EPA 520/4-81-003
BACKGROUND REPORT
PROPOSED
FEDERAL
RADIATION PROTECTION GUIDANCE
FOR OCCUPATIONAL EXPOSURE
^tO Sfy
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RADIATION PROGRAMS
-------
PROPOSED
FEDERAL RADIATION PROTECTION GUIDANCE
FOR OCCUPATIONAL EXPOSURE
BACKGROUND REPORT
Criteria & Standards Division
Office of Radiation Programs
U.S. Environmental Protection Agency
January 16, 1981
-------
SUMMARY OF PROPOSED CHANGES
IN OCCUPATIONAL RADIATION PROTECTON GUIDANCE
Requirement
1960 Guides
1. Justification of exposure required
2. Optimization of exposure required
3. Limitation of exposure
a) Whole body
b) Partial body
c) Combined internal and
external exposure
4. Radiation Protection
Requirements
5. Regulatory limits lower
than the RPGs for
specific job categories
6. Intake guides
7. Exposure of minors
8. Exposure of the unborn
9. Exceeding the RPGs
3 rems/quarter;
5(N-18) cumulative
rems, (N = age)
individual critical
organ limits*
Proposed New Guides
required (also consider
alternatives to exposure)
required (include
collective dose)
5 rems/year
(100 rems/lifetime)
limit on sum of organ
risks*
independent limits combined limit
not specified
not addressed
Radioactivity
Concentration
Guides (RCGs)
1/10 RPGs
not addressed
permitted
in three ranges for
instruction, super-
vision, monitoring,
and recordkeeping
(including lifetime
dose)
recommended
Radioactivity Intake
Factors (RIFs)
1/10 RPGs
four alternative
recommendations
permitted (disclo-
sure now required)
*Some limits are raised and some lowered; some organs are deleted and some
added. See the specific guides for numerical values.
111
-------
TABLE OF CONTENTS
INTRODUCTION 1
I. THE PROPOSED RECOMMENDATIONS 9
II. OCCUPATIONAL EXPOSURES IN THE UNITED STATES 18
III. HEALTH RISKS DUE TO OCCUPATIONAL RADIATION EXPOSURE .... 30
Units 32
The present state of knowledge 35
Risk estimates used in this review 59
IV. GENERAL PRINCIPLES FOR THE PROTECTION OF WORKERS 77
Justification of activities leading to worker exposure . . 77
Optimization of the protection of workers 80
Limitation of risk to individual workers 81
V. MINIMUM RADIATION PROTECTION REQUIREMENTS 85
Education of workers 85
Radiation protection supervision 86
Monitoring and record keeping 88
Lifetime dose 89
VI. RADIATION PROTECTION GUIDES FOR MAXIMUM ALLOWED DOSES ... 92
Cancer risks from whole-body exposure 92
Health risks to the unborn 104
Health risks from partial-body exposure 109
VII. SPECIAL EXPOSURE SITUATIONS 113
REFERENCES 118
APPENDIX A - Non-linear dose responses in human populations
APPENDIX B - The Radioactivity Intake Factors
-------
TABLES
1. The U.S. radiation work force in 1970 19
2. The U.S. radiation work force in 1975 24
3. Irradiated populations in which cancer has been studied. . 38
4. Coefficients used to estimate risk of fatal cancer .... 60
5. Probability of cancer death by occupational category ... 66
6. Average loss of life expectancy by occupational category . 69
7. Fractional risks for non-uniform exposures 71
8. Risk coefficients for mutational effects 71
9. Risk coefficients for in utero risks 75
10. Annual risk of accidental death in U.S. industries .... 94
11. Non-Fatal injuries and illness in U.S. industries 103
81. Maximum concentration of selected radionuclides in air . . B-3
VI
-------
FIGURES
1. Principal regulators of occupational radiation exposure ... 2
2. Distribution of workers by dose range in 1975 21
3. Distribution of workers and collective dose in 1975 23
4. Effect of radiation on a population with sensitive subgroups. 45
5. Dose response for microcephaly due to in utero exposure ... 54
6. Age-dependent future risk of cancer death due to an annual
dose of one rem for a working lifetime 61
7. Risk of cancer death by attained age due to an annual dose
of one rem for a working lifetime 63
8. Lifetime risk of death from radiation-induced cancer
versus annual dose 65
9. Loss of life expectancy due to radiation-induced cancer
versus annual dose 68
10. Average risk of mutational effects versus annual dose .... 73
11. Lifetime risk of death due to radiation-induced cancer
compared to occupational risks of accidental death 96
12. Loss of life expectancy due to radiation-induced cancer
compared to occupational risk of accidental death 98
13. Risk of death from radiation-induced cancer due to a single
dose of 12 rems versus age at exposure 100
Al. Dose response for leukemia in two samples of Nagasaki
survivors A-2
VI1
-------
FEDERAL RADIATION PROTECTION GUIDANCE
FOR OCCUPATIONAL EXPOSURES
INTRODUCTION
In 1975, the latest year for which comprehensive statistics are
available, there were almost one and a quarter million people potentially
exposed to ionizing radiation in their jobs or as students (En80). We
estimate there are now about one and a half million. Workers exposed to
radiation are engaged in a wide variety of medical, industrial, defense,
research, and educational activities involving many kinds of radiation
sources. These include x-ray emitting devices, a large number of
naturally-occurring and man-made radioactive materials, nuclear reactors,
and particle accelerators. Workers exposed to radiation in mining
operations are not included in the above estimates; except for underground
uranium miners, there is little information on their exposure.
No single agency regulates the exposure of workers in the United
States. This responsibility is carried out by five Federal regulatory
agencies with jurisdiction over exposure of workers or sources of
radiation exposure in private industry, several Federal agencies who
regulate exposure of their own (or their contractors') employees, and
various agencies of the fifty States (see Figure 1). Some of these State
agencies regulate exposure of workers under agreements with one or more of
the Federal regulatory agencies, and some regulate independently.
-------
AUTHORITIES FOR RADIATION PROTECTION OF WORKERS
RECOMMENDATIONS
OCCUPATIONAL RADIATION PROTECTION GUIDANCE
PRESIDENTIAL
GUIDANCE TO
REGULATORS
STATES
171
i
i
NRC
12)
(
| APPROVED STATES
AGREEMENT
i
NRC AGREE-
MENT. OSHA
APPROVED
AND STATE
PROTECTED
WORKERS
i
OSHA
13)
i
GOVERN-
MENT AND
NONGOV
ERNMENT
LICENSEE
WORKERS
*
MSHA
(4)
NRC
I
ALL WORK-
ERS NOT
OTHERWISE
PROTECTED
NON-AEAI2)
LICENSEES
1.
_]
EXPOSURES
. i
DOD
(2)
1
MINE AND
MILL WORK
ERS
11
J
DOE
12)
\
MILITARY.
AND DOD
CIVILIAN
AND CON-
TRACTOR
WORKERS
. |
OTHER
DOT
(5)
FEDERAL
AGENCIES
(3)
1
DOE AND
DOE CON-
TRACTOR
WORKERS
1
AGENCY
AND
AGENCY
CONTRAC-
TOR
WORKERS
REGULATORS
OF WORKERS
EXPOSURE
FDA
(6)
TRANSPORT
WORKERS
REGULATORS
OF SOURCES
ONLY
IMPLEMENTORS
OF GUIDANCE
AND REGULATIONS
I
WORKERS
USING ELEC-
TRONIC
PRODUCT
RADIATION
SOURCES
PROTECTED
WORKERS
Figure 1. Occupational radiation protection guidance is binding on all major regulatory agencies except
NRC and the States, in which case it is advisory. Heavy lines refer to Federal Radiation
Protection Guidance; light lines indicate regulations. The authorities cited in parentheses
are (1) Executive Order 10831; (2) Atomic Energy Act of 1954, as amended; (3) Occupational
Health and Safety Act of 1970; (4) Federal Mine Safety and Health Act of 1977; (5) Department
of Transportation Act of 1966; (6) Radiation Control for Health and Safety Act of 1968; (7)
State enabling legislation and State laws.
-------
During the three decades prior to 1960 two organizations of
professionals in radiation protection and related fields of research, the
International Commission on Radiological Protection (ICRP) of the
International Congress on Radiology and the National Council on Radiation
Protection and Measurements (NCRP) and its predecessor, provided
recommendations which served as the principal basis for the rules
established by all of these regulators. However both of these are, in
effect, private groups; they choose their own members and agree on
recommendations in private. In 1959 the President created a public body
for the United States, the Federal Radiation Council (FRC), to provide
recommendations to him on radiation matters affecting health. The
recommendations issued by the FRC were promulgated by successive
presidents as guidance to Federal agencies, and provided a uniform basis
for Federal and State regulation of many forms of public exposure to
radiation.
The Federal radiation protection guidance now in effect for most
occupational exposure (Fe60) was developed by the FRC and was promulgated
by President Eisenhower on May 18, 1960. It was implemented through
regulations of the former Atomic Energy Commission, the former Energy
Research and Development Administration, the Nuclear Regulatory
Commission, the Occupational Safety and Health Administration, the
Departments of Defense and Energy, and the States, as well as by other
Federal regulatory agencies with specialized responsibilities, such as
the former Mining Enforcement and Safety Administration, the Mine Safety
and Health Administration, and the Department of Transportation.
Although additional Federal guidance was issued in 1971 for the special
case of exposure of underground uranium miners to radon decay products
-------
(En71), the basic guidance which governs the exposure of the vast
majority of workers has not been reviewed or modified since it was
established in 1960.
In 1970 the President abolished the FRC and transferred its
functions to the Administrator of the Environmental Protection Agency
(EPA) (Re70). EPA has developed recommendations for new radiation
protection guidance for workers pursuant to this responsibility to advise
the President on radiation matters affecting health. This report
contains the support for these new recommendations, which would replace
the guidance now used by Federal agencies to regulate all occupational
exposure to ionizing radiation except the exposure of underground uranium
miners to radon decay products.
We have based these recommendations on the assumption that risks to
health should be considered in relation to the need for exposure. This
approach is similar to that used by the FRC in 1960. As the FRC said
(Fe60): "Fundamentally, setting basic radiation protection standards
involves passing judgment on the extent of the possible health hazard
society is willing to accept in order to realize the known benefits of
radiation." In this review we have also compared risks from occupational
exposure to ionizing radiation with risks of accidental death and non-
lethal occupational diseases in industries and occupations in which
workers are not occupationally exposed to radiation. We have not,
however, attempted to either assess or limit total risk to workers from
all causes.
In forming these judgments we have considered current knowledge of
how radiation affects health, the number of people now exposed, and the
size of the radiation doses they receive. We have also considered recent
reviews and recommendations of the National Academy of Sciences -
-------
National Research Council (NAS-NRC) (NA72,77,80), the United Nations
Scientific Committee on the Effects of Atomic Radiation (Un77), the
NCRP (NC71-77), and the ICRP (IP73-80). Although our estimates of risk
are based on more data and better understanding than existed in I960,
they are still uncertain. Nevertheless, we believe they provide an
adequate basis for this new radiation protection guidance. In spite of
the uncertainties we have made numerical estimates of the risks from
doses permitted by these recommendations because we believe that this
information is essential to judgments by the public of the appropriate-
ness and acceptability of these recommendations.
The primary changes from the 1960 guidance are structural. We have
also modified the numerical values of maximum allowed radiation dose
levels. The recommendations place increased emphasis on eliminating
unjustified exposure and on keeping justified exposure as low as is
reasonably achievable, both long-standing tenets of radiation protec-
tion. A principal addition is the introduction of a graded set of
minimum radiation protection requirements in three exposure ranges. We
have tried to express these recommendations in terms that dispel any
notion that the levels specified are dividing lines between "safe" and
"unsafe," and that exposure within any of the recommended ranges may be
viewed as "acceptable" without qualification.
