TECHNICAL SUPPORT DOCUMENT FOR REGULATORY
ACTION [TSCA: SECTION 6(a)] AGAINST
FRIABLE ASBESTOS-CONTAINING MATERIALS IN SCHOOL BUILDINGS
Seotember 1980
Office of Testing and Evaluation
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Washington/ DC 20460
-------
t '
When promulgating a rule concerning a chemical
substance or mixture under the Toxic Substances
Control Act (TSCA), the Administrator is required
to publish a statement on the effects of that
substance on health and the -magnitude of exposure
of human beings to that substance. This document
*
is a preliminary statement of these findings in
support of the rule "Friable Asbestos-Containing
Materials in Schools Proposed Identification and
Notification." It is a draft and is released for
comment on its technical merit and policy
implication.
-------
CONTENTS
I. INTRODUCTION 1
II. USE AND PRESENCE OF FRIABLE ASBESTOS-CONTAINING
MATERIALS IN SCHOOLS 2
A. Uses of Friable Asbestos-Containing Materials in
Building Construction 3
B. Presence of Friable Asbestos-Containing
Materials in Schools 6
C. Number of Persons Exposed to Asbestos in
Schools 12
D. Remaining Years of Use for School Buildings
.14
III. ASSESSMENT OF RISK FROM ASBESTOS IN SCHOOLS 15
A. Introduction 15
B. Hazard Assessment 17
1. Introduction 17
2. Health Hazards of Asbestos Exposure 19
a. Lung Cancer 19
b. Pleural and Peritoneal Mesothelioma ...22
c. Other Cancers 30
d. Asbestosis 32
e. Summary and Conclusions 43
3. Factors that Modify the Risk of Asbestos-
Induced Disease 44
a. Smoking 44
b. Age 51
c. Fiber Size and Type . 53
-i-
-------
d. Summary and Conclusions 55
C. Exposure Assessment 56
1. Asbestos Dispersion Mechanisms 58
2. Estimate of Prevalent Exposures 59
3. Description of Peak Exposures 69
D. Risk Assessment 70
1. Procedure for Estimating Risks of
Premature Death 70
a. Outline of the Risk Estimation
Procedure 70
b. Selection of the Underlying Study 72
c. Asbestos Exposure Among the Insulation
Workers 74
d. Increased Risk Among the Asbestos
Insulation Workers 79
e. Asbestos Exposure in Schools 82
f. Selection of the Extrapolation
Method . . . 84
2. Risk Estimates for School Building
Occupants 89
IV. IDENTIFICATION OF FRIABLE ASBESTOS-CONTAINING
MATERIALS IN SCHOOLS 94
A. Introduction 94
B. Sampling 95
C. Analysis 96
V. CONTROL OF ASBESTOS IN SCHOOLS 98
VI. REFERENCES 103
-11-
-------
I. INTRODUCTION
Exposure to asbestos fibers can lead to numerous serious and
irreversible diseases. Certain building materials in common use
can release asbestos fibers into the atmosphere. In particular,
friable asbestos-containing materials have been found to release
fibers in concentrations which, if inhaled, are sufficient to
increase the risk of developing such diseases. Some 3,000,000
students and 250,000 teachers and other staff regularly use
public school buildings which contain friable asbestos-containing
materials and which may contain such levels of contamination.
The Agency has determined that exposure to asbestos in school
buildings poses a significant hazard to public health. This
determination is based on the following considerations:
(1) the extent of use of friable asbestos-containing
materials in schools,
(2) the number of diseases which epidemiologic studies have
shown to be caused by exposure to asbestos,
(3) the evidence of elevated airborne concentrations of
asbestos in schools and other buildings where friable
asbestos-containing materials are present, and
(4) an estimate of the degree of risk posed by these
elevated concentrations.
In addition, information on the identification of friable
asbestos-containing materials and control measures that can be
taken to reduce the release of and consequent exposure to
asbestos fibers are presented.
-1-
-------
A. Uses of Friable Asbestos-Containing Materials in
Building Construction
Asbestos is used in buildings as a spray- or trowel-applied
coating to building surfaces to retard fire, deaden sound, or
decorate; it is also used in lagging on boilers and pipes, and in
cement products, plasters, vinyl tile, and miscellaneous products
such as lab table tops and ventilation hoods. Asbestos-
containing sprayed- or trowelled-on materials were first used in
the U.S. in 1935, when the material was found to be suitable for
acoustical purposes and for decorative finishes in public
buildings. In the 1950's, one of the most significant advances
in the construction industry was the replacement of concrete with
asbestos to protect structural steel against fire. Structural
steel must be insulated to ensure that it does not become soft,
bend, and collaspe during a fire. The replacement of concrete by
asbestos greatly reduced the weight and bulk of large buildings
(Sawyer 1979).
The amount of asbestos in the mixtures used in these
applications varies widely. From 1% to 80% or more of asbestos,
usually chrysotile or amosite or a mixture of the two, were
combined with other fibers (including cellulose, mineral wool, or
fiberglass), and cement or resinous binders. Table 1 shows the
results of analyses of a variety of sprayed-' or trowelled-on
asbestos samples from schools in the U.S. (Battelle 1980).
Nicholson et al. (1978a) reported similar concentrations for
schools in New Jersey. (Paint present in these samples may have
been applied at the time of spraying or during subsequent
maintenance operations.)
-3-
-------
Table 1. Site Sample Results from Bane Me Bulk Sample
(Continued)
Sample No.
15-01A
01B
QIC
01D
03A
038
16-01B
02A
02B
02C
020
03A
B
04A
048
17-01A
018
02A
028
Location
Hall
water damage
Hall
water damage
Hall
Hall
Hall
Hall
wet damage
Hall
wet damage
Hail
Hall
Music room
Music room
Caf Blot la
Cafeteria
Classroom
Classroom
Art room
Art room
Chrysolite Amosite
(%) (%)
90
80
50
40
20
5 40
5
0
5
10
10
10
10
5
5
0
0
0
0
Anthophyllite Other fibers
(%)
2% min.
wool
30% min.
wool
30% min.
wool
20% min.
wool
30% min.
wool
20% min.
wool
50% fiber-
glass
10% fiber-
glass
10% fiber-
glass
10% fiber-
glass
10% fiber-
glass
Nonfiber materials
5% calcite
3% opaque
10% gypsum, 5% glass.
5% opaque
10% gypsum, 5% glass.
5% opaque
20% gypsum
5% glass, 5% opaque
40% gypsum.
10% glass, 10% opaque
20% calcite
5% glass, opaque
50% opaque, 50% calcite.
40% opaque, 5% cotton
20% calcite
60% opaque
45% calcite,
50% opaque
50% calcite.
40% opaque
30% calcite.
50% opaque
25% calcite.
5% opaque,
5% cotton
40%calcrtB,
40% opaque
35% calcite,
10% opaque
35% calcite.
60% opaque
60% wood.
20% gypsum
10% opaque
60% wood.
20% gypsum
20% opaque
70% wood
20% gypsum,
10% opaque
60% wood.
20% gypsum.
20% opaque.
50% wood
Sample appearance
Fibrous
Fibrous
Fibrous
Granular
Granular
Chunky-
granular
Granular
Granular
Fibrous
Fibrous
Fibrous
-4-
-------
In the U.S., two principal methods have been used to apply
formulations of mineral fibers, including asbestos, for building
construction applications. In one method, dry fibrous material
was pumped through a 2- to 4-in. hose. The hose conveyed the dry
material to a nozzle at the actual site of application. As the
dry material left the nozzle, it passed through the focus of a
ring of fine water jets. The mixing took place at the focal
point, approximately 4 to 8 in. from the end of the nozzle (Reitz
1972). The mixture was directed against the building surface
from a distance of about 2 ft, and depths of application ranged
up to 3 in. The material applied by this method often was
fibrous in nature, rather than compact and granular. A coat of
resin or paint frequently was incorporated to increase the
cohesiveness of the final coating.
In the second process, the material was premixed with water
in a hopper, and the resulting slurry was pumped to a nozzle and
sprayed on the surface (Reitze et al. 1972). This usually
resulted in a less fibrous, more compact material being
applied. The depth of application generally did not exceed 1 in.
(Barnes 1976).
Material that was trowelled-on had essentially the same
composition as the sprayed-on materials, and it too was premixed
with water. This material probably formed the densest, hardest
coating of the three types. The depth'of application usually did
not exceed a fraction of an inch.
Asbestos also was applied to pipes and boilers in several
ways. In some instances, a wet slurry similar to the above
-5-
-------
material was sprayed or trowelled on. In others, a blanket
consisting principally of woven asbestos fibers was wrapped
around the pipe and secured with plaster, tape, or a sprayed-on
binder.
In 1973, EPA banned the use of spray-applied asbestos-
containing material as insulation in buildings to prevent
widespread contamination of the environment during spraying (EPA
1973). EPA amended this regulation in 1975 to include asbestos-
containing pipe lagging, regardless of the method of application
(EPA 1975). EPA extended this ban in 1978, ban to all uses of
sprayed-on asbestos (EPA 1978). EPA also regulated the methods
of removing asbestos from a building and disposing of the wastes
generated by removal (EPA 1973). These regulations apply to
"friable asbestos material," which is defined as material "that
contains more than 1 percent asbestos by weight and that can be
crumbled, pulverized, or reduced to powder, when dry, by hand
pressure." For the purposes of regulating the spraying process,
EPA defined "asbestos-containing" as containing >1% asbestos in
bulk. Thus, the regulation does not preclude the use of
materials contaminated by small amounts of asbestos (_<_!% in bulk)
in spray formulations.
B. Presence of Friable Asbestos-Containing Materials in
Schools
EPA has gathered information on the presence of friable
asbestos-containing materials in public schools by surveying
school districts and reviewing State, municipal and other
programs to identify and control asbestos hazards.
-6-
-------
Table 2 shows the results, as of April 1980, of an EPA survey
of school districts regarding friable asbestos-containing
materials. EPA mailed a guidance manual which included a survey
form to school districts across the Nation in May, 1979. A copy
of the survey form follows Table 2.
768 school districts containing 7,378 public schools (about
8% of the nation's total) responded to the survey. Of the 6,422
schools in these districts which were built or renovated between
1945 and 1973, 5,797 were inspected. 1,916, or 33% of the
inspected schools that responded were identified as having
friable asbestos-containing materials.
Although school districts across the country returned forms,
districts in 7 States responded and less than 2% of the districts
in 22 of the remaining States responded (Table 3).
EPA has preliminarily estimated that 8,545 public schools
have friable asbestos-containing materials. These estimates are
based on the survey responses and follow-up contacts with the
reporting school districts, contacts with school districts that
did not respond to the survey, and data supplied by New York
City's program on asbestos in schools.
-7-
-------
Tuble 2. U.S. Environmental Protection Agoncy, Office of Pesticides and Toxic Substances Asbestos Survey Report by School District
Region
1
\/
/
it
'II
'III
X
<
ttiuiiul
>tuls
#SCH.
IN DIST.
45
69
3.307
184
540
828
187
261
755
1.202
7.378
#SCH.
BLT/REN
45-78
32
62
3.180
64
466
675
94
199
670
980
6.422
#SCH.
INSP.
42
72
2.666
89
418
688
147
195
612
968
6.797
#SCII.
USING
PLM
10
10
357
14
106
70
22
13
63
94
749
#SCH.
W/ ASB/
DATE
10
4
1,574
11
63
33
33
14
79
105
1,916
#SCH.
EXPOSED
PROB.
7
a
267
9
20
15
21
10
36
43
436
SO. FT.
EXPOSED
ASBESTOS
9.50Q
7,073
2,414,320
125,290
327,911
366,662
844.697
1,399,991
286.677
628.970
6.295,991
j^CIIILD
EXPOS.
601
660
102,113
1,434
9,279
8,402
8,624
2,690
6,721
68,923
198,147
#SCLFT
REMOVE/COST
671,953
12,612,089
41,480
6,940,000
128,370
887,885
61,190
1,876,600
1.405
1,982,250
112,920
643.666
143.157
676,666
1,150.475
26.318,956
j^SO. FT.
ENCAP/COST
1.200
2.250
1,000
721,527
6.508,988
63.840
1.009.200
269.317
25,353,650
281,500
65.000
123.800
610.000
15.200
1,400.400
142,248
866.400
1.619,632
35.815.888
/SO. FT.
ENCLS/COST
5.000
19.000
100.435
7.155.785
10
100
125,000
125.000
13.000
116,640
6,317.000
5,244
2,875
365.329
12.619,760
//SQ. FT.
DEHtH/INSP.
4.500
6.073
1.172,854
87,720
362.100
106,291
13,500
30,347
27,300
386.968
2,197.663
u: See the following "Asbestos Survey Report", EPA Form 7710-29 (3-79). for full text of questipns 4 through 12.
-8-
-------
ASBESTOS SURVEY REPORT
.'Survey c; Ac:ivit:es to Control
A soesros—Containing Materials in School Buildings)
ForTT. Approved
OMB No. 158-R-0165
GENERAL
Th
on
to
is information is collected under the authority of the Toxic Substances Control Act, Sections 6 and 8. EPA is compiling information
the progress of State and local programs to control exposure to asbestos— containing materials in schools. This form should be used
periodically report information concerning the asbestos control activities in your school district. To obtain more forms, call this
— free number: 800 — i24— 9065 or in the Washington, D.C. area, call 354—1404. Data collected in this survey will be subject to the
of the Freedom of Information Act f 5 U.S.C. 552).
MAILING INSTRUCTIONS
MAIL ONE COPY TO:
The EPA Regional Asbestos Coordinator
for your Region. (For names and addresses
see reverse side.)
ALSO, please mail a copy to your official State asbestos program
contact (for name and address, call this toll—free number: 800—
424—9065 or i( in the Washington, D.C. area, call 554—1404),
IDENTIFICATION
1. SCHOOL DISTRICT INFORMATION
2. PERSON TO CONTACT REGARDING THIS REPORT
NAME OF SCHOOL. DISTRICT
N AME {last, nrst, as miqcTc initial)
"f OH COUNTY
J O H T I T j_ g
SPECIFIC QUESTIONS
3. Has the school district submitted an EPA Asbestos Survey
Reoort before?
i YSS
NO
UNKNOWN
4. How many schools in the district were built or renovated
between 1945 and 1978? .
NUMBER OF "SCHOOLS — — — — — —— ___
As of _ (mo./yr.), how many schools in the district
have been inspected for the presence of friable asbestos—
containing materials?
SCHOOLS
6. Howmany schools had bulk samples analyzed for asbestos with
the EPA recommended technique of P olarized Light Micro scopy i
"UMBER o>TcSbo~Ls~ "~ — — — —
I
7. As of.
.(mo./yr. of
analysis) for how many schools
in the district was friable ma-
terial analyzed as containing
asbestos?
NUMBER OF SCHOOLS
8. (a) In how many schools was friable asbestos—containing material determined to present
an exposure problem?
(b) Approximately how many square feet of this material were found?
(c) Estimate the number of children per school year exposed to this material. (Multiply
the percent of children exposed by the total number of enrolled students, e.g., An
exposure problem in five classrooms may involve 157c. of the total population of 700
students; 15%. x 700 equals 105 students exposed.)
(d) Have the names of the children been recorded and retained for future reference?
a. NO. OF SCHOOLS
b. SQUARE FEET
IC-
NO. OF CHIDREN
"Id. NAM
I 3
NAMES RECORDED
YES ~~| NO
I I_J YE5 _J N0
Questions 9 through 11 refer to the friable asbestos—containing material that presents an exposure prohlem in Question
9. (a) Aonroximatelv how many square feet of this material have 10. (a) Approximately how many square feet of this material h
9. (a) Approximately how many square feet of this material
been or will be removed?
(b) What is the estimated total cost of removal?
a. SQUARE FEET
T
COST:
10. (a) Approximately how many square feet of this material have
been or will be encapsulated?
(b) What is the estimated total cost of encapsulation?
a. SQUARE FEET
COST: $
11. (a) Approximately how many square feet of this material
been or will be enclosed?
(b) What is the estimated total cost of enclosure?
have
12. (a) For approximately how many square feet of asbestos-
containing material was action deferred?
(b) Will this material be inspected periodically to de-
termine if an exposure problem exists?
a. SQUARE FEET
a. SQUARE FEET
COST: $
b. PERIODIC INSPECTION
(H YES (H NO
13. What is the source of funding for the asbestos control
activities in your district?
14. When did (or will) the asbestos control activities tn the
district begin and end?
FUNDING SOURCE
BEGINNING YEAR
1
ENDING YEAR
COMMENTS
EPA Form 7710-29 (3-79)
-9-
-------
REGIONAL OFFICES
Region 1
Mr. Paul Heffernan
Asbestos Coordinator
Air &. Hazardous Materials Div.
Pest. & Toxic Substances Br.
EPA Region I
JFK Federal Bldg.
Boston. MA 02203
(617) 223-0585
Region 2
Mr. Marcus Kantz
Asbestos Coordinator
EPA Region II
Room 802
26 Federal Plaza
New York. NY 10007
(212) 264-9538
Region 3
Mr. Fran Dougherty
Asbestos Coordinator
EPA Region III
Curtis Building
Sixth & Walnut Streets
Philadelphia, PA 19106
(215) 597-8683
Region 4
Mr. Dwight Brown
Asbestos Coordinator
EPA Region IV
345 Courtland Street
Atlanta, GA 30308
(404) 881-3864
Region 5
Dr. Lyman Condie
Asbestos Coordinator
EPA Region V
230 S. Dearborn St.
Chicago, IL 60604
(312) 353-2291
Region 6
Dr. Norman Dyer
Asbestos Coordinator
EPA Region VI
First Internarl Bide.
1201 Elm Street
Dallas. TX 75270
(214) 767-2734
Region 7
Mr. Wolfgang Brandner
Asbestos Coordinator
EPA Region VII
324 East 11 Street
Room 1500
Kansas City, MO 64106
(816) 374-3036
Region 8
Mr. Ralph Larsen
Asbestos Coordinator
EPA Region VIII
1860 Lincoln Street
Denver. CO 80295
(303) 837-3926
Region 9
Mr. John Yim
Asbestos Coordinator
EPA Region IX
215 Fremont Street
San Francisco, CA 94105
(415) 556-3352
Region 10
Ms. Margo Partridge
Asbestos Coordinator
EPA Region X
1200 Sixth Avenue
Seattle, WA 98101
(206) 442-5560
-10-
-------
Table 3. Respondents to EPA Survey for As
(As of April 2
[-Containing Materials in Schools
0)
State
Alaska
Alabama
Arkansas
Arizona
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
No.
of forms
returned
2
0
0
122
11
3
3
8
1
1
0
6
5
0
3
3
1
4
1
0
2
44
4
3
6
3
Percentage of
total districts
that responded
3.9
0
0
57.8
1.0
1.7
1.8
50.0
1.5
0.5
0
5.2
0.5
0
0.7
0.9
0.5
6.0
0.7
0
0.8
7.6
0.9
1.9
1.1
0.5
State
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Washington, D.C.
(citywide)
West Virginia
Wisconsin
Wyoming
No.
of forms
returned
7
5
1
1
44
18
3
24
6
6
14
134
0
1
5
2
22
4
0
85
119
1
25
2
3
Percentage of
total districts
that responded
2.0
29.4
0.2
0.2
5.0
2.5
2.0
7.8
1.0
1.0
25.9
26.6
0
1.1
2.6
1.3
10.0
10.0
0
63.0
39.3
100.0
45.5
0.5
6.1
-------
The data for major school jurisdictions which reported
inspection results for large portions of their schools compare
favorably with the estimates. New York City reported that 180 of
the 1,735 city schools inspected had sprayed-on friable asbestos-
containing material in general use areas or 10.4% of the city
schools. A 1978 statewide survey of 326 schools in Rhode Island
revealed that 24 (8%) had sprayed-on asbestos material. In nine
schools some degree of deterioration was noted (Faich 1980).
Massachusetts' Special Commission on Asbestos in Schools and
Public Buildings reported that walk-through surveys had been
conducted in all 1,432 public schools in the State which were
built or renovated between 1946 and 1972. 178 schools, 12%, were
identified as containing "sprayed-on asbestos" (Commonwealth of
Massachusetts 1978).
