. K., Is Total Mass or Mass of
Alveolar-Deposited Airborne Particles of Beryllium a Better Predictor of
the Prevalence of Disease? A Preliminary Study of a Beryllium Processing
Facility, Applied Occupational and Environmental Hygiene, Volume 16(5):
539-558, (2001).

 	

Kreiss, K,. Mroz, M. M., Zhen, B., Wiedemann, H., Barna, B., Risk of
beryllium disease related to work processes of a metal, alloy, and oxide
production plant, Occupational Environmental Medicine, 54-605-612,
(1997).

 	

Kreiss, K., Mroz, M. M., Zhen, B., Martyny, J. W., and Newman, L. S.,
Epidemiology of Beryllium Sensitization and Disease in Nuclear Workers,
Am Rev Respir Dis, Volume 148 pp 985-991, (1993).

 	

Kreiss, K., Mroz, M. M., Zhen, B., Martyny, J. W., and Newman, L. S.,
Machining Risk of Beryllium Disease and Sensitization with Median
Exposures Below 2µg/m3 , American Journal of Industrial Medicine,
Volume 30: 16-25, (1996).

Kreiss, K., Wasserman, S., Mroz, M. M., Newman, L. S., Beryllium Disease
Screening in the Ceramics Industry, JOM Volume, 35(3): 267-247, (1993).

Levy, P.S., Roth, H.D., Hwang, P.M.T., and Powers, T.E., Beryllium and
Lung Cancer: “A Reanalysis of a NIOSH Cohort Mortality Study”,
Inhalation Toxicology, 14:1003-1015, (2002).

Levy, P.S., Roth, H.D., Duebner, D. C., Exposure to beryllium and
occurrence of lung cancer: a reexamination of findings from a nested
case-control study. Journal of  Occupational and Environmental Medicine,
Volume 49(1):96-101 (2007).

Madl, A. K., Unice, K., Brown, J. L., Kolanz, M.E., Kent, M. S.,
Exposure-Response Analysis for Beryllium Sensitization and Chronic
Beryllium Disease Among Workers in a Beryllium Metal Machining Plant,
Journal of Occupational and Environmental Hygiene, Volume 4(6), 448-466,
(2007).

Maier, L. A. and Newman, L. S., Beryllium Disease, Environmental and
Occupational Medicine Third Edition, edited by William N. Rom.,
Lippincott-Raven Publishers, Philadelphia, Chapter 27, 1021-1035,
(1998).

Marie, S., Janet, V., Krajewski, T., and Rogers, J., Chronic Beryllium
Disease and Beryllium Sensitization at Rocky Flats: A Case-Control
Study, AIHAJ 61:244-254, (2000).

McCanlies, E. C., Ensey, J. S., Schuler, C. R., Kreiss, K., and Weston,
A., The Association Between HLA-DPB1Glu69 and Chronic Beryllium Disease
and Beryllium Sensitization, American Journal of Industrial Medicine,
46:95-103, (2004).

McCawley, M. A., Kent, M. S., and Berakis, M. T., Ultrafine Beryllium
Number Concentration as a Possible Metric for Chronic Beryllium Disease
Risk, Applied Occupation and Environmental Hygiene, Volume 16(5):
631-638, (2001).

Medical Aspects of Beryllium Production at the Ulba Metallurgical Plant.
K. Creek, et al., [Creek, et al., 2003].  Los Alamos National
Laboratory, Los Alamos, NM, USA.  Ulba Metallurgical Plant, Ust’
Kamenogorsk, Republic of Kazakhstan.  Presented at the AIHCE, May 14,
2003.  Internet.

  	

Newman, L. S., Mroz, M. M., Balkissoon, R., Maier, L. A., Beryllium
Sensitization Progresses to Chronic Beryllium Disease, Am J Respir Crit
Care Med, Volume 171 pp 54-60, (2005).

Newman, L.S., Mroz, M.M., Maier, L.A., Daniloff, E.M., and Balkissoon,
R., Efficacy of Serial Medical Surveillance for Chronic Beryllium
Disease in a Beryllium Machining Plant, JOEM, Volume 43:(3), (2001). 

Newman, L. S., Kreiss, K. Nonoccupational beryllium disease masquerading
as sarcoidosis: identification by blood lymphocyte proliferative
response to beryllium. Am Rev Respir Dis, 1992 May;145(5):1212-4

Peyton MF, Worchester J., [Worchester, 1959], “Exposure Data and
Epidemiology of the Beryllium Case Registry 1958”, AMA Arch Ind
Health, 1959; 19:94-99.

Rosenman, K., Hertzberg, V., Rice, C., Reilly, M. J., Aronchick, J.,
Parker, J. E., Regovich, J., Rossman, M., Chronic beryllium disease and
sensitization at a beryllium processing facility.

Environ Health Perspectives, Volume 113(10):1366-72 (2005).

Sanderson, W.T., Petersen. M.R., and Ward, E.M., Estimating Historical
Exposures of Workers in a Beryllium Manufacturing Plant, American
Journal of Industrial Medicine, 36:145-157, (2001).

Sanderson W.T.,  Ward, E.M., Steenland, K., and Peterson, M.R., Lung
Cancer Case-Control Study of Beryllium Workers, American Journal of
Industrial Medicine, 36:133-144, (2001).

Schuler, C.R., Kent, M.S., Deubner, D.C., Berakis, M.T., McCawley, M., 
Henneberger, P.K., Rossman, M.D., and Kreiss, K., Process-Related Risk
of Beryllium Sensitization and Disease in a Copper-Beryllium Alloy
Facility, American Journal of Industrial Medicine, 47:195-205, (2005).

Stange, A., Furman, F. J., Hilmas, D. E., The Beryllium Lymphocyte
Proliferation Test: Relevant Issues in Beryllium Health Surveillance,
American Journal of Industrial Medicine, 46:453-462, (2004).

 Steenland K. and Ward, E., “Lung Cancer Incidence Among Patients With
Beryllium Disease:  A Cohort Mortality Study,” Journal of the National
Cancer Institute, 83 (1991), 1380-1385.

Tinkle, S.S.,  Anthonini, J. M., Rich, B. A., Roberts, J. R., Salmen,
R., DePree, K., and Adkins, E. J., Skin as a Route of Exposure and
Sensitization in Chronic Beryllium Disease, Environmental Health
Perspectives, Volume 111(9): July 2003.

Yoshida, T. Shima, S., Nagaoka, K., Taniwaki, H., Wada, A., Kurita, H.,
and Morita, K., A Study on the Beryllium Lymphocyte Transformation Test
and the Beryllium Levels in Working Environment, Industrial Health,
35:374-379, 1997.

 Note that estimates of cumulative, mean, and peak exposures in
Henneberger et al. were constructed based on individual work histories
and job-specific exposure sampling, not personal sampling of each
e灭潬敹⁥湩琠敨猠畴祤‮匠敥䠠湥敮敢杲牥攠⁴污‮
〲㄰映牯搠瑥楡敬⁤楤捳獵楳湯漠⁦数歡‬敭湡‬湡⁤
畣畭慬楴敶攠灸獯牵⁥獥楴慭整⹳഍ഃЍ഍ഃЍ഍倓䝁
⁅ᔠ഍ግ䅐䕇†ㄔക഍潃浭湥獴琠㩯†††删扯牥⁴畂
瑲‬晏楦散漠⁦敒畧慬潴祲䄠慮祬楳ⱳ䐠䅅††ⵎ㘳㜲

Date circulated:    June 14, 2007