Among the major issues we addressed in developing these
recommendations are the following (sections of the report which contain
principal discussions of each are indicated in parentheses):
1. Are the doses currently received by workers (II) and the maximum
dose permitted under existing guidance adequately low? (VI) In this
-------
regard, a) how adequate is the basis used for estimating risks to health
from radiation exposure (III), and b) what are the appropriate bases for
judging collective and maximum individual radiation doses in the work
force and the tradeoffs between these two indices of the risk from
occupational exposure? (IV)
2. Should the same guides apply to all categories of workers (e.g.,
dental workers, nuclear medicine technicians, nuclear maintenance
personnel, industrial radiographers)? (IV) Should specific guides be
developed for pregnant women, female workers who could bear children,
and/or men? (VI)
3. On what time basis should the guides be expressed? Quarterly?
Annual? (VII) Should the lifetime occupational dose be limited? (VI)
Should the age of the worker be a factor? (VI)
4. Should the guidance reflect or cover medical, accidental, and/or
emergency exposures? (VII)
5. Is existing guidance for situations that involve exposure of less
than the whole body adequate? In this respect, a) what organs and parts
of the body should have designated limits, and b) on what basis should
guidance be expressed for exposure of more than one organ or portion of
the body? (VI)
6. How should the radiation protection principles requiring
a) justification of any exposure, and b) reduction of the dose from
justified exposures to the lowest practicable or as low as is reasonably
achievable level be applied to exposure of workers? Should the concept
of lowest feasible level be applied to exposure of workers? (IV and V)
Collective dose is numerically identical to the sum of all the
doses received by the members of a group.
-------
7. What, if any, relationship should be maintained between
permissible levels of risk to health from radiation exposure and other
regulated hazards of disease or accidents? (IV and VI)
8. Should the guidance include numerical values for the factors
(called "quality" and "modifying" factors) used to convert dose (measured
in rads) to dose equivalent (measured in rems)? If so, should this be
developed now or issued later as supplementary guidance? (Ill)
9. What guidance should apply to workers who do not use radiation
sources, but who are exposed to radiation due to the activities of other
workers? (VII)
10. Are there situations that may require doses higher than normally
permitted? Should we provide special guidance for them? (VII)
The proposed recommendations for radiation protection of U.S. workers
are contained in the first chapter. The report continues with a summary
of the size, composition, and exposure of the work force exposed to
radiation (Chapter II), followed by a summary of current knowledge of the
harm from radiation exposure and estimates of the risks at the exposure
levels experienced under and the maximum levels permitted by current
Federal radiation protection guidance (Chapter III). These two chapters
describe the basic characteristics of the radiation work force and of
risks from radiation that lead to the proposed radiation protection
recommendations. Each of the recommendations is discussed in turn in the
balance of the report. In Chapter IV we discuss general radiation
protection principles. Chapter V describes our proposal for graded
Minimum Radiation Protection Requirements in three dose ranges to help
assure that workers get as small a dose as is reasonably achievable.
-------
In Chapter VI we justify the numerical values recommended as Radiation
Protection Guide's (RPGs) for the whole-body and for individual organs and
extremities of the body, and discuss alternative proposals for protection
of the unborn. In this chapter we also address some related matters, such
as additivity of risk when several organs are irradiated and the factors
used to relate intake of radioactive materials to the RPGs. Finally, in
Chapter VII we briefly cover several special exposure situations, such as
exposure of minors, emergency exposures, and overexposures; diagnostic
x rays; and some technical matters regarding implementation.
8
-------
I. THE PROPOSED RECOMMENDATIONS
We propose nine recommendations as guidance to Federal agencies in
the formulation of Federal radiation protection standards for workers,
and in their establishment of programs of cooperation with States. These
recommendations are discussed in detail in Chapters IV-VII. In all cases
but one we have made single recommendations for public comment. The
exception, Recommendation 8, addresses protection of the unborn during
gestation. Because this recommendation involves issues that go beyond
simple radiation protection of workers, including equality of employment
rights and the rights of the unborn, we have proposed four alternatives
for public consideration. The recommendations follow:
1. All occupational exposure should be justified by the net
benefit of the activity causing the exposure. The justification
should include comparable consideration of alternatives not requiring
radiation exposure.
2. For any justified activity a sustained effort should be
made to assure that the collective dose is as low as is reasonably
achievable.
3. The radiation dose to individuals should conform to the
numerical Radiation Protection Guides (RPGs) specified below.
Individual doses should be maintained as far below these RPGs as is
reasonably achievable and consistent with Recommendation 2.
-------
3. (Continued).
Radiation Protection Guides;
a. xne sum of the annual dose equivalent from external
exposure and the annual committed dose equivalent from
internal exposure should not exceed the following values:
Whole body 5 rem
Gonads 5 rem
Lens of eye 5 rem
Hands 50 rem
Any other organ 30 rem
b. Non-uniform exposure of the body should also satisfy the
condition on the weighted sum of annual dose equivalents
and committed dose equivalents, E^, that
Hj, = ]£j W£H£ < 5 rem,
i
where W£ is a weighting factor, H^ is the annual dose
equivalent and committed dose equivalent to organ i, and
the sum excludes the gonads, lens of eye, and hands.
Recommended values of
W£ are:
Breast 0.20
Lung 0.16
Red Bone Marrow 0.16
Thyroid 0.04
Bone Surfaces 0.03
Skin 0.01
Other Organs 0.08
* "Dose equivalent" means the quantity expressed by the unit "rem,"
as defined by the International Commission on Radiation Units (IU73),
** "Annual committed dose equivalent" applies only to dose equivalents
from radionuclides inside the body. It means the sum of all dose
equivalents that may accumulate over an individual's remaining
lifetime (usually taken as 50 years) from radioactivity that is
taken into the body in a given year.
*** Applies only to each of the five other organs with highest doses.
10
-------
3. (Continued).
c. When both uniform whole-body exposure and nonuniform
exposure of the body occur, in addition to the requirements
of 3a, the annual uniform whole-body dose equivalent added
to the sum of weighted annual dose equivalents from
additional nonuniform exposure, H.., should not exceed 5 rem.
4. The following Minimum Radiation Protection Requirements
should be established by appropriate authorities and carried out in
the workplace, on the basis of the range of doses anticipated in
individual work situations. The numerical values specifying the dose
ranges may be adjusted to fit the needs of specific situations by
implementing agencies.
Minimum Radiation Protection Requirements:
Range A
a. Determine that exposures result only from justified
activities and are as low as is reasonably achievable.
These determinations may often be made on a generic basis,
that is, by considering groups of similar work situations
and protective measures.
b. Monitor or otherwise determine individual or area exposure
rates to the extent necessary to give reasonable assurance
* Suggested numerical ranges are: Range A, less than 0.1 RPG; Range B,
0.1 - 0.3 RPGj Range C, 0.3 - 1.0 RPG.
11
-------
4. (Continued).
that doses are within the range and are as low as is
reasonably achievable.
Instruct workers on basic hazards of radiation and
radiation protection principles, and on the levels of risk
from radiation and appropriate radiation protection
practices for their specific work situations. The degree
of instruction appropriate will depend on the potential
exposure involved.
The above requirements, plus:
d. Provide professional radiation protection supervision in
the work place sufficient to assure that both individual
and collective exposures are justified and are as low as is
reasonably achievable.
e. Provide individual monitoring and recordkeeping.
The above requirements, plus:
f. Justify the need for work situations which are expected to
make a significant contribution to exposure in Range C and
provide professional radiation protection supervision
before and while such jobs are undertaken to assure that
collective and individual exposures are as low as is
reasonably achievable.
12
-------
4. (Continued).
g. Carry out sufficient additional monitoring of workers to
achieve Recommendation 4f.
h. Once a worker has been exposed in Range C, maintain a
lifetime dose record, including at least all subsequent
annual doses (as specified in Recommendation 3c) in
Ranges B and C.
i. Maintain lifetime doses as low as is reasonably achievable.
The accumulation of doses (as recorded under Recommendation
4h) by individual workers should be managed so that their
lifetime accumulated dose is less than 100 rem.
5. a. "Radioactivity Intake Factors" (RIFs) should be used to
regulate occupational radiation hazards from breathing, swallowing,
or immersion in media containing radionuclides. The RIF for a
radionuclide is defined as the maximum annual intake (in curies) for
which the committed dose equivalent to a reference person satisfies
the Radiation Protection Guides in Recommendation 3. RIFs may be
derived for different chemical or physical forms, and for intake by
breathing, swallowing, or for external exposure from air containing a
radioactive gas. Exposure regulated through use of the RIFs should
meet the same Minimum Radiation Protection Requirements as equivalent
exposure under the Radiation Protection Guides.
b. When a RIF for a specific radionuclide in a specific
chemical or physical form determined on the basis of part a) is
larger than that currently in use, a value no greater than that in
13
-------
current use should be adopted in regulations governing work situations
identical or similar to those currently in existence.
6. Federal agencies should establish limits and administrative
levels that are below the RPGs and the RTFs, when this is appro-
priate. Such limits or levels may apply to specific categories of
workers or work situations.
7. In addition to any other Federal restrictions, the
occupational exposure of individuals younger than eighteen should be
limited to one tenth of the Radiation Protection Guides for adult
workers.
*
8. Exposure of the unborn should be restricted more than that
of workers. This should include special consideration of ALARA
practices for women. Women able to bear children should be fully
informed of current knowledge of risks to the unborn from radiation.
In addition, employers should assure that protection of the unborn is
achieved without loss of job security or economic penalty to women
workers. Due to the complexity of the issues involved, we propose
four alternative recommendations on numerical limitation of dose to
the unborn for public comment. We would be glad to receive other
recommendations for dealing with exposure of the unborn.
"Unborn" here means the fertilized oocyte, the embryo, and the fetus.
14
-------
a. Women are encouraged to voluntarily keep total dose to any
unborn less than 0.5 rem during any known or suspected
pregnancy; or
b. Women able to bear children are encouraged to voluntarily
avoid job situations involving whole-body dose rates greater
than 0.2 rem per month, and to keep total dose to the unborn
less than 0.5 rem during any known pregnancy; or
c. Women able to bear children should be limited to job
situations involving whole-body dose rates less than 0.2 rem per
month. Total dose to the unborn during any known period of
pregnancy should be limited to 0.5 rem; or
d. The whole-body dose to both male and female workers should
not exceed 0.5 rem during any six month period.
9. In exceptional circumstances the RPGs may be exceeded, for
cause, but only if the Federal agency having jurisdiction carefully
considers the specific reasons for doing so, and publicly discloses
them unless this would compromise national security.
The following notes clarify application of the above recommendations:
1. Occupational exposure of workers does not include that due
to a) normal background radiation and b) exposure as a patient of
practitioners of the healing arts.
15
-------
2. When uniform external whole-body exposure occurs in addition
to exposure from radioactive materials taken into the body, the
requirement of Recommendation 3c may be satisfied by the condition
that
RpCb + y RIF j - lf
where Hext is the external whole-body dose equivalent, RPGwt, is
5 rem, I- is the intake of radionuclide j, and RIF- is defined as in
Recommendation 5.
3. The values currently specified by the ICRP for quality
factors and dosimetric conventions for measurement of the various
types of radiatipn may be used for determining conformance with the
RPGs. The model for a reference person and the metabolic models
currently specified by the ICRP may be used to calculate the RIFs.
We will recommend other factors, conventions, and models when and if
they are more appropriate.
4. Numerical guides for emergency exposures are not provided
by this guidance. Agencies should follow the general principles
established by Recommendations 1, 2, 7, 8, and 9 in dealing with
such situations.
5. Procedures for handling overexposures are not addressed by
this guidance. The equitable handling of such cases is the
responsibility of the employer and the Federal agency having
regulatory jurisdiction.
16
-------
6. Limits for periods other than one year may be derived by
Federal agencies from the annual RPGs and RIFs when necessary for
administrative purposes. Such limits should be consistent with
Recommendation 2 and the three ranges in Recommendation 4.
7. The existing guide for limiting exposure of underground
uranium miners to radon decay products is not changed by these
recommendations.
These proposed recommendations would provide general guidance for the
radiation protection of workers. They would replace that part of existing
guidance (see 25 F.R. 4402 of May 18, 1960) which applies to workers.
Individual Federal agencies, with their knowledge of specific worker
exposure situations, would use this guidance as the basis upon which to
develop detailed standards and regulations to meet their particular
statutory obligations. We propose to follow the activities of the Federal
agencies as they implement the final Guidance, to issue any necessary
clarifications and interpretations, and to promote the coordination
necessary for an effective Federal program of worker protection.