Several factors, in addition to the low response rate in some
States, affect the validity of the estimates. First, the
analysis is based on a small sample with a large response from
one geographic area (EPA Region III). Second, this sample is not
random, and it may reflect a bias due to the use of information
from early respondents to the survey.
C. Number of Persons Exposed to Asbestos in Schools
Throughout the country, an estimated 3,000,000 students and
250,000 teachers, administrators, and other staff, including
approximately 23,000 janitorial and maintenance workers are
potentially exposed to airborne asbestos from friable asbestos-
containing materials in public schools during the school year.
An additional unknown number of persons may be exposed in private
schools.
-12-
-------
The number of exposed students was estimated from information
contained in responses to the Agency's survey of school districts
regarding friable asbestos-containing materials. The survey form
included question 8(c): "Estimate the number of children per
school year exposed to this (friable asbestos-containing)
material." Respondents were instructed to consider whether only
a portion of the school's population used the area in which
friable asbestos-containing materials were found and to estimate
the "exposed" population accordingly (see survey form).
Adjustments were made to this data base by contacting
responding school districts, reviewing data from New York City,
and contacting school districts which did not respond to
determine whether" the response to the survey was biased.
To complete the analysis, the sample school districts were
clustered by metropolitan code (inner city, suburban, rural), EPA
region, number of schools in the district, and number of students
per district. Survey results were then extrapolated to the
aggregate of public school districts to estimate the total number
of students using areas likely to lead to exposure.
The number of exposed teachers was estimated on the basis of
National Center for Education Statistics data that, nationwide,
there is approximately 1 teacher per 20 students. Finally, the
number of exposed janitorial and maintenance workers was
estimated on the basis of the assumption that there are
approximately two such staff persons for each of the 8,545 public
schools with friable asbestos-containing materials.
-13-
-------
D. Remaining Years of Use for School Buildings
School buildings are built to last for about 50 years,
although during the late 1950''s and the 1960's, slightly shorter
lifetimes were expected. The 50-year estimate is a rule of
thumb; no studies have been found that statistically or otherwise
validate this approximation (Gardner 1980).
The schools most likely to have friable asbestos-containing
materials were built between 1945 and 1973. Using the 50-year
lifetime estimate, a school built in 1945 would have a remaining
life of 15 years,-one built in 1973 a life of 43 years. If the
construction of the 8,545 schools with friable asbestos-
containing materials was equally distributed among the years 1945
to 1973, the average .expected remaining life would be 29 years.
Factors that affect this estimated average are:
(1) spraying of asbestos was most popular in the late 1950's
and in the 1960's;
(2) more schools were built during the 1950's and 1960's, to
accommodate the postwar baby boom, than during 1945-
1950;
(3) schools built in the 1950's and 1960's may not be
expected to last as long as those built earlier or
later;
(4) many schools are being closed across the nation
because of declining enrollment.
The first two factors would increase the expected average
remaining life of schools; the last two would reduce it. In view
of the lack of definitive information on these factors as applied
specifically to schools with friable asbestos-containing
materials, an average remaining life of 30 years has been chosen.
-14-
-------
III. ASSESSMENT OF RISK FROM ASBESTOS IN SCHOOLS
A. Introduction
Friable asbestos-containing materials that have been used in
the construction of a large number of schools release asbestos
fibers, and the Agency believes that occupants of these schools
incur risks of developing diseases caused by exposure to such
airborne fibers.
This section assesses the risks of adverse health effects and
premature deaths from exposures to asbestos in schools. In
making this assessment/ it was necessary to identify the health
hazards of asbestos exposure (Part B), to estimate the amount of
asbestos to which occupants of schools are being or will be
exposed, and to estimate the length of time and the number of
occupants who are and will be exposed (Part C). This
information, in turn, was used to estimate the number of people
expected to die from asbestos-caused diseases as a result of
exposure to "prevalent" (average) levels of asbestos in schools
(Part D), if all asbestos materials currently in the schools
remain in place until the buildings are no longer used.
The application of asbestos materials by spraying produces a
friable coating. The fact that asbestos fibers may be released
from these coatings was recognized as early as 1969 (Byrora,
Hodgson, and Helms, 1969), which led to considerable concern that
asbestos-caused diseases may develop in occupants of buildings
containing the coatings (Reitze et al. 1972). Investigators
found that fiber levels in these buildings varied widely because
of a combination of many factors (Nicholson et al. 1978a,
-15-
-------
Sebastian et al. 1979a). The wide variation of asbestos
concentrations in time and space means that no single measurement
can determine prevalent levels of asbestos fibers (Nicholson et
al. 1978a). Studies do show, however, that levels in buildings
containing friable asbestos materials can frequently be very high
("peak" levels; see Table 13, Part C). Exposure to these levels
and to lower, prevalent levels are predicted, as shown in Part D,
to result in a considerable number of premature deaths among
occupants of schools.
An evaluation of the risk requires combining estimates of
asbestos concentrations in the buildings, the risk of disease due
to a given exposure, the number of people exposed, and the
duration of exposure. The accuracy of the risk evaluation is
limited, however, because all of the available data are on a
small number of areas sampled in a small number of buildings or
on the risk of asbestos-induced disease in only a few
populations. This limitation is dealt with in two ways: (1) in
most cases, reasonable assumptions are made about how well the
sampling data apply to possible situations in schools and the
validity of these assumptions is discussed; (2) when reasonable
assumptions cannot be made, cases are presented that give the
lowest or highest reasonable estimate of risk. The accuracy of
the risk estimates is defined in both of these ways.
Three sets of reasonable assumptions are made that give low,
medium, and high estimates of the risk of mortality from exposure
to the prevalent concentration of asbestos in schools. These
estimates indicate that no fewer than 100 and no more than 7,000
-16-
-------
premature deaths will be caused by exposure to prevalent levels
of asbestos in schools if no controls are instituted. Additional
premature deaths caused by short-term exposures to "peak" levels
of asbestos also are very likely to occur. However, as explained
in Part D, the number of these additional deaths as well as
morbidity due to cancer and asbestosis cannot be estimated
quantitatively.
B. Hazard Assessment
1. Introduction
The first step in assessing risk from asbestos in schools is
to identify the adverse health effects arising from human
exposure to asbestos. The evidence comes primarily from
epidemiologic research. Persons exposed to asbestos were found
in these studies to be at increased risk of developing specific
diseases/ thereby implicating the diseases as hazards of asbestos
exposure. Indications of dose-response relationships in the
studies support these findings.
The use of epidemiologic research to identify a disease as a
hazard of asbestos exposure requires consideration of four major
criteria: bias, confounding, chance, and biologic plausibility
(Cole 1979). The proper design of studies and analysis of
results to avoid misleading interpretation due to bias and
confounding are explained in detail in many epidemiologic
textbooks (e.g., MacMahon and Pugh 1970). The probability that
apparent associations between asbestos exposure and specific
diseases might be due to chance alone is distinguished by the
application of standard statistical tests. In this assessment,
it is considered biologically plausible that asbestos exposure
-17-
-------
can increase the risk of cancer at a given anatomic site if
inhaled or ingested fibers can reach that site.
Epidemiologic studies may demonstrate dose-response relation-
ships (increasing risk with increasing level of exposure) for
asbestos-induced diseases in various degrees of detail. Many
studies group exposure into only a small number of categories
(e.g., "high," "medium," or "low"). These studies provide
"qualitative" evidence of dose-response relationships and are
briefly summarized. Other studies contain sufficiently detailed
exposure data to examine in more detail the shape of dose-
response curves within the range of observed exposures.
Part 2 below, identifies the following diseases on the basis
of epidemiologic reasearch as hazards of asbestos exposure: lung
cancer; pleural and peritoneal mesothelioma; cancers of the
larynx, oral cavity, esophagus, stomach, colon, and kidney; and
asbestosis.
The next step in the assessment is to identify factors that
influence the degree of risk posed by asbestos exposure. In Part
3, smoking, age, and fiber type and size are discussed as
possible factors that modify the degree of risk. As shown in
Part B, the increase in lung cancer risk among smokers exposed to
asbestos is greater than the sum of the separate increases
produced by asbestos exposure alone and smoking alone. Smoking
also may increase the risk of developing asbestosis. Because
children have a greater remaining life span than adults, they may
have a greater likelihood of developing asbestos-induced
diseases. The overall influence of other age-related risk
factors, however, is difficult to assess. All types of asbestos
-18-
-------
present in schools have been shown to be hazardous, and
differences in fiber size and type are not likely to affect the
risk assessment substantially.
2. Health Hazards of Asbestos Exposure
a. Lung Cancer
Epidemiologic studies demonstrate clearly that the risk of
lung cancer is increased by exposure to asbestos (e.g., Doll
1955, Selikoff et al. 1964, Peto et al. 1977, Newhouse and Berry
1979). Several studies show qualitatively that the greater the
exposure, the greater the increase in risk (Table 4). In
addition, the authors of two studies of respiratory cancer
mortality (predominantly due to cancers of the lung) among
asbestos workers have drawn linear non-threshold dose-response
curves to summarize their data (Figure !).!/. According to this
curve, all asbestos exposures, even those of very brief duration
or very low intensities, intensities, increase risk of cancer.
I/ Figures 1 and 2 are presented simply to demonstrate the shape
of the dose-response relationships in the two studies. These
and similar dose-response curves appearing in this report
should not be compared directly to each other, because
substantial differences exist among study designs,
measurement techniques, and exposure conditions. For
instance, asbestos concentrations in Figures 1A and IB were
measured with the same type of instrument (midget impinger),
but the method does not distinguish asbestos particles from
other particles. Thus, the apparent difference in slope
between the two curves could have resulted from a higher
fraction of asbestos particles in samples taken in the
asbestos products factory (Figure IB) than in samples taken
in the mining and milling facility (Figure IB).
-19-
-------
Table 4. Studies Showing Qualitative Dose-Response Relationships
Between Asbestos Exposure and Increased Risk of Lung Cancer
Reference
Type of asbestos
Type of exposure
Measure of exposure
to
o
Newhouse and Berry (1979)
Hobbsetal. (1979)
Meurman et al. (1979)
Wagoner et al. (1973)
Mancuso and El-Attar (1967)
Selikoff and Hammond (1975)
Nicholson et al. (1978b)
Hughes and WeilM 1979)
Mixed
Australian crocidolite
Anthophyllite
Primarily chrysotile
Primarily chrysotile
Amosite
Primarily chrysotile
Mixed
Factory work
Mining and Milling
Mining and milling
Factory work
Factory work
Factory work
Factory work
Factory work
Intensity and duration
Duration
Intensity
Duration
Duration
Duration
Intensity
Cumulative exposure
-------
10
(O
-*
v>
to «o
H «
' CC
3
500
x through 1966
o through 1975
1000
Cumulative exposure1
1500
(million particles per cubic foot x years)
Source: Adapted from McDonald and Lidded (1979).
CO
.g
0>
(U
CC
8
7
6
5
4
3
2
1
0
B
0 100 200 300 400 500 600 700 800 900 1000
Cumulative exposure"
(million particles per cubic foot x years)
Source: Adapted from Henderson and Enterline (1979).
aRelative risk is the mortality rate in an exposed group divided by the rate in a comparison group.
In Study A, the comparison group is the group of leajt exposed workers. In Study B, it is the general
population.
Units for cumulative exposure are not directly comparable among studies. See footnote on page 22.
Figure 1. Dose-response Curves for Respiratory Cancer Mortality in Two Groups of Asbestos Workers.
A, Chrysotile miners and millers; B, retired asbestos production and maintenance workers.
-------
The increases are directly proportional to cumulative
exposure.2/ This curve, and its use in predicting risk increases
predicting risk increases at low exposure levels, is discussed in
greater detail below in Part D.l.f.
Direct evidence of elevated lung cancer risk following low
cumulative asbestos exposure is provided by a study of asbestos
production workers (Seidman et al. 1979). In this study, men with
less than 3 months of employment had a lung cancer mortality rate
more than two times higher than that expected from general
population rates. EPA has estimated that the average exposure
level in the plant was 40,000,000 f/m3 (EPA 1979a). Thus, an
increase in lung cancer risk was detected epidemiologically
following a cumulative exposure of less than 10,000,000
f-yr/m3. Although it was achieved by relatively short-term
exposure to high concentrations, this is the lowest level of
•
cumulative asbestos exposure shown epidemiologically to lead to
increased lung cancer risk.
b. Pleural and Peritoneal Mesothelioma
Malignant mesothelioma is an extremely rare type of cancer
that appears as a thick, diffuse mass inside any of the serous
membranes (mesothelia) that line body cavities.^- Considerable
2/ Cumulative exposure is calculated by multiplying the average
concentration of asbestos in the air by the duration of
exposure. When concentration is measured in fibers per cubic
meter of air and duration is measured in years, the units for
cumulative exposure are fiber-years/cubic meter (f-yr/m3).
3/ There is a benign form of mesothelioma (Taryle et al.
1976). This discussion concerns only the malignant form.
-22-
-------
epidemiologic research (e.g., Wagner et al. I960, Mancuso and
Coulter 1963, Selikoff et al. 1965, Newhouse and Thompson 1965,
Ashcroft and Heppleston 1970, Puntoni et al. 1976) has shown that
exposure to asbestos can produce mesothelioma at two sites: the
pleura, the serous membrane that surrounds the lungs and
separates them from the thorax; and the peritoneum, the serous
membrane that surrounds the abdominal organs and lines the
abdominal and pelvic cavities.
Neither pleural nor peritoneal mesothelioma can be treated
effectively, and both are nearly always fatal (Taryle et al.
1976, Kovarik 1976, Saijo et al. 1978). One-half of all patients
die during the first year after diagnosis, and few patients
survive longer than 2 years (e.g., Whitwell and Rawcliffe 1971,
Rubino et al. 1972, Lumley 1976).
As in the case of lung cancer/ a number of epidemiologic
studies qualitatively demonstrate dose-response relationships
between occupational asbestos exposure and the risk of
mesothelioma (Table 5). In addition, Hobbs and colleagues (1979)
found that the incidence of pleural mesothelioraa among Australian
crocidolite miners and millers increased in direct proportion to
increasing duration of exposure. The linear trend and the
occurrence of mesothelioma among the workers in this study who
were exposed"most briefly (<3 months) are reasonably compatible
with a linear nonthreshold dose-response relationship (See Part
D.l.f).
-23-
-------
Table 5. Studies Showing Qualitative Dose-Response Relationships
Between Asbestos Exposure and Occurrence of Pleural and Peritoneal Mesothelioma
i
to
Reference
McDonald et al. (1970)
Newhouse and Berry (1979)
Hobbsetal. (1979)
Selikoff (1977)
Type of asbestos
Chrysotile
Mixed
Crocidolite
Amosite
Anatomic site
Pleura
Pleura/peritoneum
combined
Pleura
Pleura/peritoneum
Measure of exposure
Cumulative exposure
Duration and intensity
Intensity
Duration
separately
-------
Pleural and peritoneal mesothelioma are considered "marker
diseases" for asbestos exposure. A marker disease is one that is
often, if not always, caused by a particular agent. In all cases
of pleural and peritoneal mesothelioma, extremely rare types of
cancer, there have been very strong suspicions that exposure to
asbestos was the cause. In fact, as discussed below, close
examination of individual case histories of mesothelioma patients
usually provides evidence of some identifiable exposure to
asbestos above ambient levels, even if only of brief duration or
low intensity.
It is estimated that "apparently complete" case history
information reveals some source of asbestos exposure above
ambient levels for 85%-90% of all mesothelioma patients (Wagner
et al. 1971). For some patients, however, "apparently complete"
information is actually incomplete. Milne (1976) discovered that
the last known occupations recorded on death certificates
misleadingly indicated an absence of asbestos exposure for 66% of
a series of mesothelioma patients later found, when their case
histories were traced more diligently, to have been exposed.
McEwen and colleagues (1971) found that the hospital records of
55% of another series of patients did not contain information on
the asbestos exposures that these patients had, in fact,
experienced. In addition, mesothelioma patients who, during
personal interviews, were unable to recall experiencing any
asbestos exposure were later found to have asbestos fibers in
sections of their lung tissue taken at autopsy (Chen and Mottet
1978, Hourihane 1964). These studies strongly imply that
significantly more than 90% of all persons with mesothelioma have
-25-
-------
been exposed to asbestos above the ambient outdoor exposure
levels experienced by most urban dwellers (Selikoff and Lee
1978). Ambient exposure levels in urban and rural air may have
been responsible for a substantial proportion of the <10%
remaining cases. It is, therefore, reasonable to presume that
all cases of mesothelioma in persons who have had previous
asbestos exposure are the result of that exposure.
Given the status of pleural and peritoneal mesothelioma as
marker diseases for asbestos exposure, the many well-documented
cases that have followed extremely brief exposure to high
concentrations of asbestos or long-term exposure to low
concentrations provide evidence that risk is increased at these
low levels of cumulative exposure. Table 6 lists a few of the
cases of mesothelioma that have followed brief or low-intensity
asbestos exposure both inside and outside the workplace. Table 7
lists 48 cases of mesothelioma that have occurred in persons
sharing homes with asbestos workers. Table 8 lists 144
mesotheliomas that have occurred in persons who resided within a
mile of an asbestos products factory, mine, or shipyard and who
had no other known asbestos exposure. These case histories
provide evidence that very brief or low-intensity exposure to
asbestos can cause mesothelioma.
-26-
-------
Table 6. Mesothelioma Occurring After Brief or Low-Intensity Asbestos Exposure
Reference
Anatomic site and nature of exposure
Lieben and Pistawka (1967)
McDonald et al. (1970)
Borowetal. (1973)
Newhouse(1973)
Green berg and Lloyd Davies (1974)
Nurminen (1975)
Jones etal. (1976)
Anil and Holt (1977)
Bruckman et al. (1977)
Wiutwell et al. (1977)
Cochrane and Webster (1978)
Seidman etal. (1979)
1 pleural; helped replace plaster board
during extensive remodeling of his house
1 pleural; mixed and applied asbestos insulation
to boilers in home for "a few hours"
1 pleural; recycled asbestos filters in a brewery
1 pleural; sawed pipe coverings at home
1 (site unspecified); handled asbestos sheet
and pipe in a hardware store
2 pleural; stock boys in asbestos products
factory for 10 and 18 months, respectively
1 peritoneal; played on an asbestos factory
waste pile as a child
1 (site unspecified); relined and refitted clutches
and brakes as hobby.
1 (site unspecified); lived in a house largely
composed of asbestos sheeting
1 (site unspecified); worked on and lived adjacent
to a chicken farm with asbestos-cement buildings
1 (site unspecified)); sawed asbestos-cement
sheets for 1 day to construct two sheds
1 pleural; did repair work on own house and
handled asbestos boards
1 (site unspecified); inspector at a gas mask
assembly plant, did not handle asbestos pads
used in assembly of masks
1 pleural; resided near an asbestos products
factory for 2 years
1 (site unspecified); toll collector
2 pleural; filled gas mask cannisters with
crocidolite for 6 months
1 pleural; jeweller, occasionally cut
sections from a roll of asbestos textile
1 pleural and 1 peritoneal; worked in an
amosite products factory for less than
9 months
-27-
-------
Table 7. MesotK^Wma Occurring in Persons
Sharing Households with Asbestos Workers
No. of
Reference mesotheliomas
Anderson et al. (1976) 37a
Vianna and Polan (1978) 7
Hobbsetal. (1979) 2
Edge and Choudhury (1978) 1
Hainetal. (1974) 1
Total 48
I
ro
oo
i a
Total includes cases reviewed from reports other than those listed.
-------
Table 8. Mesothelioma Occurring in Persons
Residing Near Point Sources of Asbestos Emissions
Reference No. of mesotheliomas
Hain et al. (1974) 105a
Cochrane and Webster (1978) 13
Wagner et al. (1960) 13
Borowetal. (1973) 2
Greenberg and Lloyd Davies (1974) 10
Arul and Holt (1977) 1
Total 144
aTotal includes cases reviewed from reports other than those listed.