17
-------
II. OCCUPATIONAL EXPOSURES IN THE UNITED STATES
The use of radiation in the work place has increased steadily since
current Federal occupational radiation protection guidance was established
in 1960. In a 1972 study we estimated that in 1960 about 460,000 people
were exposed to radiation in their jobs (C172). This was 0.6 percent of
all workers and about one quarter of one percent of the 1960 United States
population. The mean annual occupational dose to that work force was
roughly estimated as 300 millirem, based on data for only 30,000 workers
from two of the larger facilities operated by the Atomic Energy
Commission, the Hanford and Oak Ridge National Laboratories. In a study
begun in 1975 (En80) we improved this estimate by using additional data;
the result was a mean annual dose of 170 millirem based on records for
130,000 workers in Federal and Federal contractor facilities in 1960.
The 1972 study also contains an analysis of the 1970 work force. The
results are shown in Table 1. The total number of radiation workers was
estimated to be about 770,000, with a mean annual occupational dose of 210
millirem. This was 0.9 percent of all workers and about one-third of a
percent of the 1970 United States population. The number of radiation
workers increased by two-thirds during this decade. However, the data
bases are too different and too uncertain to tell whether there was a
significant change in mean dose. The data indicate that the largest
collective dose was received by medical workers and that those who handled
radium received the highest mean dose of any class of workers studied.
18
-------
Table 1. Occupational Exposure Summary for 1970 (C172)*
Category
Number of
Radiation
Workers
Mean Whole-
Body Dose
(millirem)
Collective
Dose
(person-rams)
Atomic Energy Commission
Contractors 102,918
Reporting Licensees
AEC 62,090
Agreement State 24,519
Non-reporting Licensees
AEC 93,000
Agreement State 3,000
Department of Defense
Army 7,445
Air Force 17,591
Navy 55,051
Other Federal
PHS 508
Miscellaneous 2,000
Medical**
Radium 37,925
Non-Federal
Medical x ray 194,451
Dental x ray 171,226
198
215
274
54
274
100
88
198
129
129
540
320
125
20,361
13,365
6,715
5,022
822
744
1,555
10,879
65
258
20,480
62,253
21,403
All Workers
772,000
210**
164,000
,'fC'fC
**
Numbers of some workers and the mean and collective dose to the entire
work force have been rounded to the nearest 1000 workers, 10 millirem, and
1000 person-reins, respectively. Sources of values quoted to more signi-
ficant figures are given in the original report.
Values of doses to medical workers were based on limited data obtained
from a few States. Based on data for comparable situations in government
facilities, as well as more complete data for later years, doses to medical
workers are probably overestimated.
19
-------
The study begun in 1975 designed and tested a procedure to monitor
trends in occupational exposure and provides a baseline for assessing the
effect of any future changes in Federal occupational radiation protection
guidance. This study was recently completed using 1975 records for over
450,000 people obtained from both governmental and commercial sources
(En80). It includes all types of workers exposed to radiation except
miners.
Figure 2 shows the distribution of occupational doses projected from
these data. We estimate that two-thirds of those exposed in their jobs
received "no measurable dose" during any monitoring period. (This means
that the dose received by these workers was not distinguishable from
background radiation for any single monitoring period during the year, and
therefore that their annual occupational dose was much less than 100
millirem, the nominal value for background radiation exposure in the
United States.) About 95% of all workers are estimated to have received
doses of less than 500 millirem. Only 0.1% of the work force is estimated
to have received doses between 5 and 12 rem. Twelve rem is the maximum
permitted under current guides.
Based on this study, we estimate that 1,106,900 workers were
potentially exposed to ionizing radiation in their workplaces in 1975.
(There were also an estimated 120,000 students and airline personnel who
are not usually considered part of the radiation work force.) This was
1.2 percent of all workers and a little over one-half of one percent of
the 1975 United States population. It is approximately two and one-half
times the number in 1960 and one and one-half times that in 1970. The
mean annual occupational dose to these workers was 120 millirem. This
mean is computed assuming that those reported as receiving "no measurable
20
-------
800,000 I-
700,000
600,000
£ 500.000
400,000
300.000
200.000
100,000
66.7%
18.8%
5.8%
3.3%
2.4%
NM NM- 0.10- 0.26- 0.6-
0.10 0.25 0.50 1.0
1.0- 1.6-
1.5 2.0
ZO- 3.0- 4.0- 5.0-
3.0 4.0 5.0 1Z
12+
Dose Rang* (REM)
Figure 2. The distribution by dose range of U.S. radiation workers
in 1975 (En80). "NM" means that the dose was not measurable.
21
-------
dose" received zero dose. If only those who received a measurable dose
during any reporting period of the year are counted (approximately 369,100
individuals), the mean becomes 350 millirem. Although it can be inferred
from these data that the average dose has probably declined during the
years 1960-1975 and that the collective dose to the entire work force may
not have increased, definite conclusions cannot be drawn because we do not
know how comparable the data from the earlier studies are.
We also estimated the number of workers, as well as mean and
collective doses, in different parts of the work force. Figure 3 shows
the distribution of workers among major occupational groups in 1975.
Medical workers make up about one half of the work force, industrial
workers 18%, and government (including defense) workers 17%. Nuclear fuel
cycle workers are 7% of the work force. The Figure also illustrates t..e
distribution of collective dose among these major occupational groups.
Despite the significantly higher mean doses noted below for some types of
nuclear fuel cycle and industrial workers, medical workers account for 40
percent of the national -jllective dose, more than all nuclear fuel cycle
and industrial we ers combined.
Table 2 summarizes the number of workers, the mean dose, and the
collective dose in individual job categories. Mean doses are shown for
all workers and for just those who received a measurable dose. Since we
calculated mean doses to all workers and collective doses using the
assumption that the dose to individuals receiving "no measurable dose" was
zero, these calculated doses may be underestimated. If one assumes that a
log-normal distribution, which fits measured doses above 100 millirem
22
-------
Figure 3. Distribution of workers (a) and collective dose (b) in the 1975
occupationally exposed work force (En80).
23
-------
Table 2. National Occupational Exposure Summary For 1975 a (En 80)
Occupational
Subgroup
MEDICINE
Hospital /Clinic
Private Practice
Dental
Podiatry
Chiropractic
Veterinary
Number of
Radiation Workers
Totalb Exposed0
100,000
137,800
265,700
10,100
14,600
18,100
55,100
53,300
41,400
2,100
3,700
6,200
Mean Whole-Body
Dose (millirem)
Totalb Exposed0
220
160
20
10
30
80
400
410
140
30
110
230
Collective
Dose
(person-rems)
22,000
21,700
5,800
100
400
1,400
Entire Subgroup
INDUSTRY
546,300
161,800
Industrial Radiography
Licensees 19,800
Other Industrial Users
9,700
Entire Subgroup 200,800 49,200
NUCLEAR FUEL CYCLE
Power Reactors 54,763 28,034
Fuel Fabrication
Entire Subgroup
74,200
39,400
90
290
130
390
340
320
580
520
760
630
51,400
5,700
Licensees 114,100
Registrants
Source Manuf. & Distr.
Licensees
Registrants
55,900
7,000
4,000
18,800
16,000
3,900
800
100
110
350
40
610
370
630
200
11,400
5,900
2,500
200
25,600
21,400
and Reprocessing
Uranium Enrichment
Nuclear Waste Disposal
Uranium Mills
11,405
7,471
300
300
5,495
5,664
100
100
270
50
310
20
560
70
920
50
3,100
400
100
-
24,900
24
-------
Table 2. (Continued)
Number of
Occupational Radiation Workers
Subgroup Total Exposed0
Mean Whole-Body
Dose (millirem)
Totalb Exposed0
Collective
Dose
(person-rems)
GOVERNMENT
Dept.
Dept.
Other
of Energy 80,954
of Defense 92,500
Federal Govt. 13,400
39,451
55,800
4,400
150 300
110 180
90 280
11,800
10,100
1,300
Entire Subgroup
MISCELLANEOUS
186,800
99,700
Entire Subgroup
ALL WORKERS
98,800
1,106,900
19,000
369,100
120
40
120
230
200
350
23,100
Education (Faculty):
2-year Institutions
4-year Institutions
Transportation
7,000
14,800
77,000
2,300
4,900
11,800
60e
80e
30
170
230
200
400
1,100
2,300
3,800
128,800
ADDITIONAL GROUPS
Transportation
(Flight attendants; 30,000 10,000
radionuclides)
Education (Students):
2-year Institutions 35,000 11,700
4-year Institutions 54.800 18,300
60'
80e
10
170
230
100
2,000
4,200
All Additional Groups 119,800
40,000
50
150
6,100
b
c
d
e
f
Extrapolated numbers of workers are rounded to the nearest 100, mean doses to
the nearest 10 millirem, and collective doses to the nearest 100 person-rems.
All monitored and unmonitored workers with potential occupational exposure.
Workers who received a measurable dose in any monitoring period during the year.
"Licensee" means NRC and NRC agreement state licensees for use of radionuclides.
Doses from electronic (e.g., x-ray) sources are also included. "Registrant"
means state registrants, who have electronic sources only.
These estimated doses are based on small samples that may not be representative.
Persons who are only incidentally exposed or not normally considered radiation
workers; the estimates listed are very uncertain.
25
-------
well, holds also for lower doses that are not measurable, then assuming
"no measurable dose" was zero dose would under-estimate the collective
dose for all workers by less than 3 percent. However, dosimeter readings
are corrected by subtracting an average value for background radiation.
When negative values result these are reported as zero. This creates an
upward bias in reported values that could more than compensate for
assuming that "no measurable dose" is zero. Since the number of monitored
but not exposed workers in any job category is also a highly variable
quantity, depending upon the degree of conservatism in administering
radiation protection programs as well as other difficult to assess
factors, we consider that the mean dose of only those workers with
measurable doses is a more reliable value to use for comparing risks in
various parts of the work force.
A recent study of personnel dosimetry services for the U.S. Nuclear
Regulatory Commission indicates that a significant number of individual
dosimetry records are not accurate (Nu80). In two rounds of tests, 22%
and 14% of dosimeters were in error by more than 50%. However, despite
the poor performance of individual dosimeters, the same study showed that
the mean value for a large number of dosimeters gives close to the correct
average and collective doses. The study showed, for example, that in
samples of more than 1000 dosimeters the mean value of measured dose was
28% high for low-energy x rays (15-30 kev), 17% high for medium-energy
x rays (30-300 kev), 3% high for cobalt-60 gamma rays (1.2-1.3 Mev), and
21% low for califomium-252 neutrons (thermal to several Mev).
We do not know to what extent the choice and calibration of personnel
dosimeters is tailored to the various kinds of radiation to which workers
26
-------
are exposed. In addition, different methods are used to adjust dosimeters
for background radiation. These factors, along with the results of the
above study, lead us to conclude that mean and collective values of dose
to most categories of workers, as well as to the entire work force, are
probably known to within no better than 30%.
Counting only those who received measurable doses, nuclear fuel cycle
workers had the highest mean annual dose. In 1975 these nuclear workers
averaged 630 millirem and included three of the six job categories with
the highest mean dose — 920 millirem for waste disposal workers, 760
millirem for power reactor workers, and 560 millirem for fuel fabrica-
tion and reprocessing workers. Industrial workers with measurable doses
had the second highest mean dose — 520 millirem. This group, which
contains the job categories with the third, fourth, and fifth highest mean
dose, are all NRC and Agreement State licensees principally exposed in
work involving the use of radionuclides — industrial radiographers at 580
millirem, source manufacturing and distribution workers at 630 millirem,
and other industrial workers at 610 millirem. Mean measurable doses to
workers in jobs in the remaining parts of the work force, which include
82% of all exposed individuals, were in most cases significantly below 500
millirem.
One can divide workers receiving measurable doses into two major
groups: 1) a group of about 66,000 in the above six highest dose job
categories who received mean doses in the neighborhood of 600-900
millirem, and 2) a much larger group of about 303,000, primarily in
medicine, government, and education, most of whom received mean doses of
100-400 millirem. Almost two-thirds of the collective dose in the entire
27
-------
work force is received by workers in this latter group. That is, the
majority of occupational exposure accrues to the 82% of exposed workers
who are in the lower dose occupations.
The study also provides some information on the distribution of dose
by age and sex. The mean dose for men is higher than that for women at
any age, and is more than double averaged over all ages. Women average
about 70 millirem per year during their normal childbearing years (age
18-40). Men average about 170 millirem per year prior to age 40. Women
comprise 66% of all radiation workers of ages 18-24, but accumulate only
42% of the collective dose to workers in that age group. From age 30 on,
men comprise about 70% of the work force, and accumulate 85% of the
collective dose. Female workers are found mostly in the parts of the work
force with lower mean doses, i.e., in medicine, government, and
education. This could explain why women contribute a lower proportion of
collective dose than their numbers might imply. However, even within
these occupations, mean doses to women are generally only 25-50% of those
to men in the same occupations.
In summary, during the period 1960 - 1975, we estimate that the
number of workers potentially exposed to radiation grew from 460,000 to
1,106,900, an average growth rate of about 6% per year during a period
when the average growth rate of the general population was only 1.2% per
year. The mean annual dose in 1960, based on exposure records for AEG
workers, has been roughly estimated as a few hundred millirem. In 1975
the estimated mean dose to 1,106,900 United States workers was 120
millirem. For the 369,100 workers receiving measurable doses it was 350
millirem. The largest group of workers and the largest contributors to
collective dose are medical workers, who accumulated 40 percent of the
28
-------
total dose for all workers. Mean doses for workers receiving measurable
doses in a few specific occupations, such as nuclear power reactor workers
and industrial radiographers, were twice as high as those to most other
workers receiving measurable doses. The mean dose to males was signifi-
cantly higher than that to women in all job categories. Finally, the
distribution of doses among workers is heavily weighted toward low doses:
two-thirds received no measurable dose, 95% received less than 0.5 rem,
and only 0.15% received 5 rem or more.
29
-------
III. HEALTH RISKS DUE TO OCCUPATIONAL RADIATION EXPOSURE
This chapter outlines the assumptions and methods we use to estimate
the risk of harm from occupational levels of radiation exposure. Section
A discusses the units used to quantify radiation dose. Section B defines
each type of harm and briefly describes the information on which our risk
estimates are based. The last section describes the parameters and risk
projection models we use and illustrates how these choices affect the
risks calculated for different levels of occupational exposure.
The following discussion represents our current understanding of the
risks from exposure to radiation. Our understanding has grown and changed
over the years. Undoubtedly it will continue to grow and change. Some of
what we now believe may, in the light of future knowledge, prove to be
wrong and much of it is obviously incomplete. Nevertheless, the degree
and mechanisms of harm from ionizing radiation are better understood than
those of almost any other carcinogen or mutagen.
Biological harm caused by ionizing radiation may be divided into two
general classes: somatic effects, which occur in exposed individuals; and
hereditary effects, which appear in their descendants. Some somatic and
all hereditary effects are generally believed to be "stochastic effects"
(IP77). We use "stochastic effects" here to mean those for which the
frequency of occurrence increases with dose, but the degree of impairment
does not. This is in contrast to some somatic effects for which the kind
or the severity of the impairment changes with dose, so that, for small
enough doses, the effects are negligible.
30
-------
Cancer is the principal stochastic risk to the exposed worker.
Radiation-induced cancers include leukemia and most commonly-occurring
solid cancers. There is no known way to distinguish them from cancers due
to other causes. Similarly, hereditary effects due to radiation are
assumed to exhibit the same range of impairment as those due to other
causes. Non-stochastic effects include cataract of the lens of the eye,
non-malignant damage to skin, cell depletion in the bone marrow leading to
hematological deficiencies, and gonadal cell damage causing impaired
fertility.
Since the 1960 Federal Guidance (Fe60) was issued, quantitative
estimates of ionizing radiation risks have been developed, particularly
for cancer. These estimates are still uncertain. Making them involves
choosing the most accurate and relevant information from the large body of
research on radiation risks, because the reliability of available data
varies.
Harmful effects in humans have been clearly shown only for doses and
dose rates much higher than those to which most workers are exposed.
Therefore, risks at occupational levels must be estimated on the basis of
the data obtained at higher levels of exposure and an assumed response at
lower levels. Our estimates of the stochastic effects from ionizing
radiation are based on the assumption that the number of stochastic
effects at low doses is directly proportional to the dose. The constant
of proportionality is derived from the number observed at larger doses and
the assumption that there is no level of radiation without some potential
for harm. More exactly, we use the straight line which fits the data best
and passes through the point representing no effect at zero dose.
31
-------
A. Units
The amount of damage done to a tissue by ionizing radiation depends
mostly on the amount of energy the tissue absorbs. Energy absorption is
commonly measured in a unit called a rad. One rad is 100 ergs absorbed
per gram of tissue. Thus, one rad to twice as much tissue means that
twice as much energy has been absorbed. A person receives a "whole-body
dose" when the absorbed energy is distributed relatively evenly throughout
the body.
One rad is a very small amount of energy absorbed per gram.
Nonetheless, a dose of a few rad to body tissues can be harmful, because
the energy is in a form concentrated enough to ionize molecules - that is,
knock off their electrons. It requires little energy to ionize a
molecule. A 160-pound person who receives a whole-body dose of one rad
absorbs enough energy to ionize 7 billion billion (7 x 1018) molecules;
this is about 100,000 ionizations per cell. Fortunately, very few of
these ionizations interact with important constituents of cells,
especially the DNA molecule. Ionizations can cause fundamental changes
(either directly or indirectly) in the body's chemical constituents,
including DNA molecules. Our genes, which regulate much of our cellular
activity, are made of DNA. Cancer is probably due, in part, to certain
types of changes in cellular DNA. Mutations are inheritable changes in
DNA molecules.
*A 160-pound person who has received a whole-body dose of one rad has
absorbed the same amount of energy required to light a 75-watt light
bulb for only one-hundredth of a second. A person absorbs from a milk
shake, French fries, and large cheeseburger enough energy to light the
same light bulb for about 21 hours; 125,000 times as much. These
comparisons do not mean that these different forms of energy
(ionizing, visible light, thermal, and chemical) have similar effects.
32
-------
All ionizing radiation is not the same. Some consists of energetic
particles such as protons (hydrogen nuclei), beta particles (electrons),
and neutrons, or combinations of these, e.g., an alpha particle may be
thought of as a combination of two neutrons and two protons. Electro-
magnetic radiation of high enough energy per photon - x rays and gamma
rays - can also ionize molecules. For doses of the same size in rad,
different types of ionizing radiation act differently. Some, like x rays,
beta rays, and gamma rays, ionize molecules which are far apart, like this;
8 fim in tissue
Some, like alpha particles, make very dense ionization tracks like this;
0.2 Mm in tissue
33
-------
Alpha particles and protons are examples of "high-LET" radiation.
LET stands for linear energy transfer, the amount of energy deposited per
unit track distance. High LET means that the particle gives up large
amounts of energy along a short, densely ionized track. Low-LET
radiation, such as gamma rays and x rays, produces a long, sparsely
ionized track. "High LET" and "low LET" are broad and rather imprecise
categories. For example, some particles have sparse ionization at the
beginning of their tracks and dense ionization at the end. Also, the fast
electrons that high-LET particles knock off atoms themselves act largely
as secondary, low-LET radiation. In general, doses of the* same size in
rad from high-LET radiation are more dangerous than from low-LET radiation.
The biological effects of ionizing radiation can depend, among other
factors, on: the type of radiation; the size of the dose and the rate at
which it is received; the mass and type of tissues irradiated; and the
age, sex, race, genetic makeup, and other characteristics of the exposed
person. Because all the relevant factors and their precise effects are
usually not known, for radiation protection purposes we only consider the
amount and type of radiation, the tissues irradiated, and in some cases,
age and sex.
The ability of different types of radiation to cause harmful effects
is related by "quality factors." The quality factor for x rays and gamma
rays is defined as one. If the quality factor for another type of
radiation is five, this means that in some general way this type of
radiation is likely to cause five times as much harm as the same dose in
rad absorbed from x rays. The International Commission on Radiation Units
and Measurements publishes tables listing quality factors as a function of
34
-------
LET (IU71, IU73, IU76). In this general review of occupational guidance
we have not re-evaluated the specific quality factors in current use.
The dose in rad from a particular type of radiation multiplied by its
quality factor gives a quantity called "dose-equivalent." Dose-
equivalent is measured in a unit called the rem. For simplicity, in this
document we call "dose-equivalent" just "dose." The dose in rem is a
rough measure of health risk. This is why most radiation protection
limits, including ours, are expressed in terms of rem.
B. The Present State of Knowledge
In this section, we discuss the basis for the risk estimates we use
in this review. (These are given in Section C.) We cover risk of cancer
caused by radiation (radiogenic cancer) first, followed by hereditary
risks and then risks to the unborn following exposure in utero. Finally,
risks of nonstochastic effects are described.
1. Radiogenic Cancer
A number of long-term epidemiological studies to evaluate the
consequences of exposure to radiation are in progress. Almost all of
these studies have been reviewed in the 1972 National Academy of Sciences
report, The Biological Effects of Ionizing Radiation, commonly called the
BEIR report, and the 1977 report of the United Nations Scientific Commit-
tee on the Effects of Atomic Radiation (UNSCEAR), Sources and Effects of
Ionizing Radiation (Na72, Un77). More recently, the Interagency Federal
Task Force on the Health Effects of Ionizing Radiation has published a
report describing the health effects associated with radiation exposures,
35
-------
Report of the Work Group on Science (In79). The General Accounting Office
has just published a report, The Cancer Risks of Low-Level Ionizing
Radiation Exposure (Ge80). The National Academy of Sciences has recently
finished a revision of their 1972 report (Na80).
A particularly important source document for any review of radiation
risk is the Life Span Study; Report 8 - Mortality Experience of Atomic
Bomb Survivors 1950-1974 (Be78), which provides the most recent results
from the long-term study of persons exposed at Hiroshima and Nagasaki.
This study is particularly valuable because it has continued for a long
time, contains a large number of persons, and has been carefully
documented. Moreover, the population at risk was exposed on a known date
so that follow-up studies give some insight into when radiogenic cancers
appear and how long exposed persons are at risk following exposure. Even
so, the Life Span Study has many limitations.
The population studied contains 82,000 A-bomb survivors, of whom over
62,000 persons were still alive in 1974. Thus, even the most recent
results are based on far from a lifetime follow-up. Of the 3,842 cancer
deaths observed by 1974 in this population, only about 200 are thought to
be due to A-bomb radiation. These cancer deaths can be grouped into broad
intervals according to the dose received to obtain rough estimates of the
cancer risk per unit dose. Further subdivision of these data to obtain an
estimate of the risk by type of cancer or by age at exposure usually
results in a small sample size and, therefore, a relatively unreliable
estimate. It follows that more is known about the total risk of solid
cancers and leukemia from the A-bomb survivor study than about individual
cancers. In addition, the type of radiation thought to be
36
-------
important at Hiroshima, neutrons, is different from the major source of
exposure at Nagasaki, gamma rays. In many cases, but not all, this makes
combining the data from two cities a possible source of error. Moreover,
both of these populations were exposed almost instantaneously at very high
dose rates. The consequences of prolonged exposures at low dose rates,
such as occur in most occupational situations, may be different.
In spite of the limitations of the study of Japanese survivors and
other exposed groups, scientists are reasonably sure about which kinds of
cancers follow radiation exposures at high doses. Even though there is
less certainty on when cancers appear, how long the excess cancer risk
persists, and the magnitude of this risk per unit dose, some quantitative
estimates can be made. This is in marked contrast to the situation when
the 1960 guides were prepared and direct knowledge of radiogenic cancer
risks was quite limited. Table 3 indicates the kinds of cancer that have
been identified as radiogenic in the Life Span Study and in some other
epidemiological studies of persons exposed to high levels of radiation
(ln79). The number of persons at risk in these other studies is quite
small compared to the number of A-bomb survivors and we cannot be sure all
types of radiogenic cancers have been identified yet.
As important as the number of persons in an epidemiological study is
the length of time after irradiation they have been studied. This is
because most radiogenic cancers begin to appear only after a rather
lengthy latent period and because radiogenic cancers usually occur late in
life. People in major exposed groups have not been followed long enough
to observe the full extent of their risk of cancer. This must be
estimated by projecting, the excess harm observed to date over the rest of
the expected lifetime of the members of the groups.