-------
c. Other Cancers
The scientific evidence reported below supports the
identification of cancers of the larynx, oral cavity, esophagus,
stomach, colon, and kidney as hazards of asbestos exposure.
Three cohort studies of asbestos workers (Newhouse and Berry
1979, Selikoff et al. 1979a, Rubino et al. 1979) and two case-
control studies (Stell and McGill 1973, Morgan and Shettigara
1976) found increases in the risk of larynx cancer following
exposure to asbestos. In one of the studies (Rubino et al.
1979), the risk increased with increasing cumulative asbestos
exposure, an indication of a possible dose-response relationship.
The rates of mortality due to cancers of the esophagus and
oral cavity the latter comprised of the (buccal cavity and
pharynx) were elevated in a group of 17,800 asbestos insulation
workers, compared with the rates in a group of other blue-collar
workers (Hammond et al. 1979, Selikoff et al. 1979a). These
cancers, like cancer of the larynx, have been shown to be related
to cigarette smoking (Hammond 1966). To allow for this
association, Hammond and colleagues (1979) accounted for the
smoking habits of the insulation workers and the comparison
group.
The asbestos insulation workers had higher stomach cancer
mortality rates than the comparison group (Hammond et al.
1979). In addition, stomach cancer rates were elevated in a
group of amosite production workers (Selikoff and Hammond •
1975). In the latter group, the risk of stomach cancer increased
with duration of asbestos exposure, an indication of a possible
dose-response relationship.
-30-
-------
A clear excess risk of colon cancer was reported in a group
of 632 asbestos insulation workers in New York and New Jersey
(Selikoff 1977a). Mortality .rates for cancers of the colon and
rectum combined were significantly elevated among the larger
group of 17,800 asbestos insulation workers (Hammond et al. 1979)
and among the amosite factory workers (Selikoff and Hammond
1975). Because the results for rectal cancer were not reported
separately in the latter two studies/ only cancer of the colon
can be said to be a hazard of asbestos exposure at this time.
The large group of asbestos insulation workers also
experienced an increase in kidney cancer mortality (Hammond et
al. 1979). This epidemiologic finding and the corroboration lent
by an experiment in which an excess of kidney cancer was seen in
rats fed ground, paper-based beverage filters containing 53%
chrysotile asbestos (Gibel et. al 1976) lead to the conclusion
that kidney cancer should be considered a hazard of asbestos
exposure.
Inhaled asbestos can be expected to reach each of the
anatomic sites at which increased risks of cancer were shown in
the epideraiologic studies discussed above. The process of
respiratory clearance results in exposure of the larynx, oral
cavity, esophagus, stomach and colon to inhaled asbestos
-31-
-------
fibers -i/. Exposure of the kidneys was shown directly when
chrysotile asbestos was found in the kidneys of rats (Cunningham
et al. 1977) and a baboon (Hallenbeck and Patel-Mandlik 1978)
that had been fed the fibers. Indirect evidence that the kidneys
become exposed was provided by the detection of fibers in the
urine of persons who drank water contaminated with the fibers
(Cook and Olson 1979).
d. Asbestosis
Exposure to airborne asbestos produces a chronic non-
cancerous disease of the lungs that/ in its severest form is
called asbestosis. As implied by its name, the disease is caused
solely by exposure to asbestos. It is characterized by a
hardening and thickening of lung tissue that is called
fibrosis. The rigidity produced by this process restricts the
normal movement of the lungs. Asbestosis is irreversible and
there is no effective treatment (Becklake 1976, Selikoff and Lee
1978). In advanced stages, the disease can be fatal. In a study
of mortality among asbestos textile workers employed under
extremely dusty conditions (Doll 1955), 63% of the death
certificates listed noncancerous respiratory disease in
conjunction with asbestosis as the cause of death.
4/ The airways of the lungs are lined with a layer of mucus that
is moved along by cilia, hairlike structures attached to the
free surface of the cells on the airway surfaces. Inhaled
particles that become embedded in the mucus eventually are
cleared to the oral cavity, where they are swallowed or
expectorated.
-32-
-------
A. Uses of Friable Asbestos-Containing Materials in
Building Construction
Asbestos is used in buildings as a spray- or trowel-applied
coating to building surfaces to retard fire, deaden sound, or
decorate; it is also used in lagging on boilers and pipes, and in
cement products, plasters, vinyl tile, and miscellaneous products
such as lab table tops and ventilation hoods. Asbestos-
containing sprayed- or trowelled-on materials were first used in
the U.S. in 1935, when the material was found to be suitable for
acoustical purposes and for decorative finishes in public
buildings. In the 1950's, one of the most significant advances
in the construction industry was the replacement of concrete with
asbestos to protect structural steel against fire. Structural
steel must be insulated to ensure that it does not become soft,
bend, and collaspe during a fire. The replacement of concrete by
asbestos greatly reduced the weight and bulk of large buildings
(Sawyer 1979).
The amount of asbestos in the mixtures used in these
applications varies widely. From 1% to 80% or more of asbestos,
usually chrysotile or amosite or a mixture of the two, were
combined with other fibers (including cellulose, mineral wool, or
fiberglass), and cement or resinous binders. Table 1 shows the
results of analyses of a variety of sprayed- or trowelled-on
asbestos samples from schools in the U.S. (Battelle 1980).
Nicholson et al. (1978a) reported similar concentrations for
schools in New Jersey. (Paint present in these samples may have
been applied at the time of spraying or during subsequent
maintenance operations.)
-3-
-------
Asbestosis is a progressive disease that has various degrees
of severity (Berry and Lewinsohn 1979). Asbestos exposure can
continue to cause damage to the lungs even after direct exposure
has ceased (Becklake et al. 1979). Symptoms likely to prompt an
exposed individual to seek medical care, such as loss of breath
or a bluish discoloration of the skin,_^ do not appear until well
after severe oxygen deprivation has occurred (Harries 1973, Robin
1979). In order to detect progressing asbestosis (i.e., the less
severe stages of the disease), exposed individuals must be
examined for clinical and diagnostic signs.
Most often, medical examinations of persons exposed to
asbestos include chest x-rays and a physical examination that
includes a determination of the presence or absence of
crepitations/ the abnormal lung sounds that are characteristic of
asbestosis (Leathart 1968, Forgacs 1967, 1969). Unlike lung
function tests, which are conducted less frequently, crepitations
and abnormal x-ray findings do not indicate directly that health
is impaired. Instead, they show that the disease process has
begun. For example, persons with crepitations have a high
probability of suffering later decrements in lung function (Berry
et al. 1979). Often, persons with lung damage visible on x-rays
already have impaired lung function (Jodoin et al. 1971, Selikoff
and Lee 1978). Abnormal x-ray findings also indicate that a
person is at high risk of subsequently developing more severe
5/ This discoloration (called cyanosis) is due to an excessive
concentration of nonoxygenated hemoglobin in the blood.
-33-
-------
stages of asbestosis. For instance, Liddell (1979) found that
asbestos miners and millers with lung damage detectable on x-rays
later experienced an asbestosis mortality rate nine times greater
than that experienced by workers with normal x-rays.
Consequently, although some researchers reserve the term
"asbestosis" for advanced stages of the disease [e.g., "clinical"
asbestosis (Murphy et al. 1971) or "certified" asbestosis
(McVittie 1965)], crepitations, x-rays findings of lung damage,
and measurements indicating decreased lung function are each
considered signs of asbestosis in the following discussion.
A large number of occupational studies have used the various
measures of asbestosis to demonstrate dose-response
relationships. The studies in Table 9 show qualitatively that
the risk of asbestosis rises with increasing asbestos exposure.
McDonald and colleagues (1979) described the dose-response curve
for asbestosis mortality among Canadian chrysotile miners and
millers as a linear relationship (Figure 2), although they
cautioned against extrapolation to very low exposure levels. An
earlier study of x-ray-detectable lung damage among South African
crocidolite miners and millers (Sluis-Cremer and duToit 1973) is
consistent with this finding (Figure 3), as is a very recent
study of asbestos textile workers in the United Kingdom (Berry
and Lewinsohn 1979). Data from the latter study are used in
Figure 4 to draw dose-response curves for three stages of
asbestosis (crepitations, "possible" asbestosis, and "certified"
asbestosis):
-34-
-------
Table 9. Studies Showing Qualitative Dose-Response Relationships
Between Asbestos Exposure arwarious Measures of Asbestosis
Reference
Selikoff and Hammond (1975)
Nicholson et al. (1979b)
Hobbsetal. (1979)
Lacquet (1979)
Selikoff (19771))
i Sclikof fetal. (1979b)
to
en
1
Selikoff (1977c)
Sluis Cremer and duToit (1973)
Baselga-Monte and Segarra (1978)
Harfetal. (1979)
Ayer and Burg (1978)
Type of
asbestos
Amosite
Primarily chrysotile
Australian crocidolite
Mixed
Chrysotile
Mixed
Primarily chrysotile
Amosite and crocidolite
Mixed
Mixed
Mixed
Type of
exposure
Factory Work
Factory Work
Mining and milling
Factory work
Mining and milling
Shipyard work
Insulation work
Mining and milling
Factory work
Spray application
Factory work
Measure of
exposure
Duration
Intensity
Duration
Cumulative
exposure
Duration and
intensity
Duration from
exposure onset
Duration from
exposure onset
Cumulative
exposure
Mean cumulative
exposure
Duration
Duration
Measure of
asbestosis
Mortality
Mortality
Incidence
Incidence
X-ray changes
X-ray changes
X-ray changes
X-ray changes
X-ray changes
Decreased vital
capacity
Decreased forced
vital capacity
-------
U)
Ch
SJ
I
Q>
cc
80
70
60
50
40
30
20
10
0
100
200 300 400 500 600
Cumulative exposure" (million particles per cubic foot x years)
700
800
Source: McDonald, as reported Acheson and Gardner (1979).
aSlope determined by the formula, slope =£x(y-1)/£x .
"Units for cumulative exposure are not directly comparable among studies. See footnote on page 22.
Figure 2. Dose-response Curve for.Asbestosis Mortality in a Group of Chrysotile Miners and Millers.
-------
I
10
8
t
60 r
50
0)
en
|
O)
| 40
o>
c
o
"? 30
20
10
0
J
1,000
2.000
3,000
4,000
5,000
6,000
7,000
8.000
Cumulative exposure (long-fiber equivalents per cubic centimeter x years)"
Source: Sluis Cremer and duToit (1973).
"Slope determined by the formula, slope = S xy/Sx .
"Converted by the authors from concentrations measured in million particles per cubic foot x years. Units for
cumulative exposure are not directly comparable among studies. See footnote on page 22.
Figure 3. Dose-response Curve for X-ray Signs of Asbestos in a Group of South African Miners and Millers of
Amosite and Crocidolite.
-------
80
60
40
slope = 0.26a
CD
>
o
VI
I
§
O
09
Q.
8
i
20
0
8
6
4
2
0
50 100 150 200 250
Cumulative exposure (f-yr/cm^) b
B
slope = 0.021a
300
50 100 150 200 250
Cumulative exposure (f-yr/cm^)b
__I
300
2
1
slope = 0.00968
o -*'•
0
1
50
i
100
1
150
i
200
i
250
i
300
Cumulative exposure (f-yr/cm^)
Source: Berry, as reported in Acheson and Gardner (1979).
aSlopes determined by the formula, slope = xy /
Units for cumulative exposure are not directly comparable among studies
See footnote on page 22.
Figure 4. Dose-response Curves for (A) Crepitations, (B) Possible Asbestosis and
(C) Certified Asbestosis in a Group of Asbestos Textile Workers
-38-
-------
Possible asbestosis represents a judgment by the factory medical
officer, based on periodic physical examinations and x-rays, that
the disease has progressed to the extent that a worker should
move to a less dusty job. The diagnosis of certified asbestosis
qualifies a patient for workmen's compensation (McVittie 1965).
Without extrapolation, the curves in Figure 4 show the following
annual incidence rates of asbestosis for workers with previous
cumulative exposure of approximately
25 f-yr/cm3:
certified asbestosis 2 cases/10,000 workers/year
possible asbestosis 5 cases/10,000 workers/year
crepitations 65 cases/10,000 workers/year.
This represents the lowest level of cumulative asbestos exposure
at which severe forms of asbestosis have been detected.
These studies of dose-response relationships imply that the
risk of asbestosis is proportional to cumulative asbestos
exposure. .Because the curves do not demonstrate a "no-adverse-
effect level" of exposure, signs of asbestosis may well result
from exposure levels lower than those present in the asbestos
factories, mines, and mills that were studied.
The above implication is borne out by the results of studies
that show that signs of less severe stages of asbestosis can
occur in individuals exposed to asbestos outside the workplace.
The most important results are those reported by Anderson and co-
workers (1979), who found a high prevalence of lung abnormalities
on the x-rays of children and other persons living in the same
households as asbestos workers (Table 10). Persons sharing
-39-
-------
Table 10. Lung Abnormalities Detected on X-rays
of Persons Sharing Households with Asbestos Workers
Group
No. of persons
examined
No. of x-rays with
one or more abnormality
I
•J^
O
Controls
All household contacts
Sons and daughters only
-<1 year of exposure only
325
679
375
192
15 (4.6%)
239 (35.2%)a
109(29.1%)a
47 (24.5%)a
Source: Anderson et al. (1979)
Probability that the difference from control value resulted by chance alone is less
than 0.001 (Two-tailed chi square test. See Fleiss 1973).
-------
households for less than 1 year with persons actively employed as
asbestos workers had a percentage of abnormalities five times
higher than that of controls. The asbestos concentrations in
these homes are not known, but they are presumed to have been
many times lower than those to which the workers were exposed at
their places of employment. Thus, persons sharing households for
less than 1 year with persons actively employed as asbestos
workers had very low levels of cumulative exposure.
Another recently reported study concerns office workers whose
only known asbestos exposure was from sprayed-on insulation
materials in office buildings in Paris (Awad et al. 1979). These
individuals received medical examinations that included a
determination of the presence-of "crackling rales"
(crepitations). Of office workers employed for 10 or more years
in building areas with "low protection" but no "specific
exposure," only 0.4% had crepitations. The prevalence was three
times higher (1.3%) among workers who were present during
construction of the building but who, at the time of the survey,
worked in buildings free of asbestos contamination. The highest
prevalence (2.5%) was found among employees having direct contact
with "ceilings, sheaths, cupboards, etc." coated with asbestos-
containing materials. These results are no reported completely
and, because of the small number of persons with crepitations,
cannot be ruled out the possibility that these findings should be
attributed to chance. Nevertheless, if validated, they will form
the first direct evidence of asbestosis among occupants of
buildings that, like many school buildings in the United States,
were constructed with sprayed-on asbestos-containing materials.
-41-
-------
In South Africa, Sluis-Cremer and duToit (1979) found lung
abnormalities characteristic of asbestos exposure (e.g., pleural
calcification) on the x-rays of nonworkers residing near asbestos
mines. The prevalence increased with duration of residence in
the area, another demonstration of a dose-response relationship.
The three studies of nonoccupational exposure cited above
support the inference from occupational dose-response curves that
lung damage characteristic of asbestosis can occur in persons
exposed to asbestos concentrations lower than those in
occupational settings. The two best sources of information for
predicting whether signs of asbestosis can result from asbestos
exposure levels found in schools would be the findings among
household contacts of asbestos workers (Table 10) and the dose-
response curves in Figure 4. Unfortunately, the absence of data
on asbestos concentrations in workers' homes prevents a direct
comparison with the situation in schools. The lowest level of
cumulative exposure actually measured in Figure 4 (25 f-yr/cm3)
is approximately 100 times the highest estimate for adults
employed in school buildings (using a conversion factor of 1
•5 -5
f/cm = 33,000 ng/m ; see Table 18). [Unlike the assessment of
cancer risks, in which extrapolation is warranted by the current
scientific understanding of carcinogenic processes and by
regulatory policy, an extrapolation of asbestosis risks over two
orders of magnitude of cumulative exposure may be unduly
speculative.] Some noncancerous lung damage probably will result
from asbestos exposure in schools, but the extent of damage
cannot be predicted with a reasonable degree of confidence.
-42-
-------
e. Summary and conclusions
Epidemiologic research has identified cancers of the lung,
pleura, peritoneum, larynx, oral cavity, esophagus, stomach,
colon and kidney as hazards of asbestos exposure. Inhalation of
asbestos also produces the non-cancerous lung disease
asbestosis. Dose-response relationships (increasing risk
correlated with increasing asbestos exposure) have been shown or
suggested for cancer of the lung, larynx, and stomach, pleural
and peritoneal mesothelioma, and asbestosis. Two studies of
respiratory cancer among asbestos workers provide results
compatible with linear nonthreshold dose-response curves.
These dose-response studies imply that asbestos exposure can
increase the risk of cancer at lower exposure levels than those
studied. This expectation is supported by evidence of adverse
health effects resulting from relatively low levels of asbestos
exposure. Increased lung cancer risk has been observed among
workers exposed to asbestos for the equivalent of 5 years at the
current workplace standard of 2,000,000 f/m3. Mesothelioma, a
"marker disease" for asbestos exposure, has occurred in persons
with exposures as brief as 1 or 2 days and in persons with steady
exposures as low as those found in the homes of asbestos workers
and in neighborhoods around asbestos mines, products factories,
and shipyards. X-ray signs of asbestosis have been detected
among persons sharing households with actively employed asbestos
workers for less than a year. Linear nonthreshold dose-response
curves predict that asbestos exposure in schools will produce
adverse health effects (see Part D). This prediction is
-43-
-------
consistent with the occupational dose-response curves and with
the observation of increased risks of disease at exposure levels
lower than those found in the workplace.
Part B 3, below, identifies factors that influence the degree
of increased risk posed by asbestos exposure. Part D, estimates
risks of cancer mortality expected to result from exposure to
asbestos in school buildings.
3. Factors that Modify the Risk of Asbestos-Induced
Disease
a. Smoking
The major factor affecting the risk of asbestos-induced lung
cancer, other than the intensity and duration of asbestos
exposure, is the smoking habits of exposed individuals.
Although, as shown below, asbestos exposure alone and cigarette
smoking alone can each cause lung cancer in humans, the combined
effects of cigarette smoking and asbestos exposure produce an
increase in lung cancer risk that is greater than the sum of the
increases produced by the two agents independently. In one
study, a group of 283 asbestos insulation workers who smoked had
a lung cancer mortality rate approximately 90 times greater than
the rate they would have had if they had been neither smokers nor
asbestos workers (Selikoff et al. 1968). In a more recent study
of 17,800 asbestos insulation workers, the rate was 50-60 times
greater (Hammond et al. 1979). As discussed below, this latter
study also showed that the combined effect of smoking and
asbestos exposure exceeded the sum of their separate effects, an
indication that the effects of asbestos and smoking interact or
modify one another in some way.
-44-
-------
In the earlier study of asbestos insulation workers (Selikoff
et al. 1968)/ there were no lung cancer deaths among 87
nonsmokers. This study prompted speculation that asbestos might
increase lung cancer risk only in smokers (Cole and Goldman 1975,
Hoffmann and Wynder 1976). The current evidence/ however, shows
that asbestos exposure induces lung cancer in smokers and
nonsmokers alike. The recent study by Hammond and colleagues
(1979), found a fivefold increase in the risk of lung cancer
among 891 nonsmoking asbestos insulation workers. Because it
covered a larger group of nonsmoking workers over a longer
follow-up period, the study of Hammond et al. had a higher
probability of detecting an increase in risk than the earlier
study. Asbestos exposure, therefore, increases lung cancer risk
even in the absence of cigarette smoking (Selikoff and Hammond
1979).
In the study by Hammond et al., the asbestos workers who
smoked cigarettes could have avoided about the same increase in
lung cancer risk if they had not been asbestos workers as they
could have if they had not been smokers. In 1978, Selikoff
supplied EPA with a set of unpublished data from this study that
enables estimates to be made of the proportion of lung cancer
deaths among the cigarette-smoking asbestos workers that can be
attributed to smoking alone, asbestos alone, interaction of the
effects of asbestos and smoking, and unknown factors (Table 11).