37
-------
U)
CO
Type of
Cancer
Leukemia
Thyroid
Female Breast
Lung
Bone
Stomach
Esophagus
Bladder
Lymphoma (incl.
mult, myeloma)
Brain
Liver
Skin
Salivary Gland
Colon
Rectum
Atom Bomb Radiation
•i s
O 01
A T)
e §
O CO rH
*J M CO
"0
0) -H rH
M > rH
01 M Cd
id co co
a M
% £
**
** *
**
**
*
*
*
Ankylosing spondyl:
tis (x-ray)
Ankylosing spondyl]
tis (radium)
Benign pelvic
disease
** *
**
**
*
*
Medical Radiation
CO
4J co >\ co fl n)
(0 OIO.-rl.~->> M
Id .B O *J d .C «-N |
0) UU-rlOl4J(0 4J X
M 01 CO P. M 4J CO
J3 0) OIO 0) T3 T3 O CD O
(0 iH M U rH 01 Id l-l M
60* a o -H eo<4-i 4-1 oi
10 -H 3 « J3 M C O 4J
•HTl4JHOIUcd-H M 9
d *d rH -• o
* A* **
* **
** **
*
*
**
* *
* *
*
Occupational Radiation
01
10 4J
rH *J O
id u> co
•H »J -H •« Ifl
•U 01 60 M
4J o B 01
| C rH 3 0
9 iH O iH -H
•HOJ -ri as
m 4 rl
o2 (2 p
**
**
**
*
*
** *
TABLE 3. Cancers Linked to Radiation in Particular Populations. Strong associations
are indicated by **, and meaningful but less striking associations by *.
-------
The "risk period" for leukemia appears to be about 25 years (Be78),
but this is not true for most cancers. Current results from the Life Span
Study indicate that for most cancers the person exposed has an excess risk
for the rest of their life. Fortunately, the numerical risk estimates for
adults that we use are not very sensitive to the assumed length of the
risk period.
Ideally, estimates of lifetime risk would be based on a person's age
at the time of exposure and the observed chance of excess cancer as a
function of age. For most cancers the available data are too incomplete
to make this a feasible approach. Instead, two different kinds of
projection models are commonly used. These were developed by the NAS-6EIR
Committee for their 1972 report. To the extent that the risks of
radiation are independent of dose rate and increases linearly with the
dose, the different numerical results obtained with these models may
indicate the possible range of the future risk.
The two projection models are called the absolute risk model and the
relative risk model. In the absolute risk model it is assumed that, after
the latent period, the risk per unit of dose remains constant throughout
the risk period. The risk coefficient for the absolute risk model is
found by dividing the observed number of excess cancers by the collective
dose to the population and by the number of person-years of risk period in
the population during the time of observation. We have used this risk
coefficient, the number of excess fatal cancers per rem per person-year at
risk, to estimate the number of excess fatal cancers in adults exposed at
various annual dose rates and having the life expectancy predicted by 1970
mortality statistics (see Section C below)(Bu80, Na70).
The risk period means the time from the end of the minimum latent
period until the exposed persons no longer have an excess risk.
39
-------
In the relative risk model it is assumed that, after the latent
period, the risk per unit of dose is a constant proportion of the normal
incidence of cancer, which depends on age. The relative risk coefficient
is therefore not based only on the absolute number of observed excess
fatal cancers per unit dose, but on the ratio of this value to the
age-dependent normal incidence for the population under observation.
Relative risk coefficients, percent increase per rem, are used in
Section C to calculate the numerical increase in fatal cancer on the basis
of age-specific U.S. cancer mortality in 1970 (Na75) and 1970 U.S. overall
mortality statistics (Bu80, Na75).
The two models yield different numerical results when the data are
extrapolated to account for years of life in the exposed population beyond
those covered by follow-up of the study group. For most, but not all,
fatal cancers the relative risk model projects a larger number of
radiogenic cancers, because for most cancers the normal incidence
increases rapidly with age. However, the relative risk model predicts
that death will occur at an older age, on the average. Thus, the two
models tend to predict a similar total number of years of life lost in an
exposed population.
Section C includes only estimates of fatal cancers, not estimates of
the total of fatal and nonfatal cancers. The risk of nonfatal radiogenic
cancers is not calculated because little information is available on their
incidence. Almost all of the epidemiological studies are based on
mortality. In the absence of specific data on nonfatal radiogenic
cancers, the total risk of radiogenic cancer can be roughly estimated from
State and national health statistics on cancer incidence and mortality in
the general population. One way to do this is to compare the ratio of the
40
-------
incidence of fatal cancers to the incidence of all clinically observed
cancers. Such estimates are not too satisfactory, not only because of the
possibility of differences in the relative frequency of cancer types
between radiogenic cancers and those caused by other factors, but also
because cancer incidence statistics are incomplete and not directly
related to cancer mortality statistics. Studies of survivorship following
treatment are another possible source of mortality to incidence ratios.
However, most of these studies are from exemplary medical centers and may
not accurately reflect the national situation.
The 1972 BEIR Committee estimated the probability of a nonfatal
cancer to be about the same as the probability of a fatal cancer (Na72).
While this ratio is reasonable for breast cancer and many other cancers,
there are exceptions. Skin and thyroid cancer have very low fatality,
probably less than 6% (Un77). On the other hand, the mortality for lung
cancer and for leukemia in adults approaches 100%. We estimate that the
total number of discovered clinically observable radiogenic cancers,
excluding skin cancer, is one and one half to two times the number of
fatal cancers.
Because breast cancer is one of the most common radiogenic cancers,
the total risk to men and women following whole-body exposure is probably
not the same. On the basis of the absolute risk model, breast cancer
makes the total radiation risk of fatal cancer for women about twice that
for men. On the other hand, because of prevalence of lung and some other
cancers among men, the relative risk model projection of mortality due to
all cancers is 7% greater for men than women. The recent trend of
increased lung cancer in- women will reduce this margin. Male
41
-------
A-bomb survivors have a higher mortality risk from radiogenic cancer than
comparably exposed women, particularly at older ages (Mo78). In view of
the ambiguity in the available data, the estimated risks of cancer
fatality in Section C have been calculated using averaged risk
coefficients for both sexes. However, even if cancer mortality is about
the same for both sexes, there will be more nonfatal cancers observed in
women because they have more curable breast and thyroid cancers.
The numerical estimates of fatal radiogenic cancer that are listed in
Section C cannot be compared directly to general cancer mortality for U.S.
population. The latter reflects the age distribution of the whole U.S.
population while our calculations assume a cohort of workers who were 18
years old at the start of their exposure to radiation. Calculations based
on 1970 age-specific cancer mortality rates indicate that a worker in this
cohort has a 16% chance of dying of a cancer unrelated to occupational
radiation exposure. Use of more recent cancer mortality data would
increase this percentage by a small amount.
In Section C we have used the risk coefficients listed in the 1972
BEIR Report to prepare numerical estimates of the potential number of
fatal cancers from occupational exposures to radiation. While there is
little controversy about doing so for high-LET radiations, there is
considerable controversy about how well a linear extrapolation estimates
the cancer risk for low doses of low-LET radiation. Because of this, our
numerical estimates may be considered too high by some and too low by
others. We believe our estimates provide a reasonably conservative basis
* For solid cancers due to adult exposures, these risk coefficients
agree rather well with those in the 1980 BEIR Committee Report
(Na80), certainly within their inherent uncertainty. A more
definitive comparison will not be possible until the 1980 report
is evaluated.
42
-------
for regulations to protect public health. The available epidemiological
evidence is insufficient to either prove or disprove the linear,
nonthreshold hypothesis used to derive these values.
Although exposures of animals and cultured cells sometimes give
responses that are consistent with a nonlinear relation to dose, they are
usually consistent with a linear relation as well. Moreover, we do not
believe it is clear how these results should be applied to quantitative
estimates of risks in human populations, which, unlike cultured cells and
most laboratory animals, are inhomogeneous. Even for a population of
genetically identical individuals the shape of the dose response curve can
be very different from the shape of dose response curve for their cells;
and the shape of a dose response curve for a genetically diverse human
population can be very different from the shape of the curve for any
individuals in the population. Some of these points are briefly discussed
below and more complete statements may be found in the literature cited.
The risk estimates in Section C are for an imaginary cohort of
radiation workers - all of the same age and receiving the same annual dose
for a working lifetime. We then estimate the chance of fatal cancer
occurring in a hypothetical "average" individual. This is not the same as
estimating the risk to a particular real individual. In an inhomogeneous
population some persons are more susceptible to cancer than others,
because of genetic predisposition, age, personal habits, or other
factors. While the extent of such variability is currently unknown, it
can have an important influence on the average response of a population to
radiation. A recent General Accounting Office report explores this in
some detail (Ge81).
43
-------
Figure 4, taken from that report, shows the expected radiation
response in a hypothetical group having a highly nonlinear dose response
(response proportional to the dose squared) and various degrees of sensi-
tivity to radiation among its members. Although the example is arbitrary,
it illustrates that the overall response can be quite different from that
of any subgroup. In particular, it shows that a linear extrapolation of
the data can lead, over most of the dose range, to an underestimate of the
risk to those who are most sensitive to radiation, and an overestimate of
the risk to most people. At low doses it can lead to an underestimate of
the risk to the population as a whole. For this reason, we believe that
the experimental induction of radiogenic cancers in inbred strains of
rodents and other mammals does not provide a very useful basis for
predicting the shape of the dose response to radiation for inhomogeneous
human populations.
The risk estimates we have used are based on epidemiological studies
which include persons exposed to relatively large amounts of radiation
compared to occupational doses. The data from these studies is consistent
with several different types of dose response functions. The functional
form chosen for estimating risks can have a large effect on the degree of
risk predicted at low doses. In Appendix A we discuss an example commonly
cited as a non-linear dose response in an inhomogeneous human population -
leukemias in the Life Span Study of Nagasaki survivors - and why we do not
find this evidence convincing. The risk estimates in Section C are based
on a straight line fit through the data and an assumed zero risk at zero
dose. We believe this is a reasonable regulatory position for predicting
the dose response to radiation for human populations.
44
-------
oc
iu
CO
O
1U
CO
IU
0
oc
iu
O
1
OC
UJ
CO
3
1. Most sensitive group (10 people)
2. Moderately sensitive group (100 people)
Majority group (9790 people)
4 Resistant group (100 people)
5. Total
6. Quadratic extrapolation
10 20 30 40 50 60
DOSE (RAD)
80 90 100
Figure 4. The presence of groups of people especially sensitive to radia-
tion can cause the overall response of the entire population to
differ from the dose-response of any one group. The figure,
taken from Ge81, illustrates the effect of radiation on a
hypothetical population of 10,000 people, each of which has a
quadratic dose-response (with no linear component) with
saturation at some dose (i.e., at that dose the person is
almost certain to die from the exposure): 10 people very
sensitive to radiation-induced cancer, 100 people moderately
sensitive, 100 people resistant, and a majority of 9790 people
having typical (modal) sensitivity. In this example, the
population dose-response curve is approximately linear even
though the basic response (before saturation) of each group is
quadratic, i.e., increases as the square of the dose. For this
population, a purely quadratic extrapolation (curve 6)
substantially underestimates the risk. Even a linear
extrapolation can underestimate the risk in such examples.
45
-------
2. Hereditary Impairments From Exposure to Radiation
A mutation is an inheritable change in the genetic material
within chromosomes. We assume that ionizing radiation causes the same
kinds of mutations as those that occur from other causes. Generally
speaking, mutations are of two types, dominant and recessive, but these
categories are rough and somewhat arbitrary. The effects of dominant
mutations usually appear in the first and subsequent generations. The
effects of recessive mutations do not appear until a child receives a
similarly changed gene for that trait from both parents. This may not
occur for many generations. It may never occur. Although mutations may
in time be eliminated from the population by chance or by natural
selection, they can persist through many generations. The 1972 BEIR
Committee estimated that radiation-induced recessive mutations are spread
over 10 to 20 generations. Dominant mutations are usually expressed (and,
if deleterious, usually eliminated) in the first few generations.
Mutations can cause harmful effects which range from undetectable to
fatal. In this report when we refer to mutational effects we mean only
those inheritable conditions which are usually severe enough to require
medical care at some time in a person's lifetime. Even as limited by this
definition the range of seriousness of mutational effects is large. The
effect of one fairly common dominant mutation is extra fingers and toes.
However, some other dominant mutations can have much more severe effects,
such as increased susceptibility to cancer, severe mental retardation, and
muscular dystrophy. McKusick has classified over 55% of 583 "proven
autosomal (not sex-linked) dominants as clinically important."(Me75)
Most identified mutations are recessive, not dominant. The severity
ranges from changes in hair and eye color (not a mutational effect as
46
-------
defined above), to such dangerous diseases as hemophilia, Tay Sach's
disease, sickle cell anemia, and cystic fibrosis. The largest class of
genetic impairments, classified by the 1972 BEIR Committee as diseases of
complex origin, includes congenital malformations and constitutional
degenerative diseases having a genetic component. These "diseases," which
are thought to be caused by the cumulative effects of many mutations and
environmental factors, can cause serious handicaps. Examples are anemia,
diabetes, schizophrenia, and epilepsy (Na72).
Risk estimates for mutational effects caused by radiation are almost
wholly based on data from inbred strains of animals. There is no
completely satisfactory way to apply these data to genetically
inhomogenous human populations. Nonetheless, the 1972 BEIR Committee
estimated the dose needed to double the human mutation rate on the basis
of the average increase of recessive mutations per rem in large
populations of inbred mice. This average "doubling dose" could be
determined only within broad limits, 20 to 200 rem for low dose rate,
low-LET radiation. Using a very similar analysis, the 1977 UNSCEAR
Committee arrived at 100 rad as their estimate of the doubling dose. Low
LET radiation is about 3 times less effective per rem at low dose rates
than at high dose rates in producing genetic damage in the progeny of male
laboratory mice (Na72). For the progeny of female mice the effect of
decreasing the dose rate on the hereditary risk is even larger, lowering
it by a factor of twenty or more (Na72). Both the BEIR and UNSCEAR
Committee concluded that radiation-induced genetic damage in humans would
be similarly reduced at low dose rates.
In addition to an* estimated doubling dose based on recessive
mutations, the UNSCEAR Committee also made a second and more direct
47
-------
estimate of hereditary risk. This estimate is based on the first direct
measurement of radiation-induced dominant mutations, in this case, those
affecting skeletal tissues in mice. This is important because these
anomalies are due to rare dominant and irregularly expressed dominant
mutations, types of mutations generally thought to be major contributors
to mutational effects in humans. Moreover, the severity of the skeletal
changes observed in these mice were related to similar skeletal defects in
humans, so that the extent of potential impairment to humans could be
considered. Both of the UNSCEAR estimates are in substantial agreement
with each other and with those proposed by the BEIR Committee in 1972.
;,
The largest source of human data that can be used to estimate genetic
risks are the records of children of A-bomb survivors. So far, there is
little statistical evidence of genetic damage in these children (Ne74).
While this does not contradict other estimates of hereditary damage, the
number studied is too small to be conclusive. For types of genetic damage
causing death before age 17, a lower limit on the doubling dose for male
parents, based on the fact that no exposure-related mortality was observed
in offspring, is 46 rem; for female parents, it is 125 rem. Both of these
estimates are at a 95% confidence level and pertain to high dose-rate
exposures. When allowance is made for the effects of dose rate, these
lower limit estimates of doubling dose are, for low doses of low-LET
radiation, increased to about 140 rem for exposed males and to more than
1000 rem for exposed females, yielding an average doubling dose for both
sexes of about 250 rem (Ne74). This lower limit is about the same as the
highest value estimated by the 1972 BEIR Committee (200 rem).
In estimating the number of mutations, we assume a linear,
nonthreshold dose-response relationship. The risk of inherited mutational
48
-------
effects in children depends on a number of factors, including the sex of
the exposed parent, whether or not both parents are exposed, and the
gonadal dose before conception. Even for a constant rate of annual
exposure the effect of the gonadal dose is a function of the age of the
worker, because younger workers are more likely to have additional
children than older workers.
The sex of the worker is also an important factor. Animal
experiments generally show that at doses permitted by current guides,
low-LET radiations have a much smaller mutational effect on oocytes
than on spermatagonia. The 1972 BEIR Committee estimated the difference
between male and female sensitivity as a factor of five for low dose,
low-LET radiations. Because of this difference, we calculate the
hereditary risk estimates in Section C separately for each sex.
In summary, there are three estimates of hereditary risk - all based
on animal data and showing reasonable agreement. The 1977 United Nations'
UNSCEAR Committee estimates of dominant mutations agree with the more
indirect estimates made by the 1972 BEIR Committee. The upper and lower
bound estimates in the 1972 BEIR Report differ by a factor of about 20, a
degree of uncertainty which is consistent with what is known now about
hereditary risks due to radiation. In Section C, we have used the
estimates of the 1972 BEIR Committee to calculate the potential hereditary
harm from occupational exposures.
3. The Risk Due to In Utero Exposure
•Jy-jtr
An exposed unborn child is subject to more risk from a given dose
* Both rodent and human oocytes are formed prior to birth and are not
a product of continuous cell division in adults, as are sperm. In
their "resting stage" before being released from the ovary, oocytes
appear to have little sensitivity to mutations from radiation.
** For simplicity we will designate all the stages from conception to
birth as an "unborn child." These stages are discussed below.
49
-------
of radiation than is either of its parents. The biggest risks are of
inducing malformations and functional impairments during the early stages
of its development. A child is also more likely to get cancer if it
receives radiation in utero. Moreover, the oocytes in the female fetus
are much more sensitive to radiation-caused mutations than are those of
adult women (Na72).
It is likely that the major detrimental effect from radiation
received in utero is the induction of malformations and functional
impairments in the developing unborn child. The particular effect and its
severity depend on the stage of development when exposure occurs. The
development of a baby is usually divided into three stages: ovum, embryo,
and fetus. A fertilized human ovum becomes an embryo after about seven
days. The initial formation of body organs (organogenesis) is nearing
completion at about eight weeks, after which the embryo becomes a fetus.
The seven-month fetal period is mainly a period of growth, although
development of the central nervous system and some other organs continues
to some extent. Laboratory animals pass more quickly through similar
stages of development. Therefore the effects of experimental in utero
radiation on animal development, described below, are probably
qualitatively related to effects in humans.
Relatively few cells are present in the fertilized ovum, and animal
studies show that the most common radiation effect at this stage is
chromosomal injury leading to cell death. If enough cells are killed,
this usually results in an intrauterine "death." Less frequently,
malformation or neonatal death occurs. The dose response shows no
evidence of a threshhold and usually a greater effect per rem at low doses
50
-------
(5 rem, low LET) than at higher doses (Un77). In the mouse, the most
studied species, a one percent lethality rate per rem is reported (Un77),
but there is considerable variation in sensitivity among the species
studied.
After the formation of organs begins (the embryonic stage),
intrauterine death is less likely for doses below 100 rem; malformations
are the most common effect. The cellular organization of the embryo is
changing very rapidly during this stage. Cells become specialized and
start processes leading to the development of specific tissues in a fixed
sequence. Consequently, the effect of radiation varies from day to day,
causing different kinds and degrees of malformations depending on exactly
when the exposure occurs.
An unborn child is more sensitive to radiation during the embryonic
stage than in earlier or later stages of development. Although the dose
response observed in animal studies is usually less than linear at low
doses, in some cases the dose response is consistent with linearity
(Un77). There is no good evidence for a threshold down to doses as low as
5 rem (low LET). The types of malformations in different laboratory
animal species correlate with the developmental stage of the embryo.
There is no evidence that the human embryo is an exception to this general
pattern.
Defects in development caused by radiation in mice and rats include
skeletal malformations, brain and spinal cord malformations, alterations
of nerve cells and cortical architecture of the brain, heart and urinary
tract malformations, and eye defects (Un77). Both the frequency and
severity of these effects increase with dose. The UNSCEAR Committee has
51
-------
estimated for animals an increased frequency of 5 x 10~3 malformations
per rem (low LET), but emphasizes that this estimate is tentative and not
applicable to humans because of large interspecies differences.
During the fetal period, malformations are less common and less
severe. The major effect is reduced growth, which may persist throughout
life.
The effects of radiation on human development are not as well known
as for animals. Most observed human exposures have occurred randomly
throughout pregnancy and intrauterine doses are not known with much
precision. The observations that are available indicate that human
response is similar to that for animals. When an ovum is killed by
radiation, the death is usually not noticed. The major observed effects
are malformations, which can occur in all stages of development, most
frequently in the embryonic and early fetal stages. The most common
radiation-induced malformations in humans are impaired development of the
brain, skeleton, and eyes (Up69).
The central nervous system has a long period of development in an
unborn child and the brain is particularly sensitive to radiation injury.
This is reflected by the frequent occurrence of microcephaly (small head
size) among persons exposed in utero. Microcephaly is commonly defined as
a head size two or more standard deviations smaller than the average (for
any specific age). Its clinical importance is that it is often associated
with microencephaly (small brain), but is much more easily measured.
Mental retardation is strongly associated with microcephaly, particularly
when the microcephaly is severe. Microcephaly and other malformations
have been observed in clinical practice after high pelvic doses (250 rem
52
-------
of low-LET radiation) from radiation therapy. The most frequently
observed radiation-induced human malformations are small size at birth,
stunted postnatal growth, microcephaly, microphthamia (small eyes),
pigmentary degeneration of the retina and other eye defects, genital and
skeletal malformations, and cataracts (Un77).
Microcephaly occurred frequently among the children of Japanese
survivors exposed in utero, particularly among the Hiroshima survivors,
where there is a linear trend of increasing incidence with the dose from
mixed gamma and neutron irradiation. Figure 5 shows the dose response for
these survivors during the time span when the unborn child was at greatest
risk, 6 to 11 weeks after conception. Estimates of the jln utero dose are
based on recent evaluations (only about 8% of the in utero dose at
Hiroshima was due to neutrons)(Ke78, Be78). Even in the lowest dose range
(average _in utero dose, 1.3 rad), the frequency of microcephaly is 11%,
nearly 3 times that for the relatively unexposed controls, which was 4%
(Mi76). Although this difference could conceivably be due to sampling
error (only two cases were observed in the lowest dose range), the risk
observed in this range is linearly proportional to the risk observed at
higher dose levels, where the frequency of microcephaly is so high that it
is almost certainly not due to chance.
As an upper limit on microcephaly, the 1977 UNSCEAR Report lists a
probability of one in a thousand per rem. This estimate may not be
conservative since it is based on the dose to the mother's skin, not the
much smaller i.n utero dose. If the data shown in Figure 5 reflects a
linear non-threshold response at Hiroshima, there were between 5 and 20
chances in a thousand of inducing microcephaly for an in utero dose of one
rem during the most sensitive period (6-11 weeks post conception), when
53
-------
100
80
tu
u
oc
Ul
a.
U
a
1U
oc
u.
Q
lu
OC
111
60
40
20
(10.27+1.9n)
(8.3> + 0.6n)
|(4.8y+0.6n)
I (1.2Y+ 0.1n)
NO DOSE CO
I
(24.27+ 1.9n)
JTROLS
.a?.+ 3.5n)
10 20 30 40 50
100
-150
TISSUE DOSE IN AIR (KERMA) DUE TO GAMMA AND NEUTRON RADIATION-RAD
Figure 5. Frequency of microcephaly as observed at Hiroshima for
different dose categories (Mi76). Average in utero gamma-ray
(8) and neutron (n) doses in rad are shown Tn parentheses for
each dose range. There are 84 children in this group, 27 of
whom were affected. The sample size at each dose level is
small (2-7 cases) and thereby subject to considerable
statistical variation.
54
-------
neutrons are assumed to be between 50 and 5 times more effective per rad,
respectively, than gamma rays. This risk is much larger than those we
estimate below for mutational effects and cancer for the same dose.
However, we do not know if a minimum dose is required to cause micro-
cephaly or how dependent the damage is on the type of radiation.