Estimates were made of the number of deaths that could be
expected among the smoking asbestos workers if they had been
neither smokers nor asbestos workers (E^), if they had smoked but
-45-
-------
Table 11. Observed and ExpectecHing Cancer Deaths > 20 Years from
Onset of Exposure in a Group of Asbestos Workers with
a History of Cigarette Smoking.
Lung cancer deaths No. of deaths9
Observed (0) 305
Expected on the basis of:
Nonsmoking non-asbestos workers (E-j) 4.4
Smoking non-asbestos workers (E?) 57.5
Nonsmoking asbestos workers (E3f 35.0
Attributable to:
Factors other than smoking or asbestos (£••) 4.4 ( 1.4%)
Smoking alone (E2 - EJ 53.1(17.4%)
Asbestos alone (E3 - Ej) 30.6(10.0%)
Asbestos /smoking interaction (0-lEj + (E2 - E^ + (Eg-E^j) 216.9(71.1%)
Source: Unpublished data supplied to EPA by Selikoff (1978)
aThe most recently published results from this study (Selikoff et al. 1979a, Hammond et al. 1979) report
0 = 306, E.J = 4.7; E2 and E3 were not reported. We are requesting updated figures for 0, E^, E2, and E3
from these researchers.
-------
had not been exposed to asbestos (E2)/ and if they had never
smoked but had been exposed to asbestos (EO. For £•,, the lung
cancer mortality rates of a group of nonsmoking non-asbestos
workers from a large study (Hammond 1966) sponsored by the
American Cancer Society (ACS) were used. For E2/ the rates for
smokers in the ACS study were used. The rates of the nonsmoking
colleagues of the smoking asbestos insulaton workers were used to
derive En*
The results, summarized in Figure 5, are nearly identical to
the estimates derived by Lloyd (1979) using published data from
the same study and a different method of derivation. Less than
2% of the lung cancer deaths among the cigarette-smoking asbestos
workers were attributable to causes other than smoking and
asbestos exposure; over 70% were the result of some sort of
interaction between the effects of the two agents. The most
important implication is that approximately 81% of the lung
cancer deaths could have been prevented if none of the men had
been asbestos workers and about 88% could have been prevented if
none had been smokers. Thus, from a preventive standpoint, the
impacts of smoking and asbestos exposure on lung cancer risk were
approximately equal in this group of workers.
-47-
-------
00
I
Smoking
alone
(17.4%)
Asbestos
alone
(10.0%)
Asbestos-smoking
interaction
(71.1%)
Unknown factors
(1.4%)
Figure 5. Proportions of Lung Cancer Deaths Attributable to Known and Unknown
Factors in a Group of Cigarette-smoking Asbestos Workers
-------
Smoking may also be an important factor in increasing an
individual's susceptibility to asbestosis. Although asbestosis
occurs in persons who smoke and in those who do not smoke, (e.g,
Berry et al. 1979, Hammond et al 197.9), Frank (1979) reported
that asbestos insulation workers with a history of cigarette
smoking had an asbestosis mortality rate 2.9 times higher than
that of workers who had never smoked regularly. In addition,
several morbidity studies have found that clinical and diagnostic
signs of asbestosis in asbestos workers are more prevalent among
smokers than nonsmokers (Langlands et al. 1971, Weiss 1971, Weiss
and Theodos 1978, Harries et al. 1975, Ayer and Burg 1978,
Mitchell et al. 1978, Rossiter and Berry 1978, Berry et al.
1979).
Because none of these morbidity studies included a comparison
group of persons who smoked, but who had not been occupationally
exposed to asbestos, the "polyvalent" (Becklake 1973) or
nonspecific nature of many of the .diagnostic signs of asbestosis
(i.e., signs that can be produced either by smoking or by
exposure to asbestos) could not be taken into account.
Consequently, the mortality study (Frank 1979) provides the
strongest evidence that smokers are at greater risk of asbestosis
than nonsmokers under similar conditions of asbestos exposure.
Data from this mortality study also suggest that there may be
a greater risk of pleural mesothelioma among cigarette smokers
exposed to asbestos than among nonsmokers similarly exposed
(Selikoff 1977a). As shown in Table 12, the rate of pleural
mesothelioma mortality among the smokers was more than twice the
-49-
-------
Table 12. Mesothelioma Mortality Rates in a Group of
Asbestos Insulation Workers, by Smoking History
Mesothelioma deaths/10,000 person-years
Smoking history
Pleura! Peritoneal
, Never smoked regularly 1.6 7.1
Ul
^ Cigarettes 3.8 7.3
Pipe or cigar 9.7 11.3
Unknown 2.5 3.7
Source: Selikoff (1977a).
-------
rate among the nonsmokers. (There is no evidence that smoking,
by itself, can cause pleural mesothelioma.) In contrast, the
peritoneal mesothelioma mortality rates were very similar for
both cigarette smokers and nonsmokers. The rates in Table 12
were not adjusted for age or duration of asbestos exposure, but
they suggest that reevaluation of the conclusion that pleural
mesotheliomas "occur with equal frequency among smoking and
nonsmoking asbestos workers" (IARC 1977) would be worthwhile.
The high mortality rates from both pleural and peritoneal
mesothelioma among workers who smoked only pipes or cigars are
also worthy of note. Unpublished data supplied to EPA by
Selikoff in 1978 indicate that this small group of workers also
had a very high asbestosis mortality rate. Although the possible
effect of smoking on the risk of pleural mesothelioma should be
explored, it is not likely that this effect (if any) will be
found to be nearly as large as the effect of smoking on the risk
of asbestos-induced lung cancer.
b. Age
The highly active nature of school children and their
physical characteristics generate concern that, under similar
circumstances, their degree of actual exposure to asbestos may be
greater than that of adults (Kane 1976). Because children
generally are more active than adults, they have a higher
breathing rate. They also inhale relatively more often through
the mouth than through the nose. Consequently, more fibers would
be inhaled and fewer would be trapped by the nasal hairs and
mucosa. Young children are shorter than adults and their mouths
-51-
-------
and noses are closer to the floor. Therefore, they are likely to
inhale higher concentrations of dust that is stirred up from the
floor. Children also have a greater remaining life span, during
which the chronic effects of asbestos exposure can become
manifest.
It has also been suggested that children may be more
biologically susceptible than adults to carcinogens, including
asbestos (Kotin 1977, Wasserman et al. 1979). Kotin (1976)
stated that "...in the induction of cancer, it is the very young
that is always the most susceptible." Other observers (Doll
1962, Cole 1977), hold the issue to be far from settled. Kotin
(1979) reflected the uncertainty by observing more recently that
"special biological susceptibility has not been demonstrated" for
children exposed to asbestos.
One epideraiologic study and one experiment with rodents shed
some light on this question with regard to pleural mesothelioma.
After examining the incidence of this cancer in an epidemiologic
study of a group of asbestos textile workers, Peto (1979) stated
that "the incidence 30 years after first exposure appears to be
much the same irrespective of age at first exposure." The
incidence rates were not provided in his report; nevertheless, if
the annual incidence is not affected by age at first exposure,
then persons exposed earlier in life experience higher lifetime
risk. Consistent results were reported in the experiment with
rodents by Berry and Wagner (1976), who injected crocidolite into
the pleurae of two groups of rats: one at age 2 months and the
other at age 10 months. In the group exposed at the earlier age,
-52-
-------
40% developed mesothelioraa; in the latter group, the incidence
was only 19% (0.005 < p < 0.01, two-tailed chi square test).
Neither of these studies could evaluate the age-dependent
decline in the respiratory clearance of fibers that occurs in
humans, at least among smokers (Cohen et al. 1979), and possibly
among nonsmokers (Wanner 1977) as well. This decline in
clearance capacity might greatly increase the proportion of
inhaled fibers that reach the pleura. Therefore, the possibility
cannot be ruled out that pleural tissue in young persons may be
more susceptible but, because of the relatively unimpaired
respiratory clearance in these individuals, less severely exposed
than pleural tissue in older persons.
The two studies discussed above apply solely to pleural
mesothelioma, which is only one of the hazards of asbestos
exposure. The empirical relationship of age at first exposure to
the risk of other asbestos-induced diseases remains an unexplored
subject.
c. Fiber size and type
A great deal of research and discussion has been devoted to
possible variations in risk posed by durable fibers differing in
size and chemical composition. Because these factors are not
expected to play a major role in the assessment of risks due to
asbestos exposure in schools (see part d, Summary and
Conclusions, below), they are treated only briefly here.
The primary research relating fiber size to carcinogenic
potency applies only to pleural mesothelioma, and it involves the
direct injection or implantation of fibers into the pleurae of
-53-
-------
rats. These studies strongly suggest that fibers of certain
sizes are more potent in producing mesothelioma than other-sized
fibers of identical or different chemical composition (Stanton
and Layard 1978; Stanton 1973; Stanton et al. 1977; Smith et al.
1969; Wagner et al. 1970, 1973 and 1977; Smith and Hubert
1974). As a whole/ this research indicates that fibers less than
1.5 microns in diameter and between 5 and 60 microns long,
regardless of chemical composition, are likely to be more
carcinogenic in the pleura than shorter or wider fibers. The
evidence is not sufficient, however, to label fibers with
dimensions falling outside this range (especially short, thin
fibers) noncarcinogenic.
Fiber size also helps to determine the ability of inhaled
fibers to reach the pleura. Because the airways of the lung
diminish in size as they branch outward, longer fibers are more
likely to become deposited on the ciliated surfaces of the upper
airways than shorter fibers (Dement and Harris 1979). This early
interception of longer fibers may account for the autopsy finding
of a higher percentage of longer fibers in the lung tissue than
in the surrounding pleura among persons with asbestos-related
disease (Sebastien et al. 1979b). Additionally, longer fibers
are less readily cleared from the lung than shorter fibers,
especially from the alveoli: the small, saclike pouches that
terminate the airways of the lungs (Morgan 1979). Thus, fiber
size is an important factor in the transport of inhaled fibers.
Evidence of variation in toxicity according to the chemical
composition of asbestos fibers is less firm. There are some
-54-
-------
indications of slight differences in toxicity among the various
types of asbestos (Acheson and Gardner 1979), but these
differences may result in part from the different fiber size
characteristics of the asbestos types.
There is no evidence that the fiber size distributions to
which the epidemiologically studied insulation workers were
exposed differed substantially from the distribution of sizes of
fibers released from asbestos materials in schools. In addition,
all asbestos fiber types found in schools (e.g., chrysotile,
amosite, crocidolite) are carcinogenic. Consequently, separate
consideration of the health effects of the individual
mineralogical types or fiber sizes of asbestos in this assessment
is not warranted.
d. Summary and conclusions
Smoking greatly increases the risk of asbestos-induced lung
cancer. Although asbestos causes lung cancer in both nonsmokers
and smokers, smokers exposed to asbestos have a greater risk of
developing this disease than would be expected by adding the
separate effects of smoking and asbestos exposure. Smokers also
may be at a higher risk of asbestosis than nonsmokers with
similar asbestos exposure. Current data on the possible
influence of smoking on the risk of asbestos-induced pleural
mesothelioma are not persuasive one way or the other. In
estimating the risk of lung cancer from exposure to asbestos in
schools, smokers and nonsmokers will be considered separately
when appropriate data become available from the insulation
workers study. Current evidence indicates that most of the
-55-
-------
increase in lung cancer risk among a group of smokers
occupationally exposed to asbestos could have been prevented
either by their never having been exposed to asbestos or by their
never having smoked.
Although children may be more susceptible to the effects of
asbestos exposure than adults, little firm evidence is available
to determine the differences in risk. The longer remaining life
expectancy of children compared with that of adults is the only
factor that can be incorporated into quantitative risk estimates.
Experimental evidence strongly suggests that fibers of
certain sizes that reach the pleura, regardless of chemical
composition, are more potent in producing mesothelioma than
fibers of other sizes. The use of data from a study of asbestos
insulation workers for quantitative risk estimates (see Section
III, Part D) should avoid any major uncertainties that might
*
otherwise have been presented by this finding. Because there are
no data indicating that the fiber types or sizes to which the
insulation workers were exposed were substantially different from
those present in schools, the types and sizes in both settings
will be assumed to be similar.
C. Exposure Assessment
This section assesses the amount of asbestos that inhabitants
of schools containing friable asbestos materials are being
exposed to by applying current data on airborne asbestos
concentrations in various types of buildings to the situation in
schools. The results are a quantitative estimate of exposures to
the "prevalent" level of asbestos in schools and a qualitative
-56-
-------
description of exposures to "peak" levels. A description of the
methods used to make the quantitative estimates and a discussion
of how closely the estimates apply to schools are included.
The' prevalent concentration of airborne asbestos fibers is
the one present most of the time in areas of activity in
buildings. Peak concentrations are those resulting from specific
activities such as damage to or repair of asbestos-containing
materials, and they generally are high, localized, and of short
duration. For our purposes, prevalent levels are those
determined by monitoring areas and taking measurements of
asbestos concentrations over long periods of time, and peak
levels are determined by taking measurements of concentrations
resulting from specific activities over short periods of time.
Because area monitoring data are the only consistent data
currently available on concentrations of airborne asbestos fibers
in buildings and because these data are not likely to include
peak concentrations systematically, only exposure to prevalent
levels of asbestos in schools can be estimated quantitatively.
The area monitoring data do not include peak concentrations
systematically for two reasons: (1) peak releases occur
sporadically; (2) peak concentrations are limited to very small
areas. Only if continuous area monitoring were being conducted
at the same time as and very near a specific peak release would
the monitoring data reflect peak exposure concentrations. In
addition, the possible mechanisms by which asbestos is dispersed
in buildings also preclude a quantitative estimate of exposures
at peak levels, as explained below.
-57-
-------
1. Asbestos Dispersion Mechanisms
The concentration of airborne asbestos fibers is determined
by how quickly fibers enter the air and by how quickly they are
removed. In buildings containing friable asbestos-containing
materials, fibers can be released from these materials and enter
the air in several ways (Sebastien et al. 1978, Nicholson et al.
1978a, Sawyer and Spooner 1978). Mechanisms of fiber release
such as disturbance of the building materials by air currents
will release fibers over a wide area and thus an elevate the
prevalent concentration. Other mechanisms of release such as
cutting the materials will release a large amount of fibers
locally and over a short period of time and, thus, cause peak
fiber concentrations (up to several thousand times higher than
prevalent concentrations). In addition, fibers that have been
removed from the air by settling or by impacting on surfaces
(i.e., desks, light fixtures, and floors) can be resuspended in
air either diffusely or in the form of peak releases by
activities such as dusting, sweeping, maintenance work, etc.
Fibers released during peak episodes eventually become widely
dispersed, and are either removed slowly (over periods of hours
to days) from the air by settling or impacting on surfaces or are
removed when ventilation exchanges indoor and outdoor air. This
wide dispersion also elevates the prevalent concentration. Table
13 gives airborne asbestos fiber concentrations that were
measured in various buildings (Sawyer and Spooner 1978). The
measurements include those of peak concentrations produced by the
peak release of fibers directly from asbestos materials (for
-58-
-------
example, 2f, g, and h in Table 13) and as a result of
resuspension (for example, 2c, d, e, i, and j in Table 13).
2. Estimate of Prevalent Exposures
The prevalent exposure levels in schools containing friable
asbestos materials were estimated by averaging the asbestos
concentrations measured in various buildings. The exposure
estimates were based on data from a study by Sebastien et al.
(1978) of several buildings in Paris. These data, which are
given in Table 14, were not taken in a way which would represent
the contribution of peak episodes of exposure. The choice of
which specific measurements of asbestos concentration within a
building to use in exposure estimates depended on certain
assumptions as to what the measurements would represent. Three
different assumptions were made to give three different estimates
of prevalent exposure (Table 15) that are applicable to all
buildings containing accessible friable asbestos materials. A
discussion of how these different assumptions apply to exposure
in schools and why the data of Sebastien et al. (1978) were used
to make these estimates is given below.
-59-
-------
Table 13. Optics I Microscope Analysis of Airborne Asbestos Fiber Concentrations in Various Buildings
Sampling conditions or situation
Mean counts ND. of
If/cm"*) samples
Standard
deviation
1. University dormitory, UCLA.
Exposed friable surfaces, 98% amosite.
General student activities
2. Art and Architecture Building. Yale
University. Exposed friable ceilings,
20% chrysotile.
a. Ambient air, City of New Haven
Fallout
b. Quiet conditions
Contact
0.1
0-0.8
(range)
c. Cleaning, moving Looks in stack area
d. Relamping light fixtures
e. Removing ceiling section
f. Installing track light
g. Installing hanging lights
h. Installing partition
Reentrainment
i. Custodians sweeping, dry
j. Dusting, 'dry
k. Proximal to cleaning (bystander exposure)
General Activity
3. Office buildings. Eastern Connecticut. Exposed
friable ceilings, 5 - 30% chrysotile.
• Custodial activities, heavy dusting
4. Pnvfite homes, Connecticut.
Remaining pipe lagging (dry)
amosrte and chrysotile asbestos
5. Laundry: contaminated clothing, chrysotile
6. Office building, Connecticut. Exposed sprayed.
•ceiling, 18% chrysotile.
Routine activity
Under asbestos ceiling
Remote from asbestos ceiling
7. Urban grammar school. New Haven. Exposed
ceiling, 15% chrysotile asbestos.
Custodial activity: sweeping, vacuuming
8. Apartment building. New Jersey; heavy
housekeeping. Tremolite and chrysotile
9. Office buildings. New York City
Asbestos in ventilation systems
Quiet conditions and rountine activity
15.5
1.4
17.7
7.7
1.1
3.1
1.6
4.0
0.3
0.2
2.8
4.1
0.4
79s
99s
403
643*
296a
2.5-200*
3
2
3
6
5
4
5
6
J ~v
36
8
8
12
3
2
1
2
1
6.7
0.1
8.2
2.9
0.8
1.1
0.7
1.3
0.3
0.1
1.6
,1.8-5.8
(range)
0.1-1.2
(range)
40-110
(range)
186-1,100
(range)
c.
0-800
(range)
Source: Sawyer and Sponner (1978)
aNanograms/cubic meter. Determined by electron microscope.
-60-
-------
I
Ot
Table 14. Measurements of Asbestos Concentrations in Several
Paris Buildings Used to Estimate Prevalent Exposure Levels of Asbestos
Building8
Ground floor of
research building "A"'
Rooms in research
building "A"
"B" hangar
"C"
"H" .
"K" railroad
station (open walls)
n» tt
"O"
Sampling sites
Parking lots,
laos, workshops
Libraries, labs
workshops
Workshops
Dining room
Labs, workshops
Parking lot
(open walls)
Mail room
Laboratory
Mean cone.
(ng/m3)
215
55
70
29
23
16
20
38
Max. cone.
(ng/m3)
750
630
490
29
130
24
34
62
Individual samples
(ng/m3)
751,518,19,2,0.6,0.
630,460,420,225,106.
48,46.37,31.28,15,15
14,13.9,7.6,6,(21 measurements
less than 5 ng/m3)
492,65,30,24,7,6,5,2,1
28.8
134,23.14.12.11,6,5,2,1
24,12,11
34,18.17,12
62,13
i
Source: Sebastien et al. 1979.
aDesi(jnations are those used by the authors.
-------
Table 15. Estimated Prevalent Exposure Levels of Asbestos
(Applicable to all buildings containing exposed friable asbestos materials)
Assumptions used in Predicted concn.
making estimates (ng/m^)
I. Mean for a building represents
the prevalent level 58
o>
•^ II. Maximum for a building
represents the prevalent level 270
III. Average of the few highest
concentrations for all buildings
represents the prevalent level 500
-------
Estimate I is the mean of the mean asbestos concentration in
each building (the mean of column 3, Table 14). For each
building, this estimate gives-equal weight to each measurement of
the airborne fiber concentrations but overall Estimate I gives
greater weight to individual measurements in buildings where many
measurments were taken (e.g. building A). An accurate estimate
of population exposure would require that each measurement be
weighted according to the number of people exposed at that
concentration. Estimate I approximates this by using the mean of
the means rather than the mean of all measurements. The
measurements listed in Table 14 were made in areas where
activities similar to school activities take place, and,
therefore, they represent the asbestos concentrations that occur
in activity areas in schools. However, they are not accurately
weighted according to the distribution of school populations.
Estimate II in Table 15 is the mean of the maximum
concentrations of asbestos measured in each building (the mean of
column 4, Table 14). The Agency believes this estimate gives
more weight to areas of maximum human activity because it is
likely that areas where the maximum asbestos concentrations are
measured are areas of maximum activity. This hypothesis is
supported by data which show that human activity can increase
airborne asbestos concentrations by 50 to 200 times (Sebastien et
al. 1979a). The major limit to using estimate II is that there
is no way to verify that the highest measurements were obtained
in the areas of greatest human activity. The overall average
exposure to asbestos in buildings containing exposed friable
asbestos materials is likely to be between estimates I and II.
-63-
-------
Estimate III in Table 15 is the mean of une four highest
maxima in column 4 of Table 14 (750, 630, 490, and 130 ng/m3)..
Estimate III may account for the future deterioration of friable
asbestos materials now in place. Sebastien et al.(1978),
describe the sites where the value used in estimate III were
obtained as areas where asbestos materials have deteriorated. In
the future, friable asbestos materials that are now in place
likely will deteriorate through damage or be subjected to
maintenance activities such as cutting and drilling. The
assumptions in using estimate III to predict future exposure are
that deteriorated material is responsible for the few highest
measured levels, that all materials eventually will deteriorate,
and that when materials do deteriorate, they will cause
significantly high prevalent asbestos concentrations. There are
insufficient data to document these assumption-s.
These three estimates lead to the conclusion that the current
average exposure to asbestos in buildings containing accessible
friable asbestos materials^/ is not likelv to be less than 58
ng/m , it may be as high as 270 ng/itr, and, in the future, it may
become as high as 500 ng/m . Of course, these are estimates of
exposure to the prevalent concentration. Peak exposures add
significantly to the overall exposure of specific groups of
people. For example, as shown in Part D, janitors can easily be
exposed to an average level of asbestos fibers that is more than
twice the prevalent level.
6/ Materials not enclosed by a solid partition such as a
suspended, or false, ceiling.
-64-
-------
The estimates are based on data from the Sebastien et al.
(1978) study because this study meets the following three
criteria: (1) applicability of the data to exposure in schools; •
(2) consistency, reliability, and accuracy of the measuring and
sampling techniques; and (3) adequate data on "control"
buildings. The study meets these three criteria because: (1)
the areas and materials studied are similar to those in U.S.
schools (see discussion below); (2) the measurements were made by
transmission electron microscopy (the only technique which is
accurate for environmental sampling at low concentrations - see
below), the measurements were checked by statistical quality
control techniques, and the samples were taken over relatively
long time periods (5 days); and (3) comparisons were made with
outdoor air and with a significant number of buildings that did
not contain asbestos materials. Data on asbestos concentrations
in U.S. buildings from a study (Nicholson et al. 1978a) that did
not meet these criteria were carefully assessed and used to
verify that the results of the Sebastien et al. study are
consistent with data for U.S. buildings (Logue 1980).
The specific data selected from the study of Sebastien et al
represent the exposure situation in U.S. schools. These data are
measurements of asbestos concentrations in buildings with
accessible friable asbestos materials. Enclosure of the material
(for example, with a suspended ceiling) may greatly reduce
exposure, and different enclosures will have different effects.
Too little data are available to determine whether the types of
-65-
-------
enclosures in the buildings sampled by Sebastien et al. are
similar to those in U.S. schools. Therefore, exposure estimates
are too high for buildings in which enclosed or covered asbestos
materials are present.
Friable materials were selected for the estimates because
they represent the materials of greatest concern.
Asbestos levels in buildings containing friable asbestos
materials are significantly greater than asbestos levels in
buildings which do not contain asbestos surface materials. A
statistical study (Levy, 1980) showed that there is less than a
5% probability that chance alone caused this difference.
Therefore the Agency concludes that the presence of friable
asbestos caused the difference. In addition to elevated
prevalent exposure in these buildings, peak exposures are likely
to. be frequent when friable materials are present because these
materials are easily damaged.
The selection of friable materials for the estimates does not
mean that non-friable materials do not make a significant
contribution to both prevalent and peak asbestos exposures. The
statistical study cited above, however, shows that there are
insufficient data at this time to say whether the observed
elevation of asbestos concentration in buildings containing non-
friable asbestos materials (see Table 16) is caused by the
presence of these materials or due to chance alone. Peak
exposures from non-friable asbestos materials can also occur if
-66-
-------
Table 16. Comparison of Mean and Maximum Levels of Airborne Asbestos in Buildings Containing
Friable and Nonfriable Asbestos Materials3'0
Buildings containing
Buildingb
Mean concen
friable materials
Max. concen.
airborne asbestos . airborne asbestos
A-G
A-St
B
E
H
K
L
O
A-Ct
C
D
(ng/m3)
220
55
70
29
23
16
20
20
13
0.1/
3.0
(ng/m3)
750
630
490
29
130
24
34
62
28
0.2
5
Building
Buildingb
containing nonfriable materials
Mean concen.
airborne asbestos
F
T
S
G
1
J
P
Q
R
(ng/m3)
19
21
0.1
1.7
0.4
3.2
3.2
0.83
8.6
Max concen.
airborne asbestos
(ng/m3)
40
68
0.1
2.8
2.1
7.1
7.1
1.3
12
Source: Sebastien et al. (1978)
aAII asbestos measurements in buildings without asbestos-containing materials were less than 5 ng/m3.
"Building notations are those of Sebastien et al.
cBoth enclosed and exposed asbestos-containing materials are included in this table.
-------
they are cut or drilled. However, such materials are far less
susceptible to damage than are friable materials.
The application of estimates based on various types of French
buildings to U.S. schools is possible/ because (1) the materials
containing asbestos and, as shown by comparing Table 1 and Table
14, the uses of areas in the French buildings are the same as
those in U.S. schools; and (2) the French data on asbestos levels
in rooms with friable materials are not statistically different
from comparable data (Nicholson et al. 1978a) for U.S. buildings
(Logue 1980). Asbestos-containing materials in France and the
United States are also similar in that the French processes for
applying these materials (Sebastien et al. 1978) are similar to
the processes used in the United States (cf. Section II of this
document). In both cases, "friable" coatings are produced by
mixing the asbestos with binders after the material leaves the
spray nozzle. In addition, both French and United States data
reveal that most of the airborne fibers detected are chrysotile
and that the accessible asbestos material is located in similar
places.
The use of transmission electron microscopy techniques is
necessary for the identification and measurement of asbestos
fibers outside of the workplace.. The optical microscopy
techniques, especially the phase contrast microscopy technique
recommended for use in the workplace (HEW 1976), are not suitable
for measurement of low airborne asbestos concentrations in
buildings because the phase contrast technique cannot distinguish
between asbestos and many other fibers and, because the
-68-
-------
measurement accuracy is limited by the small number of fibers
counted (HEW 1976). At low fiber concentrations a larger
proportion of the fibers will be non-asbestos fibers and, thus,
the ability to distinguish asbestos fibers from other fibers
becomes important. For this reason, transmission electron
microscopy, supplemented when necessary with electron diffraction
to specifically identify asbestos, is the only tool suitable for
measuring airborne asbestos concentrations outside of the
workplace. -L-
3. Description of Peak Exposures
Significant "peak" releases of asbestos fibers will occur and
cause the total exposure of an individual to be higher than the
estimated prevalent asbestos concentration. Students, teachers,
and school administrators will only occasionally encounter peak
exposures. Janitors, custodians, and maintenance workers will
encounter them more frequently. Available data are insufficient
to estimate the frequency with which either group would encounter
peak exposures.
7/ Because both asbestos and non-asbestos fibers are counted,
measurements made by phase contrast microscopy are expected
to be higher than those made by electron microscopy. This is
found in the results of Byron, Hodson and Holms, (1969).
They measured fiber concentrations in schools (and other)
buildings by the phase contrast microscopy technique and
found that 11 of 18 schools with sprayed asbestos had
concentrations greater than .005 fibers/cm . This
corresponds to (using 30 fibers/ng as the conversion) to
about 200 ng/m which is much higher than what Sebastien et
al. (1978) found by electron microscopy (see Table 14).
-69-
-------
Peak exposures occur during episodes of damage to friable
asbestos materials, repair or renovations involving the
materials, cleaning operations in buildings that contain the
materials, or maintenance work performed in spaces that enclose
the materials (e.g., in crawl spaces over false ceilings). Peak
exposures that occur during damage episodes will affect all
i
occupants, including students, teachers, and school
administrators; in general, however, maintenance, janitorial, and
custodial personnel will experience peak exposures most
frequently. The total exposure of these groups to asbestos may
be much greater than their exposure to the prevalent level.
D. Risk Assessment
1. Procedure for Estimating Risks of Premature Death
a. Outline of the Risk Estimation Procedure*
The number of people expected to die prematurely from
exposure to asbestos in school buildings can be predicted within
limits from available epideraiologic data. The initial step in
the risk assessment procedure is to choose the most appropriate
epidemiologic study or studies to serve as the basis for making
the estimates. The key criteria are that a study contain a
quantitative characterization of cumulative asbestos exposure and
that the study population exhibit an increase in risk of
premature death following asbestos exposure. A statistical model
of the relationship between cumulative asbestos exposure and
subsequent increases in risk (dose-response model) is then
established. The cumulative asbestos exposure of the building
occupants is estimated using the prevalent asbestos
-70-
-------
concentrations (see Part C above), the number of persons likely
to be exposed and the probable duration of exposure for those
persons. Combining the exposure estimates with the dose-response
model allows risk estimates to be expressed as the number of
premature deaths expected to occur. In this analysis, such
estimates are made for each of three groups of school occupants:
students, teachers and administrative staff, and custodians and
maintenance workers. Because of the necessary assumptions and
uncertainties in quantitative risk assessment, three risk
estimates will be presented for each group: the minimum, maximum,
and most reasonable predictions of the increases in carcinogenic
risk expected to result from asbestos exposure in schools.
Although asbestos exposure in schools likely will also
produce signs of asbestosis that are not severe enough to result
in death, this risk cannot be estimated quantitatively with
currently available data (see Section III, Part B above). In
addition, available mortality studies do not reflect the
increased incidence of cancer because some cancers (e.g., larynx
cancer) frequently are treated with success. These necessary
omissions lead to underestimates of the risk of developing
nonfatal asbestosis and treatable cancers. This risk assessment
is further restricted to a consideration only of exposure to
prevalent levels of asbestos in schools. Increased risks
resulting from exposure to peak levels have not been included in
the overall risk estimate because the frequency of these
exposures is unknown.
-71-
-------
b. Selection of the Underlying Study.
The epidemiologic study selected to be the basis for making
quantitative estimates of the risk of premature death from
exposure to asbestos in schools is a large study of asbestos
insulation workers reported most recently by Hammond et al.
(1979) and Selikoff et al. (1979a). In the original report of
mortality among these workers, the men were described as
"building trades insulation workers" who were chosen for their
"asbestos exposure of limited extent and intensity" (Selikoff et
al. 1964).
The data that will be used from this ongoing study concern
12,051 men who were employed in asbestos insulation work for at
least 10 years (Hammond et al. 1979, Selikoff et al. 1979a). The
results for this group are restricted to the time commencing at
the 20th year after each worker's first exposure, each worker's
20th year since first exposure. The diseases caused by asbestos
exposure appear after an induction period-- a minimum length of
time following initial exposure that must elapse before risk
begins to increase. In this study and in others (Peto 1978,
Berry et al. 1979, Seidman et al. 1979), the minimum induction
period for mortality from asbestos-induced diseases generally has
been reported to be 10-20 years. The use of data that cover only
the period that starts X20 years following first exposure allows
for the induction of asbestos-induced tumors. The results,
therefore, pertain only to the time subsequent to each worker's
20th anniversary of employment, the time at which he was at
increased risk of dying from the diseases that are hazards of
asbestos exposure.
-72-
-------
The mortality experience of the 12,051 workers was observed
during the 10-year period from January 1, 1967, to December 31,
1976 (the "follow-up period"). As each of these workers was
actively employed on January 1, 1967, and had reached the 20-year
point from initial exposure at some time before the end of the
follow-up period, each worker was exposed to asbestos for at
least 10 years. Workers who had reached the 20-year mark prior
to January 1, 1967, were traced throughout the follow-up period
(i.e., they entered observation on January 1, 1967). Those who
reached the 20-year mark at some time- during the follow-up period
were followed only from that point on (i.e., they entered
observation on the date of the 20th anniversary of employment).
The reported increases in risk, therefore, took place >_20 years
from initial exposure among 12,051 workers, each of whom
previously had been exposed to asbestos for at least 10 years.
In addition to allowing for cancer induction by providing
data restricted to the period that starts >2Q years from first
exposure, the asbestos insulation workers study has a number of
attributes that make it uniquely suitable as a basis for
quantitative risk estimation. No other study combines all of
these useful attributes:
0 The sample of 12,051 workers surviving >2Q years from
first exposure is very large, minimizing the probability
of chance results.
0 Reasonable estimates of the average asbestos
concentrations to which insulation workers were exposed
are available (see Part c below).
0 Each of the diseases identified as hazards of asbestos
exposure was investigated and was found to be in excess
(see Table 17 below).
-73-
-------
0 The death certificates were assiduously verified with
supplemental information (e.g., autopsy reports,
histological specimens) in 86% of the deaths (Selikoff
et al. 1979a), leading to a greatly improved detection
of frequently misdiagnosed mesotheliomas (Newhouse and
Wagner 1969, Selikoff et al. 1979a).
0 A highly appropriate control group was used, for which
smoking-specific results are available (Hammond 1966).
0 The material to which the insulation workers were
exposed (commercial asbestos, primarily chrysotile) was
very similar and, in some instances, identical to the
asbestos present in school buildings.
c. Asbestos Exposure Among the Insulation
Workers.
The measure of exposure that will be used in this risk
assessment is "cumulative exposure1*—the product of the average
asbestos concentration (in this study, it is expressed as
nanograras per cubic meter of air, ng/m3) times the number of
years of exposure to this concentration. Therefore, cumulative
^
exposure, which is expressed here in units of ng-yr/mj,
incorporates the intensity and duration of exposure into a single
measure. This system of measuring exposure has the disadvantage
of assuming implicitly that brief, high-intensity exposure is
equivalent to extended, low-intensity exposure. For instance, a
person exposed to 100 ng/nr for 10 years and a person exposed to
1,000 ng/m for 1 year would both be assigned cumulative exposure
values of 1,000 ng-yr/ra . The influence of this assumption on
the risk estimates for cancer mortality under the linear
nonthreshold dose-response curve will be discussed later.
During the 1940's and 1950's, when the insulation workers in
the underlying study received the bulk of the asbestos exposure
responsible for risk increases observed during the follow-up
-74-
-------
period (1967-1976), airborne asbestos concentrations in the work
environment probably were higher than the concentrations measured
in more recent years. Because reliable monitoring data are not
available for the earlier period, approximations must be made on
the basis of the recent measurements and in light of changing
work practices and conditions. Nicholson (1976) reviewed several
monitoring studies and concluded that "the overall time-weighted
average exposure of United States asbestos [insulation] workers
in the late 1960's was less than 3 f/ml" (3,000,000 f/m3). This
estimate, made under the assumption that insulation workers in
the late 1960's worked with asbestos-containing materials only
half of the time, agrees closely with the figure of 4,200,000
f/ra3 derived by the National Institute for Occupational Safety
and Health (NIOSH 1972) for full time asbestos insulation work.
Consequently, 3,000,000 f/ra3 represents the lowest reasonable
estimate of .the average asbestos concentration to which workers
in the underlying study were exposed.
Nicholson (1976) also found that, during the late 1960's,
"work practices were virtually identical to those of the past,
and ...few controls of significance were in use." Nevertheless,
he identified two major changes in the conditions of insulation
work over the years. First, workers in the 1940's and 1950's
were in contact more often with insulation containing asbestos,
as opposed to insulation containing fibrous glass and other
materials that recently have become more popular. Second, the
asbestos content of insulation materials containing asbestos
declined by as much as one-half over the period ranging from the
-75-
-------
1940's and 1950's to the late 1960's. These factors helped lead
Nicholson (1976) to state that "insulators' average exposures in
the United States during the past years could have ranged from 10
to 15 f/ml" .(10,000,000-15,000,000 f/m3). Therefore, 15,000,000
f/m3 is the highest reasonable estimate of the average exposure
level or the asbestos insulation workers.
The most reasonable estimate lies between 3,000,000 and
15,000,000 f/m3. If it is assumed that insulation workers in the
late 1960's handled asbestos one-half of the time, that previous
workers handled asbestos three—fourths of the time (a 50%
increase), and that older insulation materials containing
asbestos had twice the asbestos content of newer asbestos-
containing materials, the recent average exposure level
(3,000,000 f/m3) can be multiplied by a factor of 3 to yield an
estimate of the earlier concentration. This conversion yields a
"most reasonable" estimate of approximately 9,000,000 f/m3 for
the average asbestos concentration to which the workers in the
underlying study were exposed.
The units in which the minimum, maximum, and most likely
average exposure levels are expressed can be converted from
fibers per cubic meter to nanograms per cubic meter. In a study
conducted for the Office of Pesticides and Toxic Substances, EPA
(Versar 1980)r it was concluded that for insulation work, a
fiber-to-mass conversion ratio of 30 f/ra to 1 ng/m is the best
approximation if fibers are counted by light microscopy. This
factor is in general agreement with data on fiber size
distributions for the asbestos industry as a whole (Dement and
-76-
-------
Harris 1979) and for insulation work in particular (Nicholson
1976). It should be remembered that this conversion factor is
rough and currently cannot be verified because the nature of the
industry has changed. It is, however, the best available
estimate.
Conversion of the three estimates of average asbestos
exposure for the insulation workers studied by Selikoff's group
yields the following estimates:
maximum 500/000 ng/m3
most reasonable 300,000 ng/nr
minimum 100,000 ng/m3
It is important to determine the period of exposure to these
concentrations that should be held reponsible for the increases
in risk detected "during the observation period (1967 through
1976). Asbestos-induced increases in risk do not appear until
>IQ years after exposure (Peto 1978, Seidman et al. 1979), so the
attributable exposure period for each worker in this risk
assessment ends 10 years prior to the time the worker entered
observation.
A . B C D
| < > | < 10 yr > I < 10 yr > |
Under the approach described above, A-B (ending for most
workers on December 31, 1956) is the period of attributable
exposure. C-D is the follow-up period, during which all exposure
is presumed to be "wasted" in the sense that it is not
responsible for increases in cancer risk during the same
period. 3-C is an additional period of time during which
exposure is considered to be "wasted." If the minimum induction
-77-
-------
period for death from asbestos-induced neoplasms were exactly 10
years, the greatest degree of underestimation of exposure would
result from the fact that exposure during year 3+1 ended, not 10
years, but 13 years before year D-l began. The choice, however,
of a length of 10 years for the additional "wasted" exposure
period (B-C) is a conservative one. The minimum induction period
for mesothelioma, for instance, appears to be closer to 20 years
than to 10 years (Selikoff et al. 1979a). For lung cancer,
Peto's (1978) data show the minimum induction period to be about
15 years. If the minimum induction period for these diseases
were >20 years, no relevant exposure would be ignored by choosing
10 years for the length of period B-C.
Thus, although a certain degree of exposure relevant to
increased risk during the follow-up period (C-D) is likely to be
ignored under this approach, 10 years for B-C is considered a
reasonable length in order to optimize three goals: (1) to make
use of the published mortality data from the insulation workers
study; (2) to avoid attributing "wasted" exposure to increased
risk during the follow-up period; and (3) to avoid labeling
exposure "wasted" that actually contributed to increased risk
during the follow-up period. The attributable exposure period
for each of the 12,051 workers was his period of employment
ending 10 years before he entered observation. The total number
of years of relevant exposure for the group divided by 12,051
yields the average exposure period. [Note: We are requesting
this total from the researchers. For the time being, a figure of
20 years will be used as the mean attributable exposure period in
the calculations. When the actual value becomes available, it
will replace the 20-year figure.]