Data on the frequency of microcephaly at Nagasaki would be useful for
estimating the dose response from low-LET radiation alone, but the number
of cases occurring in Nagasaki (15 total, and only 5 during the most
sensitive period) is too small to allow this. There is essentially no
difference in the reported incidence of microcephaly among all persons
exposed in utero in the two cities: 17% in Hiroshima, and 15% in
Nagasaki. Similarly, during the most sensitive period (6 to 11 weeks
after conception) the overall incidence was 32% at Hiroshima and 23% at
Nagasaki. A difference this large would occur by chance about 30% of the
time and is not statistically significant. In both cities the incidence
was 100% for doses larger than 60 rad during the most sensitive period.
For _in utero doses less than 60 rad during the most sensitive period, a
17% incidence was observed at Hiroshima and only 5.5% at Nagasaki. There
is a probability of 0.04 (i.e. 4%) that a difference this large would
occur by chance. This may indicate that in the lower range of in utero
doses causes other than radiation were involved, or possibly that the
neutron component at Hiroshima was particularly effective. However, at
doses higher than 60 rad, microcephaly was more frequent at Nagasaki than
at Hiroshima, so that for all exposures occurring before 18 weeks of
* All tests are for the null hypothesis, no difference between cities,
hypergeometric distribution for sampling from a finite population
without replacment (Wa60).
55
-------
pregnancy the incidence in the two cities was nearly the same. In any
case, the samples are too small to provide a firm basis for any
conclusions on the cause of differences between the two cities,
particularly since sources of in utero and maternal trauma other than
radiation were probably not the same within the two cities (Mi72).
Severe mental retardation was also observed in Japanese survivors
exposed in utero. This was often, but not always, accompanied by
microcephaly (Wo67). At Hiroshima an increased frequency of severe mental
retardation occurred at all exposure levels, but was not statistically
significant (at the 0.01 level) for in utero doses below 20 rad (B173).
Although at Nagasaki there was no increase in severe mental retardation
related to in utero doses less than 120 rad, the Nagasaki sample is so
small there would be a 25% chance of obtaining this result even if there
were no difference between the two cities (Wa60). Among all persons
exposed in utero there is no difference between the two cities.
Microcephaly and mental retardation are not the only dose-related
effects observed in Japanese survivors exposed in utero. Their height and
weight during childhood and as adults is less than for those not exposed
(Un77). Long-term mortality studies of survivors exposed in utero
indicate higher than expected death rates in the first year of life and
after ten years of age (Ka71). Among those receiving high in utero doses,
fetal and neonatal deaths were common (Un77).
Because of the sensitivity of the unborn to radiation, a number of
epidemiological studies have been performed to see if developmental
effects occur due to low doses of diagnostic radiation (Di73,Ha69,Ki68,
Op75). In contrast to the Japanese experience, such studies have shown
negative or equivocal results (Un77). Because these studies were
56
-------
comparable in size to that of the Japanese survivors, this may indicate
the importance of dose rate in initiating these effects. Studies of
laboratory animals indicate fewer effects per rem at low dose rates for
some, but not all, in utero effects (Un77).
The genetic and cancer risks per unit dose from in utero exposure
also exceed those for adult workers. Unlike those in adults, oocytes in
the female fetus are not in a resting stage, and may be nearly as
sensitive as male spermatogonia. According to the 1972 BEIR Committee
Report, this increases the risk of hereditary damage being transmitted by
the female line by about a factor of five (Na72). The most sensitive
period for genetic damage in both sexes is probably the last two
trimesters.
The 1972 BEIR Committee estimated the leukemia risk from in utero
exposure as ten times greater than that for adults who get the same dose
(Na72). The follow-up period for excess solid tumors, which have a longer
latency period than leukemia, has probably not been long enough to allow a
good estimate of the total risk for other cancers due to in utero
exposures. The absolute risk of getting fatal cancer, other than
leukemia, in the first ten years of life due to in utero exposure,
however, has been estimated as five times the risk that an adult has of
getting cancer within ten years of receiving the same dose (Na72).
4. Other Effects of Normal Occupational Levels of Exposure
Nonstochastic effects following large radiation exposures are
due to extensive cell killing coupled with imperfect repair. Laboratory
animals show little or none of these effects at small doses and severe
impairment at high doses. Loss of fertility by males is an example.
57
-------
Doses of several hundred rem to the human testes can lead to a permanent
loss of fertility; smaller doses cause only a temporary reduction in the
number of sperm cells (He67, Rw74). Fertility is not impaired at doses
permitted by the current guides limiting occupational exposure.
The blood-forming organs show nonstochastic effects at relatively low
doses. A single dose of 20 rad can cause a measurable drop in the number
of lymphocytes, but such changes are transitory (Wh71). Chromosomal
aberrations in circulating lymphocytes have often been observed after low
doses of radiation (Un77,Ev79). Some of these aberrations are permanent,
but they have not been identified as a cause of any clinical condition.
For other organs, acute doses of about 1000 rad are needed to cause a
demonstrable non-stochastic impairment (NC71).
A threshold for skin erythema (reddening) occurs at doses of a few
hundred rad for medium energy x-rays. Low dose rates or fractionation
increase the threshold enormously; skin doses of several thousand rads
occur in radiotherapy without permanent damage. Occupational radiation
protection limits for the skin are designed to limit the incidence of skin
cancer. Skin erythema does not occur at these dose levels.
Perhaps the most important nonstochastic radiation effect is cataract
induction. The lens of the eye differs from other organs in that dead and
injured cells are not removed. The size, location, and growth with aging
determine how much a cataract interferes with vision. Single doses of a
few hundred rem have induced opacities which interfere with vision within
a year. When the dose is fractionated over a period of a few years,
larger doses are required and the cataract appears several years after the
last exposure (Me62,Me72). Judgments on the adequacy of exposure limits
58
-------
for the lens are based on extrapolating these findings to exposure periods
well beyond the range of clinical observation (Ch79). For this reason,
such extrapolation should include a large degree of safety.
Another major problem in selecting an occupational dose limit for the
lens is that animal studies indicate that minor opacities are produced at
dose levels as low as 30 rads of x-rays or 0.5 rads of neutrons (Ba71).
How much these minor opacities may increase in size with age is not known,
particularly in long-lived species such as man.
C. Risk Estimates Used in this Review
As used here, "risk" is the probability of harm from radiation
exposure. The term "risk coefficient" means the risk per unit of dose
equivalent (rem). Three kinds of risk are considered: radiogenic cancer,
hereditary effects, and effects from in utero exposures.
1. Radiogenic cancers.
We have used the risk coefficients and other parameters shown in
Table 4 to estimate the risk of cancer death for whole-body exposure
extending uniformly over a working lifetime, based on the absolute risk
and relative risk models. Except for leukemia, the expression period
(risk period) following the latent period is assumed to be the balance of
a lifetime. The 1969-71 life table for the U.S. population was used to
represent the normal mortality of workers (Na75).
Estimated future annual risk of cancer death is shown in Figure 6 for
an 18 year-old who will receive one rem per year to age 65, unless death
59
-------
Table 4.
Coefficients and Projection Models Used to Estimate the Risk
of Fatal Cancer due to Whole-Body Exposure of Adult Workers (Na72)
Model
Expression
Latent Period Period
Risk Coefficient
(cancer) (years)
Absolute Risk
Leukemia 2
All Other Cancers 15
Relative Risk
Leukemia 2
All other cancers 15
(years at risk)(per rem; average
for both sexes)
25
lifetime
25
lifetime
(cases/person-year at risk)
lxlO~6
5x10
-6
(percent increase)
2%
0.2%
60
-------
I-
X
cc
oc
Ul
0.
c
Ul
u
I
u.
(0
tt
8
Ul
S
Ul
RELATIVE RISK MODEL
ABSOLUTE RISK MODEL
100
110
Figure 6 Average future annual risk of radiogenic cancer death for an
individual of age 18 who will receive one rem per year to age
65 or until death from any cause, whichever occurs first. For
higher or lower dose rates between zero and five rems per year
the curves are changed proportionately. With either of the two
risk models shown most radiogenic cancer deaths are projected
to occur beyond the age of retirement. The curves fall to zero
at old age. because other causes of death overwhelm the risk of
radiogenic cancer.
61
-------
from any cause occurs earlier. The curve falls off at high ages because
the chance of dying from causes other than radiogenic cancer increases
during old age. The lifetime risk faced by such an 18-year-old is
obtained by summing the annual risks shown in Figure 6 over all ages. The
age-dependent risk of cancer increases nonlinearly and remains at an
elevated level long after exposure is over. This is due to the effect of
latency, variation with age of mortality rates, and, in the case of the
relative risk model, the age-dependence of "natural" incidence of cancer.
Although the estimated risk of death is greater for the relative risk
model, death from radiation is predicted to occur earlier, on the average,
by the absolute risk model.
Figure 6 shows projected risk at age 18. Different results would be
obtained for initiating exposure at a later age, or for a worker who has
already survived to any age beyond 18. The risk to a worker who has
survived an advanced age will be greater. Figure 7 shows the annual risk
at any attained age for a worker exposed to one rem per year from age 18.
(For attained ages beyond 65 the exposure is assumed to cease at age 65.)
The figure includes the cumulative effect of all previous doses, but does
not drop to zero at old age because it assumes that the worker has
survived to each age shown.
Because annual risks vary so much, they are not very useful for
evaluating occupational exposure limits. Lifetime risk and the average
number of years of life lost associated with a constant level of exposure
throughout a working lifetime are more useful quantities for this
purpose. Lifetime risk is defined here as the probability of incurring a
specified radiation-induced effect due to receiving a specified dose
62
-------
o
X
DC
UJ
DC
Ul
O.
00
III
Q
K
IU
O
v>
E
o
X
111
10
AGE (YEARS)
Figure 7 Average annual risk of radiogenic cancer death for a surviving
average individual receiving one rem per year from age 18
Co 65. The figure shows the risk for the year at each attained
age; it does not show risks in either future or past years.
The risks do not fall to zero because those shown are for
workers who survive all prior causes of death; it falls off
slightly in old age because the expression period for leukemia
from the last doses received has expired.
63
-------
annually over a working lifetime, that is, from age 18 to 65 unless death
from any cause intervenes. When the dose received annually is the
maximum permitted by a guide (e.g., 5 rem), this risk is called the
maximum lifetime risk for that guide. Average lifetime risk is defined
as the lifetime risk associated with the average annual dose actually
experienced under the guide by the national work force or by any
specified subgroup. Analogous quantities can be defined for years of
life lost. A life table analysis (Bu70,Co78), which adjusts for the
competing effect of normal causes of death, was used to estimate these
lifetime risks of death and lost years of life.
Depending upon which risk model is used, the maximum lifetime risk
for death from radiation-induced cancer is estimated to be from 3 to 6 in
a hundred for the extreme case of an annual whole-body dose of five rems
per year received throughout a working lifetime. Figure 8 shows lifetime
risks faced by an 18-year-old entering the work force for doses ranging
from zero to five rems, the maximum range of average annual exposure
rates permissible under current guides. As illustrated, limiting the
expression period of cancers, other than leukemia, to 30 years does not
have a large effect on the lifetime risk.
Table 5 lists the average lifetime risks of death due to cancer for
radiation workers in various occupational categories assuming they are
exposed each year from age 18-65 at the average dose rates observed in
1975. These annual average doses are well below one rem per year; the
average lifetime risks are therefore correspondingly smaller than the
maximum lifetime risk.
A life table analysis provides two other indicators of the cancer
risk due to occupational exposure: (a) the average reduction in life
64
-------
IU
Q
UJ
cc
111
OC
0.
UL
O
bi
CO
E
UJ
Ul
u.
oc
UJ
.07
.06
.05
.04
.03
.02
.01
0
RELATIVE RISK,
LIFETIME EXPRESSION
RELATIVE RISK.
30 YEAR EXPRESSION
ABSOLUTE RISK,
LIFETIME EXPRESSION
ABSOLUTE RISK,
30 YEAR EXPRESSION
ANNUAL DOSE EQUIVALENT (REM)
Figure 8. Average lifetime risk of death due to radiogenic cancer by
annual dose level for four risk models. It is assumed that
this dose level remains constant from age 18 to 65. Limiting
the expression time for cancer to 30 years has relatively
little effect on the lifetime risk.
65
-------
Table 5
Estimated Lifetime Risk of Death Due to Radiogenic Cancer
for Constant Annual Exposure in
Various Occupational Categories*
Lifetime Risk
Annual Dose Relative Absolute
Occupation (rem) Risk Model Risk Model
Education
Government
Medicine
Industry
Nuclear fuel cycle
Average for all
Present maximum
0.