-------
It was estimated above that the average exposure level for
the insulation workers was between 100,000 and 500,000 ng/m^,
with the most reasonable estimate being 300,000 ng/nr. Over a
20-year average exposure period, these figures yield the
following estimates of cumulative exposure:
maximum 1.0 x 10 ng-yr/m^
most reasonable 6.0 x 10^ ng-yr/nr
minimum 2.0 x 10 ng-yr/m
The three values are estimates of the average cumulative
asbestos exposure of the 12,051 workers at the time 10 years
before observation began. Many continued to be exposed, but
exposure beyond that point is not thought to have contributed to
the observed increases in risk.
d. Increased Risk Among the Asbestos Insulation
Workers.
Following the 20-year induction period, the researchers
compared the observed number of deaths from specific causes among
the 12,051 workers to the number expected on the basis of
mortality rates in an appropriate comparison group ._^_ The
greater number of observed than expected deaths indicates
increased risk. The results for cancer deaths are shown in Table
17. The use of 95% confidence intervals for the observed number
8/ The data for the control or comparison group were obtained
from a large study sponsored by the American Cancer Society
(Hammond 1966) of the age-, calendar year-, and smoking-
specific experience of white males with at most a high school
education and a history of occupational exposure to dust,
fumes, vapors and gases, excluding farmers.
-79-
-------
Table 17. Mortality Data Taken from the Study of 12,051 Asbestos Insulation
Workers and Used To Make Quantitative Estimates of Risk from
Asbestos Exposure in Schools
1
00
o
1
Cause of death
Cancer, all asbestos-
relaied sites
Lung
Pleura
Peritoneum
Larynx, buccal
cavity, pharynx
Esophagus
Kidney
Colon-rectum
Stomach
Expected
deaths (E|)a
145.8
81.7
0
0
7.5
5.1
8.5
30.5
12.5
Observed
deaths (Oj)
692
397
61
109
21
17
15
54
18
95% Confidence
limitsb
Lower
641.4
358.9
46.6
89.5
13.0
9.9
8.4
40.6
10.7
Upper
745.6
438.1
78.4
131.5
32.1
27.2
24.7
70.5
28.4
Statistical
significance
level0
< 0.001
<0.001
<0.001
< 0.001
< 0.001
<0.001
0.027
< 0.001
0.084
Source: Hammond et al. (1979)
aNumber of observed deaths based on death certificate information only, except for pleura! and
peritoneal mesothelioma. Supplemental information was used for these two cancers. This
[procedure was recommended by Hammond et al. (1979).
^Method of Bailar and Ederer (1964), assuming a Poisspn distribution of observed deaths. Values
from Documunta Geigy (1970), some by linear interpolation.
cMethod of Bailar and Ederer (1964), assuniing a Poisson distribution of observed deaths. One-tailed
test, values from Molina (1942).
-------
of deaths is a way of accounting for the role of chance variation
in the results. In comparisons with the expected number of
deaths, the minimum risk estimate is provided by the lower 95%
confidence limit for observed deaths, the maximum estimate by the
upper 95% confidence limit for observed deaths, and the most
likely risk estimate by the actual number of deaths observed.
The measure of increased risk most useful for predicting
premature mortality from asbestos exposure in schools is the
difference between the observed ( Oj_) and expected ( Ej)
numbers of deaths from the cancers related to asbestos exposure
divided by the total number of deaths expected from all causes
(ET = 1,148.0) (Hammond et al. 1979). This measure is the
fraction of all expected deaths that were "in excess" because of
the asbestos exposure. It is called "lifetime risk" (LR) by the
EPA Carcinogen Assessment Group, Office of Research and
Development, EPA, and is defined as follows: LR= (Oi-Ei)/ET. If
the study were carried out until all were deceased (actually,
only 16% of the 12,051 workers died during the observation
period), (Qj_-Ej_) would equal the total number of "excess," or
premature deaths.
In using lifetime risk, as defined above, as the measure of
increased cancer risk in this assessment, certain assumptions
must be made. First, it must be assumed that the estimate of
lifetime risk when only 16% of the workers have died will be the
same when all 12,051 have died. Second, it must be assumed that
this estimate of lifetime risk among persons exposed as adults
will be indicative of the lifetime experience of exposed school
-81-
-------
children, who have a greater remaining period of expected life
during which the effects of asbestos exposure can become
manifest. The use of the lifetime risk measure, therefore, does
not allow the greater remaining life expectancy of children to be
taken into account and for this reason may underestimate risk.
The mortality rates for each of the cancer hazards of
asbestos exposure were increased among the insulation workers
(Table 17). The data for the separate causes of death can be
combined in order to estimate overall lifetime risk for all
asbestos-induced cancers:
maximum » (745.6 - 145.8)/l,148.0 = 0.522
most reasonable = (692 - 145.8)/l,148.0 = 0.476
minimum = (641.4 - 145.8)71,148.0 = 0.432
The results give a most reasonable estimate that the overall
mortality rate was increased by 48% above the expected value by
asbestos-induced deaths from cancer. (Additional premature
deaths from asbestosis are not included here.) The above
lifetime risk estimates will be used to predict the risks of
mortality from asbestos exposure in schools.
e. Asbestos Exposure in Schools
In Section III, Part C, three prevalent asbestos
concentrations were estimated for school buildings. Estimates I
(58 ng/m3) and II (270 ng/m3) were developed to reflect current
concentrations and Estimate III (500 ng/m ) to reflect
concentrations in the future, as the building materials
deteriorate. The risk assessment concerns exposures over the
next 30 years. Consequently, Estimates I and III will be used as
-82-
-------
the minimum and maximum future concentrations, respectively.
Estimate II, although developed as a maximum estimate of the
current concentration, will be used as the most reasonable
estimate of future concentrations.
Three groups of school building occupants are considered:
students; teachers and administrative staff; and custodians and
maintenance workers. It was estimated in Section II that
approximately 3,000,000 students, 222,000 teachers and
administrative staff, and 23,000 custodians and maintenance
workers are occupying schools that contain friable asbestos
materials (See Section II above). These numbers of exposed
school occupants at risk of death from asbestos-induced cancers
need to be adjusted to reflect the number expected to die before
a minimum period of time from first exposure has elapsed.
Adopting the same 20-year minimum induction period as in the
insulation workers study and assuming that the average student is
first exposed at age 12 and the average adult school occupant at
age 30, national life tables (NCHS 1978) can be used to estimate
that 2.2% of exposed students and 5.9% of exposed adults will die
before 20 years have elapsed from first exposure. The estimated
number of school occupants at risk, then is 3,000,000 - 2.2% =
2,934,000 persons exposed while attending school, 222,000 - 5.9%
= 208,900 teachers and administrative staff, and 23,000 - 5.9% =
21,600 custodians and maintenance workers. The average remaining
service time for the buildings is approximately 30 years (See
Part II above). Therefore, the average cumulative exposure for
the adult occupants is equal to the prevalent asbestos
-83-
-------
concentration times 30 work years.l^ For students, it is the
prevalent concentration times 15 work years. These estimates are
shown in Table 18.
The calculations were made under the assumption that each of
the current school occupants will be exposed for the entire 30-
year period that the buildings will remain in service. Although
this assumption is not technically correct (as students graduate
and adults leave their positions, others will replace them)/ as
long as the exposed populations continue to average 3,000,000,
222,000, and 23,000, respectively, the risk estimates will not be
affected. This is because of the nature of the cumulative
exposure measure (1,000 ng-yr/m resulting from either 10 years
at 100 ng/m3 or 1 year at 1,000 ng/m3) and the linear
nonthreshold dose-response model (cumulative exposure of 1,000
persons to 1,000 ng-yr/m yielding the same number of premature
deaths as cumulative exposure of 100 persons to 10,000 ng-yr/m3).
f. Selection of the Extrapolation Method
Once the most suitable epideraiologic study of asbestos
workers has been chosen for the risk assessment, a method must be
selected for using the results of the study to predict the
9/ A work year is assumed to be made up of 50 weeks at 5 days a
week and 8 hours a day. A school year is assumed to be made
up of 33 weeks at 5 days a week and 6 hours a day. /
Therefore, 2 school years equal one work-year:
33 weeks x 5 days x 6 hours = 0.5.
50 weeks 5 days 3 hours
-84-
-------
Table 18. Estimates of Cumulative Asbestos Exposure in Schools3
o
Cumulative exposure levels (ng-yr/m J) for:
i
CO
en
1
Population group
Students
Teachers, administrative
. staff
Custodians, maintenance
workers
Average
number exposed
2,934.00
208,900
21.600
Minimum
estimate of risk
870
1,740
1.740
Most reasonable
estimate of risk
4,050
8,100
8.100
Maximum
estimate of risk
7,500
15,000
15,000
aAssuming 30 work years of exposure per adult and 15 work years of exposure per student. See discussion on page 92.
-------
increase in carcinogenic risk that will result from asbestos
exposure in schools. The prediction is made by extrapolating^0^
from the exposure levels experienced by the asbestos insulation
workers to the lower levels of asbestos to which school occupants
are exposed. A dose-response curve is developed to describe the
relationship between cumulative asbestos exposure and subsequent
increases in cancer risk. From this curve, predictions of
increased risk can be derived that correspond to cumulative
exposure levels lower than those for which epidemiologic data are
available. Because the empirical relationship of dose to
response at these exposure levels is unknown, criteria must be
established for selecting the most appropriate dose-response
curve.
Two general types of evidence can be used to show that a
dose-response curve or model is unsuitable: (1) knowledge of the
biological processes that influence the degree to which inhaled
asbestos increases carcinogenic risk, and (2) the dose-response
data available from epidemiologic studies or experiments with
laboratory animals. The first type of information, often called
"pharmacokinetics," includes a carcinogen's "absorption,
distribuiton, reactions with cellular components, and
elimination," as well as its interaction with physiologic
10/ If, as in most risk assessments, it is assumed that the dose-
response curve passes throught the point corresponding to
zero exposure and zero increase in risk, the prediction is
technically an interpolation. Nevertheless, the conventional
term, extrapolation, will be used here.
-86-
-------
mechanisms of activation and detoxification (Gehring et al.
1977). This kind of information, if available, can lead to
inferences about the shape of the dose-response curve at low
doses, including the possibility of a threshold dose below which
the risk of cancer would not be increased (Cornfield et al.
1977). EPA is unaware of information about the pharmacokinetics
of asbestos that would enable such inferences to be drawn.
The second type of evidence, dose-response data from
epidemiologic and toxicologic studies, is available for asbestos-
induced carcinogenicity. Table 19 shows the results of
statistically "fitting" several dose-response models that have
been developed for cheraicas carcinogens to data from two studies:
an epidemiologic study of asbestos workers (Henderson and
Enterline 1979, see also Figure 1-B) and an experiment with rats
(Wagner et al. 1974). By the usual and widely accepted criterion
that a p-value greater than 0.05 or 0.10 indicates an adequate
fit (Remington and Schork 1970), none of the models can be
dismissed on the basis of these studies.
Current scientific evidence alone cannot be used to select
the most appropriate dose-response curve for this extrapolation
because none,of the curves in Table 19 can be ruled out on the
basis of pharmacokinetics or available dose-response data. Of
these curves, however, linear nonthreshold regression (see, e.g.,
Figure 1) usually provides the highest predictions of increased
risk and there is no strong scientific reason to prefer any of
-87-
-------
le 19. Dose-respite Curves Applied to Two Studies of
Asbestos Exposure and Carcinogenic Response
Goodness-of-f it p-value
Dose-response curve Reference Epidemiologic study9*" Experiment with rats"
One-hit Brown (1976) 0.58 °-°°
. Multi-hit Van Ryzin and Rai (1980) 0.42 °13
Multi-stage (1 stage) Crump (1980) 0.58 011
Multi-stage (5 stages) Crump (1980) 0.53° °°fjC
Linear regression Neter and Wasserman 0.87 0.10
(1974)
"Henderson and Enterline (1979)
bWagneretal. (1974)
cThe results of a Monte Carlo simulation
, "The "p" values calculated from the chi square (X 2) statistic are based on the difference
oo between the observed (Oj) and the expected (Ej) counts in the ith dose group. The
^ degrees of freedom equal number of dose levels-1-(number of parameters estimated).
X2 = E (O--E-)2/E-
eln the epidemiologic study, each measure of response concerns a group of people with a
unique age distribution; hence, the "background" mortality rates will differ among the
groups. To fit models with this type of data, it is necessary to adjust the observed response
to what they would be if there were a common "background" rate. The risk attributable
to the carcinogen is calculated from Abbott's equation:
P = "close ""control*' *-control*
The adjustment to common "background" is done by recalculating the observed response,
pdose- as p' dose • Where P' con|ro| represents the common "background" rate:
p' — P/1 P' \ + P'
' r dose m-r control' r control
-------
the dose-response curves that yield lower estimates of increased
risk. Consequently, the Agency has chosen to take the most
prudent course with respect to public health by using a linear
nonthreshold dose-response curve to extrapolate from the asbestos
exposure levels experienced by the insulation workers to the
lower levels of asbestos to which school occupants are exposed.
2. Risk Estimates for School Building Occupants.
Figures from Tables 17 and 18 and the. preceding discussion
are summarized in Table 20. They are arranged in the way that
yields minimum, most reasonable and maximum estimates of
increased risk for school building occupants.
When the linear nonthreshold model is used, calculation of
predicted risk levels is fairly simple. For instance, as shown
in Table 21, the minimum lifetime cancer risk for school children
would be:
(0.432 x 870 ng-yr/m3)/(1.0 x 107 ng-yr/ra3) = 3.8 x 10~5.
Multiplying this figure by the number of students yields the
minimum estimate of premature deaths:
(3.8 x 10 ~5) x (2,934,000) = 111.
-89-
-------
I
vo
o
I
Table 20. Summary of Cumulative Exposure and Lifetime Risk Estimates To Be Used in
Quantitative Risk Assessment of Asbestos Exposure in Schools
Estimates leading to:
Minimum
Population group ™k estimate
Insulation workers 1.0 x 10'
Students 870
Adult schools occupants 1,740
Insulation workers 43.2
Most reasonable
risk estimate
O
Cumulative exposure (ng-yr/m );
6.0 x 106
4,050
•
8,100
Lifetime cancer risk (%)
47.6
Maximum risk
estimate
2.0 x 106
7.500
15,000
52.2
-------
Table 21. Quantitative Risk Estimates of Mortality from Exposure to Asbestos in Schools
Lifetime risk
o , Most
Population group Minimum reasonable
2.934.000 students 3.8 ;< 10~5 3.2 x 10~4
208.900 teachers, t- ,,
administrative svaff 7.5 x 10"° 6.4 x 10~*
21.600 custodians, K -
maintenance workers 7.5 x 10~° 6.4 x 10"^
• -
Premature deaths
Most
Maximum Minimum reasonable Maximum
2.0 x 10~3 111 960 5.868
3.9 x 10~3 16 142 814
3.9 x 10'3 2 15 84
Totals 129 1,117 6,766
-------
The results of this and similar calculations ara shown in
Table 21. A total of approximately 100 to 7,000 premature deaths
are. anticipated to occur as a result of exposure to prevalent
concentrations of asbestos in schools containing friable asbestos
materials over the next 30 years. The most reasonable estimate
is approximately 1,000 premature deaths. About 90% of these
premature deaths are expected to occur among persons exposed as
school children. The remaining 10% include teachers, custodians,
and other adult occupants of the buildings. The most reasonable
estimates represent extrapolations of approximately four orders
of magnitude from the exposure levels experienced by the
insulation workers.
The risk estimates in Table 21 are subject to further
refinement. For instance, the influence on risk of the greater
remaining life expectancy of children compared with that of
adults has not yet been incorporated into the assessment. In
addition, when supplemental information requested from the
investigators who conducted the insulation workers study is
supplied, it will alter these estimates to some degree. The
information requested is:
0 the total number of person-years of exposure accumulated
by the 12,051 workers up to the time 10 years before
they entered observation;
0 the number of person-years of observation and the number
of observed and expected deaths from lung cancer,
asbestosis, and pleural raesothelioma by smoking history;
0 the number of observed and expected deaths due to cancer
of the colon separate from those due to cancer of the
rectum.
-92-
-------
This additional information, with the possible exception of
the smoking-specific data for lung cancer, is not expected to
have a major effect on the overall results of the risk
assessement. At present, an assumption is implicit that the
distribution of smoking habits among exposed school occupants
will be the same as among the insulation workers.
It is important to emphasize that the risk estimates in Table
21 concern only a portion of the total adverse impact on health
expected to result from asbestos exposure in schools. The less-
than-fatal effects of asbestos exposure on lung function and the
number of cases of certain types, of cancer that may be treated
successfully (e.g., larynx cancer) are not included in this
quantitative risk assessment. The substantial but unquantifiable
risks resulting from peak exposures also are absent from the
assessment.
-93-
-------
IV. IDENTIFICATION OF FRIABLE ASBESTOS-CONTAINING MATERIALS IN
SCHOOLS
A. Introduction
In order to control the release or resuspension of asbestos
in schools or other types of buildings/ it is necessary to
determine whether asbestos is, in fact, present in the bulk
building materials. This determination can be made by examining
building records and by analyzing samples of the materials.
Records will not establish conclusively that asbestos is not
present in a building, as they may be incomplete or there may
have been a substitution (e.g., of asbestos for nonasbestos
fibers) of the components in the material that was applied. The
magnitude of the risks involved makes it necessary to take
additional, more extensive steps such as sampling/analysis to
ensure the accurate identification of friable asbestos-containing
materials.
To locate friable materials, it is necessary to visually
inspect the steel support beams, columns, ceilings, and walls of
all areas of the school. Asbestos-containing materials also may
have been applied to hidden areas, such as those above a
suspended ceiling, and they must be checked. Inspectors should
direct particular attention to boiler rooms and other equipment
areas, in view of the frequent use of asbestos as insulating
material.
Procedures for inspecting buildings and taking samples of
friable materials and guidelines for establishing the presence of
asbestos in these materials are described in two recent EPA
publications (EPA 1979b and EPA 1980, respectively).
-94-
-------
3. Sampling
A sample of friable material can be obtained by penetrating
the depth of the material with a small canister or jar or by
dislodging the material with a knife. An amount equal to 2
tablespoons is sufficient for analysis. Proper sampling requires
that each sample container be tightly sealed, wiped clean with a
damp cloth, and labeled. The label should be recorded by the
sampler (EPA 1979b).
Samples must be taken in a manner that will provide a
representative indication of the composition of the material.
The amount of asbestos in the friable asbestos-containing
materials on a building surface may vary. If the material is
homogeneous in appearance and was applied at one time/ the amount
may not vary greatly over one surface area. From three to seven
samples may be needed, however, to establish whether asbestos is
present and, if so, the approximate percentage that is present.
EPA recommends that 3 samples be taken for homogeneous surfaces
2
that are up to 1,000 ft / 5 samples for surfaces that are between
1,000 and 5,000 ft , and 7 samples for surfaces that are >5,000
ft2 -(EPA 1980).
A random selection of sampling sites is necessary to
eliminate the bias that may result from taking samples from
convenient locations. Representative sampling can be achieved by
extracting material from different places within a sampling area
(close to walls, at joints, etc.). A more involved method for
the random selection of sample sites (EPA 1980) involves the use
of a random number table and a diagram of the area to be sampled.
-95-
-------
Friable material is disturbed during the sampling process,
and asbestos fibers, if present, may be released. Release of and
consequent human exposure to asbestos can be minimized by taking
samples when the area is not in use and limiting the number of
persons present, lightly spraying water on the area to be sampled
to discourage dust formation, holding the sample container away
from the face, and wet cleaning the area if any pieces are
dislodged and fall to the floor.
C. Analysis
Three analytical methods can be used to identify asbestos
fibers in bulk materials. The first, polarized light microscopy
(PLM), uses the different refractive indices, birefringence, and
other optical crystallographic properties of asbestos minerals to
distinguish them from nonasbestos ones. PLM also characterizes
and identifies other fibers such as glass fibers and cellulose.
The second method, x-ray diffraction (XRD), uses the unique
diffraction pattern produced when x-rays strike any crystalline
material to identify specific asbestos minerals. The third,
electron microscopy (EM), uses electron diffraction or energy-
dispersive x-ray analysis to identify asbestos fibers by
examining the structure of individual -fibers.
EPA's Guidance Manual on Asbestos Analytical Programs (SPA
1980) recommends PLM as the method of choice for determining
asbestos in suspect material and XRD as a backup technique to
confirm the PLM analysis. Although electron microscopy can be
used, it is not recommended, because only very small quantities
of sample can be analyzed at one time and the analysis of
multiple samples is prohibitively expensive.
-96-
-------
The EPA Environmental Monitoring Systems Laboratory, Research
Triangle Park, North Carolina, has prepared and currently is
field testing interim PLM and XRD analytical protocols to be
followed in identifying asbestiform minerals .in bulk samples.
The protocols clarify and refine the guidance originally offered
in Appendix H of the Guidance Manual. They have been circulated
to laboratories currently participating in the Technical
Assistance Program. An Asbestos Particle Atlas with color PLM
photomicrographs has been developed by McCrone Research
Institute. The Atlas is available from Ann Arbor Press.
EPA has identified and complied a list of laboratories that
analyze bulk samples for asbestos using PLiM. This list is based
in part on the laboratories' successful participation in a
proficiency analytical testing program. A report on this testing
program will be available in September, 1980. Copies of the list
can be obtained from EPA by calling the following toll-free
number: 800-344-8571, extension 6892.
-97-
-------
V. CONTROL OF ASBESTOS IN SCHOOLS
This section presents information on the steps that can be
taken to control exposure to asbestos in school buildings once
friable asbestos-containing materials have been identified.
EPA has published guidance materials on the corrective
actions that can be taken in schools and other buildings if
asbestos-containing materials are found to be damaged or
deteriorating. Long-term solutions to the release of asbestos
fibers from these materials are removal, encapsulation, or
enclosure. Removal eliminates the source of contamination,
enclosure (with a barrier such as a suspended or false ceiling)
reduces the likelihood that incidental contact with the asbestos-
containing material will occur, and encapsulation (with an
effective sealant) reduces the likelihood that fibers will be
released into the building environment.
Exposure to asbestos in buildings also can be controlled to
some extent by a number of other actions, most of which are aimed
at reducing physical contact with asbestos-containing surfaces.
These actions simply interrupt the process by which asbestos
fibers enter building air. Asbestos fibers enter a school
environment from friable asbestos-containing materials as a
consequence of:
(1) disturbance of the material during maintenance or
renovation operations, implementation of the long-term
corrective actions described above, and vandalism;
(2) fallout encouraged by normal activity in the building;
and
(3) resuspension of settled fibers caused by normal activity
or custodial dusting or cleaning.
-98-
-------
Usually, asbestos enters the air as a result of physical
contact with asbestos-containing material. Contact can cause
significant amounts of fibers to be released to the air,
resulting in airborne concentrations that frequently exceed
industrial standards (Sawyer 1977).
Loosely compacted friable materials are more likely to
release fibers than tightly bound materials. When a friable
material was brushed by hand to simulate mild damage, fiber
counts as high as 3.3 f/cm3 were measured as far away as 10 feet
from the site of the damage (Nicholson et.al. 1978a). In
contrast, counts of 0.2 f/cm were noted when a cementitious
material was brushed (Table 17 in Nicholson et.al.).
Repair, renovation, or maintenance of buildings may bring
about the highest airborne concentrations of airborne fibers,
because these activities disturb asbestos-containing materials
directly. Sanding or cutting asbestos-containing solid materials
during construction or repair produces the greatest release of
fibers. Incidental contact that occurs when other maintenance
chores (e.g., installing a lighting unit) are performed can lead
to significant release. In addition, damage to the material from
vandalism, maintenance work, or, simply, deterioration can
increase the rate of fiber release by fallout (Sawyer 1977). In
schools, there is the additional opportunity for damage of
friable materials by students. Whether it is the result of
normal school activity (such as throwing a ball around a
gymnasium) or acts of vandalism, damage caused by students can be
significant.
-99-
-------
Small amounts of asbestos can fall spontaneously from
ceilings or walls/ building up the airborne fiber concentration
over time. Low harmonics and vibrations caused by machinery and
other sources also can increase the release of fibers. Once a
wall or ceiling is damaged, it can shed fibers without
significant further disturbance. These fibers can accumulate
around a room and be continually resuspended any time there is
movement of the air. Accumulation of asbestos fibers caused by
fallout can be significant.
Finally, the resuspension of fibers that have been released
can continue to cause asbestos exposure. Cleaning or other
maintenance work or the movement of people through an area can
cause settled fibers to be resuspended. Suspended ceilings can
hide the accumulation of fibers until maintenance work causes the
suspended ceiling to be disturbed; this could result in the
release of a large amount of fibers to the air (Sawyer 1977).
Custodial services such as sweeping and dusting also can elevate
fiber levels by disturbing material that has collected on floors
and other surfaces. Asbestos, fibers tend to stay suspended in
the air for a long time; for smaller fibers, this time may be on
the order of days. When they do settle out, the fibers can
easily be resuspended. They do not diffuse as a gas does;
rather, they tend to be confined to a given area.
Exposure to asbestos can be controlled to some extent by
reducing the physical contact of individuals with friable
asbestos-containing materials. Sawyer reported on the beneficial
effects of wet cleaning, wet handling during maintenance, and
barrier systems in inhibiting the movement of fibers in a
-100-
-------
building. The simple rearrangement of schedules so that direct
work on asbestos-containing material will occur when the building
is not in use and provision of workers with respirators also can
reduce inhalation of asbestos. Regular wet cleaning of building
surfaces can remove accumulated fallout, thus reducing the
resuspension of asbestos. Sawyer reported that wet cleaning
reduced fiber concentrations due to custodial activity from 4.0
f/cm (before control) to 0.3 f/cm . Wet cleaning is
particularly effective in reducing the exposure of the person
doing the cleaning.
General exposures throughout the building also might be
reduced somewhat as a result of wet cleaning, although no studies
have been done to show the effectiveness of regular wet cleaning
per se on the building environment. Unless care is taken in
disposing of any fibers collected during either wet or dry
cleaning, fibers will remain available to be resuspended in
building air.
Sawyer reported that during removal operations, wetting bulk
asbestos-containing materials with water containing wetting
agents reduced mean fiber counts to 8.1 f/cm3, compared with the
mean count of 82.2 f/crn3 that was calculated for dry
conditions. [Nicholson et al. (1978a) reported fiber counts of
up to only 1.78 f/cra during wet removal of asbestos-containing
materials in a New Jersey school.] Sawyer also demonstrated that
fiber levels dropped more quickly when wet methods were used.
The migration of fibers to non-work areas can be inhibited by
barriers. When removing asbestos in a New Jersey school,
Nicholson et al. isolated the work area with plastic barriers.
-101-
-------
Fiber counts outside the work area ranged from 0.01 to 0.03
f/cra3, but counts within the removal area ranged from 0.02
(during wetting) to 1.78 f/cm .
Vacuum cleaners equipped with high-efficiency particulate
absolute (HEPA) filters can collect asbestos dust. Sawyer showed
that, whereas dry dusting of shelves and books in a library
raised fiber counts to 4.02 f/cm^, use of HEPA filters raised
counts to only 0.4 f/cm . Wet wiping the shelves produced a
count of 0.2 f/cra . Household and normal industrial vacuums
without HEPA filters cannot collect asbestos fibers.
-102-
-------
VI. REFERENCES
Acheson ED, Gardner MJ. 1979. The ill effects of asbestos upon
health. In: Advisory Committee on Asbestos. Asbestos, Final
report of the Advisory Committee. Volume 2. Papers commissioned
by the Advisory Committee. London: Her Majesty's Stationery
Office.
Anderson HA, Lilis Rr Daum SM, Fischbein AS, Selikoff IJ.
1976. Household-contact asbestos neoplastic risk. Ann NY Acad
Sci 271: 311-323.
Anderson HA, Lilis R, Daum S, Selikoff IJ. 1979. Asbestosis
among household contacts of asbestos factory workers. Ann NY
Acad Sci 330: 387-399.
Arul KJ, Holt PF. 1977. Mesothelioma possibly due to
environmental exposure to asbestos in childhood. Int Arch Occup
Environ Health 40: 141-143.
Ashcroft T, Heppleston AG. 1970. Mesothelioraa and asbestos on
Tyneside: a pathological and social study. In: Shapiro HA, ed.
Pneumoconiosis: proceedings of the international conference,
Johannesburg, 1969. Capetown: Oxford University Press.
Awad L, DiMenza L, Lazar P, Bonnaud G, Bignon J. 1979. An
attempt to determine a medium-term low-dose exposure indicator
Based on clinical and radiological lung modifications. Paper
presented at the symposium on the Biological Effects of Mineral
Fibres held on 25-27 September 1979. Sponsored by the
International Agency for Research on Cancer, Lyon, France.
Ayer H, Burg J. 1978. Cumulative asbestos exposure and forced
vital capacity. Unpublished draft manuscript. Washington, DC:
National Institute for Occupational Safety and Health.
Bailar JC, Ederer F. 1964. Significance factors for the ratio
of a Poisson variable to its expectation. Biometrics 20:639-643.
Barnes R. 1976. Asbestos spraying, an occupational and
environmental hazard. Med J Aust 2: 599-602.
Baselga-Monte M, Segarra F. 1978. Epidemiological study of
asbestosis in the Barcelona area: a dose-response relationship.
Unpublished manuscript provided by the Occupational Safety and
Health Administration.
Battelle Memorial Institute, Columbus Laboratories, Columbus,
OH. 1980. Asbestos in schools. Washington, DC: Survey and
Analysis Division, Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. Contract No. 68-01-3858.
Becklake MR. 1973. Lung function. In: Bogovski ?, Gilson JC,
Timbrell V, Wagner JC, eds. Biological effects of asbestos.
Lyon, France: International Agency for Research on Cancer.
-103-
-------
Becklake MR. 1976. Asbestos-related.diseases of the lung and
other organs: their epidemiology and implications for clinical
practice. Am Rev Respir Dis 114: 187-227.
Becklake MR, Liddell FDK, Manfreda J, McDonald JC. 1979.
Radiological changes after withdrawal from asbestos exposure. 3r
J Ind Med.36: 23-28.
Berry G, Gilson JC, Holmes S, Lewinsohn HC, .Roach SA. 1979.
Asbestosis: a study of dose-response relationships in an asbestos
textile factory. Br J Ind Med 36: 98-112.
Berry G, Lewinsohn HC. 1979. Dose-response relationships for
asbestos-related disease: implications for hygiene standards.
Part 1. Morbidity. Ann NY Acad Sci 330:185-194.
Berry G, Wagner JC. 1976. Effect of age at inoculation on
occurrence of mesotheliomas in rats. Int J Cancer 17: 477-483.
Borow M, Conston A, Livornese L, Schalet N. 1973. Mesothelioma
following exposure to asbestos: a. review of 72 cases. Chest 64:
641-646.
Brown C. 1976. Mathematical aspects of dose-response studies in
carcinogenesis: the concept of thresholds. Oncology 33:62-65.
Bruckman L, Rubino RA, Christine B. 1977. Asbestos and
mesothelioma incidence in Connecticut. J Air Pollut Control
Assoc 27: 121-126.
Byrom JC, Hodgson AA, Holmes S. 1969. A dust survey carried out
in buildings incorporating asbestos-based materials in their
construction. Ann Occup Hyg 12:141-145.
Chen W, Mottet NK. 1978. Malignant mesothelioma with minimal
asbestos exposure. Hum Pathol 9: 253-258.
Cochrane JC, Webster I. 1978. Mesothelioma in relationship to
asbestos fibre exposure. A review of 70 serial cases. S Afr
Med J 54: 279-231.
Cohen D, Arai SF, Brain JD. 1979. Smoking impairs long-term
dust clearance from the lung. Science 204: 514-517.
Cole P. 1977. Cancer and occupation: status and needs of
epidemiologic research. Cancer 39(April Suppl.): 1788-1791.
Cole P, Goldman MB. 1975. Occupation. In: Fraumeni JF Jr,
ed. Persons at high risk of cancer. An approach to cancer
etiology and control. New York: Academic Press,
Cole P. 1979. The emerging case-control study. J. Chron Dis
32:15-27.
-104-
-------
Commonwealth of Massachusetts. 1978. Annual report of the
Special Commission relative to evaluating the extent of the use
of asbestos as fireproofing in the schools and public buildings
of the Commonwealth. Boston, MA. Report No. 11, 641-125-11-79-
GR.
Cook PM, Olson GF. 1979. Ingested mineral fibers: elimination
in human urine. Science 204: 195-198.
Cornfield J. 1977. Carcinogenic risk assessment. Science
198:693-699.
Crump KS. 1980. An improved procedure for low-dose carcinogenic
risk assessment from animal data. J. Environ. Pathol. Toxicol.
(in press).
Cunningham HM, Moodie CAf Lawrence GA, Pontefract RD. 1977.
Chronic effects of ingested asbestos in rats. Arch Environ
Contam Toxicol 6: 507-513.
Dement JM, Harris RL. 1979. Estimates of pulmonary and
gastrointestinal deposition for occupational fiber exposures.
Washington, DC: Department of Health, Education, and Welfare,
National Institute for Occupational Safety and Health. DHEW
(NIOSH) Publication No. 79-135.
Documenta Geigy. 1970. Scientific tables. Basel, Switzerland:
J.R. Geigy S.A.
Doll R. 1955. Mortality from lung cancer in asbestos
workers. Br J Ind Med 12: 81-86.
Doll R. 1962. Susceptibility to carcinogenesis at different
ages. Geron Clin 4: 211-221.
Edge JR, Choudbury SL. 1978. Malignant mesothelioma of the
pleura in Barrow-in-Furness. Thorax 33: 26-30.
Edge JR. 1979. Incidence of bronchial carcinoma in shipyard
workers with pleural plaques. Ann NY Acad Sci 330: 289-294.
EPA. 1973. U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Emission Standards and
Engineering Division. National emission standards for hazardous
air pollutants: asbestos, beryllium, and mercury. Fed Regist,
6 April 1973, 38: 8820.
EPA. 1975. U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Emission Standards and
Engineering Division. National emission standards for hazardous
air pollutants: amendments to standards for asbestos and
mercury. Fed Regist, 14 October 1975, 40: 292.
EPA. 1978. U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Emission Standards and
-105-
-------
Engineering Division. National emission standards for hazardous
air pollutants: amendments to asbestos standards. Fed Regist,
19 June 1978, 43: 26372..
EPA. 1979a. U.S. Environmental Protection Agency, Criteria and
Standards Division. Asbestos: ambient water quality criteria.
Fed Regist, Oct. 1, 1979, 44: 56626,
EPA. 1979b. U.S. Environmental Protection Agency, Office of
Pesticides and Toxic Substances, Survey and Analysis Division.
Asbestos-containing materials in school buildings: a guidance
document. Washington, DC: U.S. Environmental Protection
Agency. Report No. C00090.
EPA. 1980. U.S. Environmental Protection Agency, Office of
Pesticides and Toxic Substances, Survey and Analysis Division. A
guidance document for asbestos analytical programs. Washington,
DC: U.S. Environmental Protection Agency. Draft report.
Faich G. 1980. Asbestos hazard evaluation in Rhode Island
schools. Am J Public Health 70: 2.
Fleiss JF. 1973. Statistical methods for rates and proportions.
New York: John Wiley and Sons.
Forgacs P. 1967. Crackles and wheezes. Lancet 2: 203-205.
Forgacs P. 1969. Lung sounds. Br J Dis Chest 63: 1-12.
Frank AL. 1979. Public health significance of smoking-asbestos
interactions. Ann NY Acad Sci 330: 791-794.
Gardner D, Council of Educational Facilities Planning Inter-
national, Chicago, IL. April 1, 1980. Personal communication to
L. Longanecker, Control Action Division, Office of Pesticides and
Toxic Substances, U.S. Environmental Protection Agency,
Washington, DC.
Gehring PJ7 Watanabe PG, Young JD. 1977. The relevance of dose-
dependent pharmacokinetics in the assessment of carcinogenic
hazards of chemicals. In:Hiatt HH, Watson JD, Winsten JA, eds.
Origins of human cancer. Book A. Incidence of cancer in humans.
Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Gibel W, Lohs K, Horn KH, Wildner GP, Hoffman F. 1976.
Experimental study on carcinogenic activity of asbestos filters.
Arch Geschwulstforsch 46: 437-442.
Greenberg M, Lloyd Davies TA. 1974. Mesothelioma register
1976-68. Br J Ind Med 31: 91-104.
Gross P, Harley RA, Seinburne CM, Davis JMG, Greene WB. 1974.
Ingested mineral fibers. Do they penetrate tissue or cause
cancer? Arch Environ Health 29: 341-347.
-106-
-------
Hain E, Dalquen P, Bohlig H, Dabbert A, Hinz I. 1974.
Retrospective studies on the genesis of mesothelioma. Report on
150 cases of mesothelioma in the Hamburg area. Int Arch
Arbeitsmed 33: 15-37.
Hallenbeck WH, Patel-Mandlik K. 1973. Fate of ingested
chrysotile asbestos fiber in the newborn baboon. Washington DC:
U.S. Environmental Protection Agency. EPA-600/1-78-069.
Hammond EC. 1966. Smoking in relation to the death rates of
one million men and women. In: Epidemiologic study of cancer and
other chronic diseases. Natl Cancer Inst Monogr 19. Bethesda,
MD: National Cancer Institute.
Hammond EC, Selikoff IJ, Seidman H. 1979. Asbestos exposure,
cigarette smoking and death rates. Ann NY Acad Sci 330: 473-490.
Harf A, Hirsch A, Di Menzal L, Bertrand C, Bignon J. 1979.
Functionnal (sic) tests in 49 asbestos spray workers. Paper
presented at the symposium on the Biological Effects of Mineral
Fibres held on 25-27 September 1979. Sponsored by the
International Agency for Research on Cancer, Lyon, France.
Harries PG. 1973. Clinical signs. In: Bogovski P, Gilson JC,
Timbrell V, Wagner JC, eds. Biological effects of asbestos.
Conference held at International Agency for Research on Cancer,
Lyon, France, 2-6 October 1972.
Harries PG, Rossiter CE, Coles RM. 1975. Royal Naval Dockyards
Asbestosis Research Project, Report No. 1.
Henderson VL, Enterline PE. 1979. Asbestos exposure: factors
associated with excess cancer and respiratory disease
mortality. Ann NY Acad Sci 330: 117-126.
Hobbs MST, Woodward S, Murphy B, Musk AW, Elder JE. 1979. The
incidence of pneumoconiosis, mesothelioma and other respiratory
cancer in men engaged in mining and milling crocidolite in
Western Australia. Paper presented at the symposium on the
Biological Effects of Mineral Fibres held on 25-27 September
1979. Sponsored by the International Agency for Research on
Cancer, Lyon, France.
Hoffmann D, .Wynder EL. 1976. Smoking and occupational
cancers. Prev Med 5: 245-261.
Hourihane D O'B. 19.64. The pathology of mesotheliomata and an
analysis of their association with asbestos exposure. Thorax 19:
268-278.
Hughes J, Weill H. 1979. Lung cancer risk associated with
manufacture of asbestos cement products. Paper presented at the
symposium on the Biological Effects of Mineral Fibres held on
25-27 September 1979. Sponsored by the International Agency for
Research on Cancer, Lyon, France.
-107-
-------
I ARC. 1977. International Agency for Research on Cancer. IARC
monographs on the evaluation of the carcinogenic risk of
chemicals to man: asbestos. IARC monogr 14. Lyon, France:
IARC.
Jodoin G, Gibbs GW, Macklem PT, McDonald JC, Becklake MR.
1971. Early effects of asbestos exposure on lung function. Am
Rev Respir Dis 104: 515-535.
Jones JSP, Pooley FD, Smith PG. 1976. Factory populations
exposed to crocidolite asbestos-a continuing survey. In:
Rosenfeld C, Davis W. eds. Environmental pollution and
carcinogenic risks. Lyon, France: International Agency for
Research on Cancer.
Kane DN. 1976. Bad air for children. Environment 18: 26-34.
Kotin P. January 26, 1976. Presentation before the
Occupational Safety and Health Administration, Committee on
Construction Safety and Health.
Kotin P. June 9, 1977. Briefing before the Consumer Product
Safety Commission. Fed Regist, July 29, 1977, 22: 38786.
Kotin P. January 8, 1979. Written statement to the United
States House of Representatives, Committee on Education and
Labor, Subcommittee on Elementary, Secondary, and Vocational
Education, Re: Oversight hearing on the possible hazards
associated with the presence of asbestos.in schools.
Kovarik JL. 1976. Primary pleural mesothelioma. Cancer 38:
1816-1825.
Lacquet LM, van der Linden L, LePoutre J. 1979.
Roentgenographic lung changes, asbestosis, and mortality in a
Belgian asbestos-cement factory. Paper presented at the
symposium on the Biological Effects of Mineral Fibres held on 25-
27 September 1979. Sponsored by the International Agency for
Research on Cancer/ Lyon, France.
Langlands JHM, Wallace WFM, Simpson MJC. 1971. Insulation
workers in Belfast. 2. Morbidity in men still at work. Br J
Ind Med 28: 217-225.
Leathart GL. 1968. Pulmonary function tests in asbestos
workers. Trans Soc Occup Med 18: 49-55.
Levine R. ed. 1978. Asbestos: an information resource.
Washington DC: Department of Health, Education and Welfare,
National Institutes of Health. DHEW-(NIH)-79-1681.
Levy R. Design and Development Branch, Survey and Analysis
Division, U.S. Environmental Protection Agency (EPA), May 22,
1980. Comparison of building asbestos levels grouped by surface
-108-
-------
material. Memorandum to H. Teitelbaum, Chemical Review and
Evaluation Branch, Assessment Division, EPA.
Liddell FDK. 1979. Latent periods in lung cancer mortality in
relation to asbestos dose and smoking habit. Paper presented at
the symposium on the Biological Effects of Mineral Fibres held on
25-27 September 1979. Sponsored by the International Agency for
Research on Cancer, Lyon, France.
Lieben J, Pistawka H. 1967. Mesothelioma and asbestos
exposure. Arch Environ Health 14: 559-563.
Lloyd JW. 1979. Cancer epidemiology. Paper presented at the
18th Annual Medical-Legal-Industrial Symposium held on 9 November
1979. Sponsored by the Mount Sinai Medical Center,
Milwaukee, WI.
Logue E, Center for Health Studies, Research Triangle Park, NC.
20 March 1980. Similarity of Sebastien's asbestos air levels in
buildings data and the data included in Nicholson's report to
NIEHS. Memorandum to Cindy Stroup, Task Manager, Design and
Development Branch, U.S. Environmental Protection Agency.
Lumley KPS. 1976. A proportional study of cancer registrations
of dockyard workers. Br J Ind Med 33: 108-114..
MacMahon B, Pugh TF. 1970. Epidemiology: principles and
methods. Boston: Little, Brown and Company.
Mancuso TF, Coulter EJ. 1963. Methodology in industrial health
studies-the cohort approach, with special reference to an
asbestos company. Arch Environ Health 6: 210-226.
Mancuso TF, El-Attar AA. 1967. Mortality pattern in a cohort
of asbestos workers: a study based on employment experience. J
Occup Med 9: 147-162.
McDonald JC, Gibbs GW, Liddell FDK. 1979. Chrysotile fibre
concentration and lung cancer mortality: a preliminary report.
Paper presented at the symposium on the Biological Effects of
Mineral Fibres held on 25-27 September 1979. Sponsored by the
International Agency for Research on Cancer, Lyon, France.
McDonald AD, Harper A, El-Attar OA, McDonald JC. 1970.
Epidemiology of primary malignant mesothelioraa tumors in
Canada. In: Shapiro HA, ed. Pneuraoconiosis: proceedings of the
international conference, Johannesburg, 1969. Capetown: Oxford
University Press.
McDonald JC, Liddell FDK. 1979. Mortality in Canadian miners
and millers exposed to chrysotile. Ann NY Acad Sci 330: 1-9.
McEwen J, Finlayson A, Mair A, Gibson AAM. 1971. Asbestos and
mesothelioma in Scotland: an epidemiological study. Int Arch
Arbeitsmed 23: 301-311.
-109-
-------
McVittie JC. 1965. Asbestosis in Great Britain. Ann NY Acad
Sci 132: 123-138.
Meurman LO, Kiviluoto R, Hakaraa M. 1979. Combined effects of
asbestos exposure and tobacco smoking on Finnish anthophyllite
miners and millers. Ann NY Acad Sci 330: 491-495.
Milne J. 1976. Thirty-two cases of mesothelioma in Victoria,
Australia: a retrospective survey related to occupational
asbestos exposure. Br J Ind Med 33: 115-122.
Mitchell CA, Charney M, Schoenberg JB. 1978. Early lung
disease in asbestos-product workers. Lung 154: 261-272.
Molina EC. 1942. Poisson's exponential bionomial limit.
Princeton: D. VanNostrand Company,. Inc.
Morgan A. 1979. Effect of length on the clearance of fibres
from the luny and on body formation. Paper presented at the
symposium on the Biological Effects of Mineral Fibres held on 25-
27 September 1979. Sponsored by the International Agency for
Research on Cancer, Lyon, France.
Morgan RW, Shettigara PT. 1976. Occupational asbestos
exposure, smoking and laryngeal carcinoma. Ann NY Acad Sci 271:
308-310.
Murphy RLH, Ferris BG, Burgess WA, Worcester J, Gaensler EA.
1971. Effects of low concentrations of asbestos. Clinical,
environmental, radiologic and epidemiologic observations in
shipyard pipe covers and controls. N Engl J Med 285: 1271-1278.
Neter J, Wasserman W. 1974. Applied linear statistical models.
Homewood, IL: Richard Irwin, Inc.
Newhouse M. 1973. Cancer among workers in the asbestos textile
industry. In: Bogovski ? et al., eds. Biological Effects of
Asbestos. Lyon, France: International Agency for Research on
Cancer.
Newhouse ML, Berry G. 1979. Patterns of mortality in asbestos
workers in London. Ann NY Acad Sci 330: 53-60.
Newhouse ML, Thompson H. 1965. Mesothelioma of pleura and
peritoneum following exposure to asbestos in the London area. Br
J Ind Med 22: 261-269.
Newhouse ML, Wagner JC. 1969. Validation of death certificates
in asbestos workers. Br J Ind Med 26: 302-307.
Nicholson WJ. 1976. Case study I: asbestos and the TLV
approach. Ann NY Acad Sci 271: 152-169.
-110-
-------
Nicholson WJ, Rohl AN, Sawyer RN, Swoszowski EJ, Tocaro JD.
1978a. Control of sprayed asbestos surfaces in school
buildings: a feasibility study. Report to the National
Institute of Environmental Health Sciences. New York: iMount
Sinai School of Medicine, City University of New York.
Nicholson WJ, Selikoff IJ, Hammond EC, Seidman H. 1978b.
Mortality experience among asbestos factory workers; effect of
differing intensities of asbestos exposure. Paper presented at
the International Conference on Health Hazards of Asbestos
Exposure. Sponsored by the New York Academy of Sciences, June
24, 1978.
NCHS. 1978. National Center for Health Statistics. Life
tables. Vital statistics of the United States, 1976. Volume II,
Section 5, Table 5-2. Rockville, MD: NCHS, U.S. Department of
Health Education and Welfare.
NIOSH. 1972. National Institute for Occupational Safety and
Health. Criteria for a Recommended Standard: Occupational
Exposure to Asbestos.
NIOSH. December 1976. National Institute for Occupational
Safety and Health. Revised recommended asbestos standard.
Washington, DC: NIOSH, U.S. Department of Health, Education, and
Welfare. DHEW (NIOSH) Publication No. 77-169.
NIOSH-OSHA. 1980. (National Institute for Occupational Safety
and Health -'Occupational Safety and Health Administration).
Workplace exposure to asbestos: review and recommendations.
Memorandum from Asbestos Work Group to Dr. Eula Bingham, OSHA,
and Dr. Anthony Robbins, NIOSH. Washington DC: U.S. Department
of Health, Education, and Welfare, NIOSH; U.S. Department of
Labor, OSHA.
Nurminen M. 1975. The epidemiologic relationship between
pleural mesothelioma and asbestos exposure. Scand J Work Environ
Health 1. 128-137.
OSHA. 1976. Occupational Safety and Health Administration.
Standard for exposure to asbestos. Rules and regulations. Fed
Regist 44: 11504.
Peto J. 1978. The hygiene standard for chrysotile asbestos.
Lancet 1: 484-489.
Peto J. 1979. The incidence of pleural mesothelioma in
chrysotile asbestos textile workers. Paper presented at the
symposium on the Biological Effects of Mineral Fibres held on 25-
27 September 1979. Sponsored by the International Agency for
Research on Cancer, Lyon, France.
Peto J, Doll R, Howard S, Kinlen LJ, Lewinsohn HC. 1977. A
mortality study among workers in an English asbestos factory. 3r
J Ind Med 34: 169-173.
-Ill-
-------
Puntoni R, Valerio F, Santi L. 1976. II mesothelioma pleurico
frai. lavoratori del porto di Geneva. Tumori 62: 205-210.
Reitze WB, Nicholson WJ, Holaday DA, Selikoff IJ. 1972.
Application of sprayed inorganic fiber containing asbestos:
occupational health hazards. Am Ind Hyg Assoc 33: 178-191.
Remington RD, Schork MA. 1970. Statistics with applications to
the biological and health sciences. Englewood Cliffs, NJ:
Prentice-Hall, Inc., p. 234.
Robin ED. 1979. Chapter 14: Respiratory medicine Sections IV,
Obstructive lung diseases, and V, Restrictive lung diseases. In:
Rubenstein, Federman, eds. Medicine. New York: Scientific
American.
Robinson CF, Lemen RA, Wagoner JK. 1979. Mortality patterns,
1940-1975, among workers employed in an asbestos textile,
friction and packing products manufacturing facility. In. Lersen
RA, Dement JM, eds. Dusts and disease. Park Forest South, IL:
Pathotox Publishers, Inc.
Rossiter CE, Berry G. 1978. The interaction of asbestos
exposure and smoking on respiratory health. Bull Eur Physiopath
Respir 14: 197-204.
Rubino GF, Piolatto G, Newhouse ML, Scansetti G, Aresini GA,
Murray R. 1979. Mortality of chrysotile asbestos workers at
the Balangero Mine, Northern Italy. Br J Ind Med 36: 187-194.
Rubino GF, Scansetti G, Donna A, Palestro G. 1972.
Epidemiology of pleural mesothelioma in North-western Italy
(Piedmont). Br J Ind Med 29: 436-442.
Saijo N, Nishiwaki Y, Kawase I, Kobayashi T, Suzuki A, Niitani
H. 1978. Effect of ACNU on primary lung cancer, mesothelioma,
and metastatic pulmonary tumors. Can Treat Reps 62: 139-141.
Sawyer R. 1977. Asbestos exposure in a Yale building:
analysis and resolution. Environ Res 13(1): 146-168.
Sawyer R. 1979. Statement before the U.S. House Education and
Labor Committee, Subcommittee on Elementary, Secondary, and
Vocational Education. U.S. Government Printing Office 42-975-0.
Sawyer RN, Spooner CM. 1978. Sprayed asbestos-containing
materials in buildings: a guidance document. Research Triangle
Park, NC: U.S. Environmental Protection Agency. EPA-450/2-78-
014.
Sebastien P, Billon MA, Dufour G, Gaudichet A, Bonnaud G, Bignon
J. 1979a. Levels of asbestos air pollution in some
environmental situations. Ann NY Acad Sci 330: 401-415.
-112-
-------
Sebastian P, Gaudichet A, Dufour G, Bonnaud G, Bignon J, Goni
J. 1978. Metrological survey of atmospheric pollution inside
buildings insulated by asbestos projection. Paris, France:
Paris University Laboratory Department. [EPA translation].
Sebastien P, Janson X, Hirsch A, Gaudichet A, Bignon J.
1979b. Asbestos retention in human respiratory tissues:
comparative measurement in lung parenchyma and in parietal
pleura. Paper presented at the symposium on the Biological
Effects of Mineral Fibres held on 25-27 September 1979.
Sponsored by the International Agency for Research on Cancer,
Lyon, France.
Seidman H, Selikoff IJ, Hammond EC. 1979. Short-term asbestos
work exposure and long-term observation. Ann NY Acad Sci 330:
61-89.
Selikoff IJ. 1977a. Cancer risk of asbestos exposure. In:
Hiatt HH, Watson JD, Winsten JA, eds. Origins of human cancer.
Book C. Human risk assessment. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory.
Selikoff IJ, Environmental Sciences Laboratory, Mount Sinai
School of Medicine, City University of New York, New York.
1977b. Clinical survey of chrysotile asbestos miners and millers
in Baie Verte, Newfoundland - 1976. Report to the National
Institute of Environmental Health Sciences, Research Triangle
Park, NC.
Selikoff IJ, Environmental Sciences Laboratory, Mount Sinai
School of Medicine, City University of New York, New York, New
York. 1977c. Asbestos-associated disease in U.S. shipyards.
Memorandum to the National Institute of•Environmental Health
Sciences, Research Triangle Park, NC.
Selikoff IJ, Churg J, Hammond EC. 1964. Asbestos exposure and
neoplasia. J Am Med Assoc 188: 22-26,
Selikoff IJ, Churg J, Hammond EC. 1965. Relation between
exposure to asbestos and mesothelioma. N Engl J Med 272: 560-
565.
Selikoff IJ, Hammond EC. 1975. Multiple risk factors in
environmental cancer. In: Fraumeni JF Jr, ed. Persons at high
risk of cancer. New York: Academic Press.
Selikoff IJ, Hammond EC. 1979. Editorial: asbestos and
smoking. J Am Med Assoc 242: 458-459.
Selikoff IJ, Hammond EC, Churg J. 1968. Asbestos exposure,
smoking and neoplasia. J Am Med Assoc 204: 106-112.
Selikoff IJ, Hammond EC, Seidman H. 1979a. Mortality
experience of insulation workers in the United States and Canada,
1943-1976. Ann NY Acad Sci 330: 91-116.
-113-
-------
Selikoff IJ, Lee DHK. 1978. Asbestos and disease. New York:
Academic Press.
Selikoff IJ, Lilis R, Nicholson WJ. 1979b. Asbestos disease in
United States shipyards. Ann NY Acad Sci 330: 295-312.
Sluis-Cremer GK, duToit RSJ. 1973. Amosite and crocidolite
mining and milling as causes of asbestosis. In: Bogovski P,
Gilson JC, Timbrell V, Wagner JC, eds. Biological effects of
asbestos. Lyon, France: International Agency for Research on
Cancer.
Sluis-Cremer GK, duToit RSJ. 1979. Correlation of asbestos
fibre concentrations in the general atmosphere in residential
areas on amphibole mining areas with the prevalence of
radiological, changes in the non-occupationally exposed population
over 20 years of age resident in those areas. Paper presented at
the symposium on the Biological Effects of Mineral Fibres held on
25-27 September 1979. Sponsored by the International Agency for
Research on Cancer, Lyon, France.
Smith WE, Hubert DD, Badollet MS. 1969. Biological differences
in response to long and short asbestos fibers. Am Ind Hyg Assoc
J 236-244.
Smith WE, Hubert DD. 1974. The intrapleural route as a means
for estimating carcinogenicity. In: Karbe E, Park JF, eds.
Experimental lung cancer. Carcinogenesis and bioassays.
New York: Springer-Verlag.-
Stanton MF. 1973. Some etiological considerations of fiber
carcinogenesis. In: Bogovski P, Gibson JC, Tirabrell V, Wagner
JC, eds. Biological effects of asbestos. Lyon, France:
International Agency for Research on Cancer.
Stanton MF, Layard M. 1978. The carcinogenicity of fibrous
minerals. In: Proceedings of the workshop on asbestos:
definitions and measurement methods. Held in Gaithersburg, MD,
July 18-20, 1977. Washington, DC: National Bureau of
Standards. Special publication No. 506.
Stanton MF, Layard M, Tegeris A, Miller E, May M, Kent E.
1977. Carcinogenicity of fibrous glass: pleural response in the
rat in relation to fiber dimension. J Natl Cancer Inst 58: 587-
603.
Stell PM, McGill T. 1973. Asbestos and laryngeal carcinoma.
Lancet 2: 416-417.
Taryle DA, Lakshminarian S, Sahn S. 1976. Pleural
mesotheliomas — an analysis of 18 cases and review of the
literature. Medicine 55: 153-162.
-114-
-------
Van Ryzin J, Rai K. 1980. The use of quantal response data to
make predictions. In:. Witschi HR, ed. The scientific basis of
toxicity assessment. Elsevier:North-Holland Biomedical Press.
Versar, Incorporated, Springfield, VA 22151. February 15,
1980. Exposures to commercial asbestos. Section 3:
comparability of asbestos data. Preliminary draft report.
Washington, DC: Office of Pesticides and Toxic Substances, U.S.
Environmental Protection Agency. Contract No. 68-01-5791.
Vianna NJ, Polan AK. 1978. Non-occupational exposure to
asbestos and malignant mesothelioma in females. Lancet 1: 1061-
1063.
Wagner JC, Berry G, Skidmore JW. 1977. Studies of the
carcinogenic effect of fiber glass of different diameters
following intrapleural inoculation in experimental animals. In:
i^IOSH symposium on occupational exposure to fibrous glass, Univ
Md, 1974. Washington, DC: National Institute for Occupational
Safety and Health. DHEW (NIOSH) 76-151.
Wagner JC, Berry G, Skidmore JW, Timbrell V. 1974. The effects
of the inhalation of asbestos in rats. Br. J. Cancer 29:252-269.
Wagner JC, Berry G, Timbrell V. 1970. Mesothelioma in rats
following the intrapleural inoculation of asbestos. In: Shapiro
HA, ed. Pneumoconiosis: proceedings of the international
conference. • Capetown: Oxford University Press.
Wagner JC, Berry G, Timbrell V. 1973. Mesothelioma in rats
after inoculation with asbestos and other materials. Br J Cancer
28: 173-185
Wagner JC, Gilson JC, Berry G, Timbrell V. 1971. Epidemiology
of asbestos cancers. Br Med Bull 27: 71-76.
Wagner JC, Sleggs CA, Marchand P. 1960. Diffuse pleural
mesothelioma and asbestos exposure in the North Western Cape
Province. Br J Ind Med 17: 260-271.
Wagoner J, Johnson WM, Lemen R. 1973. Malignant and non-
malignant respiratory disease mortality patterns among asbestos
production workers. In: Congressional Record, Senate Proceedings
and Debates of the 93rd Congress, First Session, 119, part 6.
Washington, DC: U.S. Government Printing Office, S-4660-S-4662.
Wanner A. 1977. Clinical aspects of mucociliary transport. Am
Rev Respir Dis 116: 73-125.
Wassermann M, Steinits R, Katz L, Lemesch C. 1979.
Mesothelioma in children. Paper presented at the symposium' on
the Biological Effects of Mineral Fibres held on 25-27 September
1979. Sponsored by the International Agency for Research on
Cancer, Lyon, France.
-115-
-------
Weiss W. 1971. Cigarette smoking, asbestos and pulmonary
fibrosis. Am Rev Respir Dis 104: 223-227.
Weiss W, Theodos PA. 1978. Pleuropulmonary disease among
asbestos workers in relation to smoking and type of exposure.
J Occup Med 20: 341-345.
Whitwell F, Rawcliffe RM. 1971. Diffuse malignant pleural
mesothelioma and asbestos exposure. Thorax 26: 6-22.
Whitwell F, Scott J, Grimshaw M. 1977. Relationship between
occupations and asbestos-fibre content of the lungs in patients
with pleural mesothelioma, lung cancer, and other diseases.
Thorax 32: 377-386.
-116-
------- |