0.
0.
0.
0.
0.
5.
20
23
32
52
63
35
0
Chance without occupational
* Assumed exposur
1
1
1
1
1
1
1
radiation
in
in
in
in
in
in
in
370
320
230
140
120
210
16
1 in 6
1
1
1
1
1
1
1
in
in
in
in
in
in
in
e is from age 18 to 65 at the average dose
910
790
570
350
290
520
37
rates
observed in 1975 to workers measurably exposed only.
66
-------
expectancy for the work force, and (b) the average number of years of
life lost for each excess cancer death (Co78). Figure 9 shows the average
reduction in life expectancy due to excess cancer for a group of 18
year-olds entering the work force as a function of lifetime exposure at
annual doses ranging from zero to five rems. These are average reductions
for the whole group. For those individuals who actually die of
radiation-induced cancer the reduction in life expectancy is much greater
than the average value for the work force shown in Figure 9. The risk of
such death is that shown in Figure 8. The average number of years of life
lost for each cancer death has a relatively constant value over the range
of dose levels normally experienced in occupational situations - 12 years
and 18 years for the relative and absolute risk projection models,
respectively.
Table 6 lists the average loss of life expectancy projected due to
death from radiogenic cancer for radiation workers in various occupational
categories, assuming they are exposed each year at the average dose rates
observed in 1975. These average lifetime losses of life expectancy are
much smaller than the maximum lifetime loss of life expectancy for annual
doses of 5 rem.
Risk estimates for individual types of cancer are considerably less
reliable than for the total of all cancer fatalities, as previously noted
in Section A of this Chapter. Of the various radiogenic cancers,
leukemia, breast cancer, and lung cancer occur more frequently in exposed
populations than fatal cancers of other types and are currently thought to
account for about half the total risk of fatal radiogenic cancer.
The International Commission on Radiological Protection has developed
weighting factors for the individual organs. These describe the
67
-------
I-
o
o
1
o
111
a.
X
UJ
ui
z
o
Q
ui
cc
IU
§
cc
UJ
10
9
8
7
6
5
4
3
2
1
RELATIVE RISK.
ABSOLUTE RISK
ANNUAL DOSE EQUIVALENT (REM)
Figure 9. Average reduction in life expectancy due to radiogenic cancer
by annual dose level for two risk models. It is assumed that
the annual dose rate remains constant from age 18 to 65.
68
-------
Table 6
Estimated Loss of Life Expectancy in Days Due to Radiogenic
Cancer Death for Constant Annual Exposure in
Various Occupational Categories*
Lost Life Expectancy (months)
Annual Dose
Occupation (rem)
Education
Government
Medicine
Industry
Nuclear fuel cycle
Average for all
Present maximum
0.20
0.23
0.32
0.52
0.63
0.35
5.0
Relative
Risk Model
0.4
0.5
0.6
1.0
1.2
0.7
9
Absolute
Risk Model
0.2
0.3
0.4
0.6
0.7
0.4
6
Assumed annual exposure is from age 18 to 65 at the average dose
rates observed in 1975 to workers measurably exposed only.
69
-------
proportion of the total risk (including both fatal cancer and mutational
effects in the first two generations) from whole-body exposure of adult
workers which is assumed to arise from each of the various organs (IP77).
The proportion of total cancer risk allocated to various organs by the
ICRP is comparable to that identified by the 1972 NAS-BEIR Committee.
These weighting factors were adopted by the ICRP to estimate the risk due
to non-uniform exposure of workers, such as by inhalation or ingestion of
radioactive materials. We have adopted the weighting factors used by ICRP
after excluding the ICRP weighting factor for the gonads (which applies
only to mutational effects) .and renormalizing the sum of weighted risks to
unity. These renormalized weights, which apply to fatal cancers only, are
listed in Table 7. Only six organs are identified by name. Organs
usually considered under the heading "other" are four portions of tne
gastrointestinal tract, kidneys, liver, pancreas, spleen, uterus,
adrenals, muscle, and bladder wall - organs in which inhaled or ingested
radioactive materials may be concentrated. Each of the five "other"
organs accumulating the vighest doses from any such material are given an
equal weight (0.' ) in the above scheme.
2. Hereditary Effects
Ranges of the estimated chance of mutational effects per live
birth due to an accumulated gonadal dose of one rem before conception are
listed separately for fathers and mothers in Table 8 (Na72). For
perspective, the current incidence in a child of unexposed parents is
about 10%. If both parents are exposed, the risks shown should be added.
These estimates are for low-LET radiation. Dose equivalents from high-LET
70
-------
Table 7
Assumed Risks of Fatal Cancer for Individual
Organs Relative to Cancer Risk for
the Whole-Body.
Organ Relative Risk
Breast
Lung
*
Red Bone Marrow
Thyroid
Bone Surfaces
Skin
**
Other Organs
0.20
0.16
0.16
0.04
0.03
0.01
0.08
* Assumes leukemia only.
** Applies to each of the five other organs with
highest dose.
Table 8
Range of Risk Coefficients for Mutational Effects (Na72)
Effects per 100,000 live births per rem
First Generation All Generations
Fathers 1-16 5-120
Mothers 0.2-4 1-30
* These workers are only applicable to doses of low-LET radiation.
71
-------
radiation, e.g., neutrons and internal alpha emitters, have a greater
hereditary risk.
The risk coefficients shown are for mutational effects for two
different cases: 1) first generation liveborn children, and 2) all
generations of liveborn children. The former can be applied directly
to the preconceptual gonadal dose of parents to determine the average
risk to each liveborn first generation child. Both cases require
assumptions on the expected number of children to parents in order to
derive an estimate of total risk, either to first or to all
generations of children, from exposure of a worker.
The expected number of mutational effects in children of an
exposed parent is a function of his or her accumulated gonadal dose.
We have calculated these risks for first generation children for
assumed constant exposure of parents starting at age 18, for average
parenting rates and ages at conception. The resulting values are
shown in Figure 10. The number of mutational effects in all
generations will be about six times greater than those estimated for
the first generation alone. The expected number of first generation
effects was calculated for the average number of children in 1975 as a
function of parental age. This average, 2.1, includes childless
married persons, but not unmarried parents. The expected number of
children is probably lower now, but the average age of parents at
conception (and therefore their average preconceptual gonadal dose)
may be higher.
72
-------
oc
Ul
oc
o
5
Q
Ul
CO
e
X
z
u.
o
z
Ul
oc
0
-1
I
o
Ul
X
2
p
o
Ul
u.
u.
Ul
_l
z
o
1-
s
u.
o
CO
E
0.012 -
0.011 -
•
0.010 -
•
0.009 -
«
0.008 -
.
0.007 -
m
0.006 -
0.005 -
-
0.004 -
-
0.003 -
0.002 -
-
0.001 -
RANGE FOR WOMEN
1234
ANNUAL DOSE (REM)
Figure 10. Risk that first generation children of men or
women exposed beginning at age 18 will have a
radiation-induced mutational effect as a
function of the parent's annual dose rate. The
risk to all generations combined is about six
times greater.
73
-------
3. Effects of In Utero Radiation
Table 9, taken from the 1972 BEIR Report (Na72), lists risk
coefficents for leukemia and solid tumors due to in utero exposures from
low LET radiations. These risk coefficients are much greater than those
for adults for equal doses. However, the period over which the risk
continues is assumed to be appreciably shorter, cf Tables 4 and 9. The
resulting absolute lifetime risks, for equal doses, are about twice those
for adults. Unlike the case for adults, numerical estimates of the cancer
risk for in utero exposure using the absolute risk model exceed estimates
based on the relative risk model. This is because the normal rate of
cancer in children is low. Hereditary risks due to in utero exposure are
not well known, but we assume that the risk per rem for men shown in
Table 8 applies to both sexes, since animal studies indicate that the
radiosensitivity of the prenatal oocyte is comparable to that of
spermatogonia (Na72). This leads to a hereditary risk about twice that
for adults, for equal doses.
The above cancer and hereditary risks from in utero exposures may be
small compared to the risk of malformations and other developmental
effects. This would be so, if there is no threshold dose for
developmental effects and the response increases at least linearly with
dose. As outlined in Section B of this Chapter, the data for Japanese
children may indicate, for microcephaly, a risk coefficient as large as
5 x 10~3 to 2 x 10~2 per rem for an instantaneous dose of mixed
neutron and gamma radiation delivered during the most sensitive period.
This is more than ten times the lifetime risk for leukemias and solid
cancers, which is 5 x 10-4 per rem (see Table 9). Moreover, the
74
-------
Table 9.
Coefficients and Projection Models Used to Estimate the Risk
of Fatal Cancer due to In Utero Exposure (Na72)
Model
Latent Period
Expression
Period
Risk Coefficient
(cancer; Tyears;
Absolute Risk
Leukemia 0
All Other Cancers 0
Relative-Risk
Leukemia 0
All other cancers 0
(years at risk;
10
10
10
10
(.per rem; average
for both sexes)
(cases/person-year at risk)
25xlO~6
25xlO~6
(percent increase)
50%
50%
75
-------
occurrence of other kinds of malformations adds to these risks. However,
for several reasons, we do not believe the data on Japanese children are
alone sufficient to be a basis for numerical risk estimates. Although the
Japanese results are clearly related to dose, a number of other traumas,
including malnutrition and disease (Mi72), could have contributed to the
effects observed, and would not contribute to the in utero risk from
occupational exposures. Moreover, if the risks observed in Japan occurred
proportionally in other populations at lower doses and dose rates, it is
unlikely that the studies of in utero effects due to low doses of
diagnostic and other sources of in utero exposure would be negative
(Un77). These negative results do not indicate there is no danger to the
unborn from low doses of occupational radiation, but they do indicate the
Japanese results may not be applicable to all types of exposures. The
presence of high LET radiation at Hiroshima and the instantaneous nature
of the dose in both cities may be important confounding factors.
Developmental effects of radiation on an unborn child depend to a
large extent on the time of exposure. In general, the most vulnerable
period for these effects is the first several months after conception,
when a woman is least likely to know whether or not she is pregnant. This
is in contrast to radiogenic hereditary effects in future children of the
workers, to which the unborn are more susceptible later in the gestation
period. A major concern is that, without special precautions, it will be
possible for an unborn child to receive a significant dose when the mother
does not know that she is pregnant and when the unborn child is especially
sensitive to developmental effects from radiation. Our inability to
quantify this risk more completely does not lessen this concern.
76
-------
IV. GENERAL PRINCIPLES FOR THE PROTECTION OF WORKERS
Three basic principles have governed radiation protection of workers
in recent decades, both in the U.S. and in most other countries. Although
the precise formulation of these principles has evolved over the years,
the basic intent has remained unchanged. The first requires that any
activity producing occupational exposure be useful enough to society to
warrant the exposure of workers; i.e., a process of "justification" must
be carried out. The second requires that, for justified activities,
exposure of the work force be the lowest that is reasonably achievable;
this has most recently been characterized as "optimization" of radiation
protection (IP73, IP77). Finally, in order to provide an upper limit on
risk to individual workers, "limitation" of the maximum allowed individual
dose is required. This limitation is required above and beyond the
protection provided by the first two principles, because their primary
objective is to minimize the total harm from occupational exposure in the
entire work force, and they do not limit the way that harm is distributed
among individual workers. These three principles are discussed in turn
below.
A. Justification of Activities Leading to Worker Exposure
Since any exposure to ionizing radiation is assumed to involve risk
of harm, no exposure should be permitted unless it cannot reasonably be
avoided and it will result in a benefit - both to the worker exposed and
to society in general. This requires two risk-benefit decisions. The
77
-------
first can be made by the worker, if he is properly informed of the risks,
who can judge for himself whether the benefit of employment is sufficient
compensation.
The judgment of benefit to society is less easily made. Only
recently has the U.S. Government explicitly required that such general
judgments be made for major Federal activities - through the National
Environmental Policy Act of 1970 (NE70). There is no comparable general
requirement for other activities. An obvious difficulty in drawing these
judgments is the lack of common units of measurement (or in some cases the
lack of any units of measurement) for a quantitative analysis of costs
(including risks) and benefits. Given this situation, informed value
judgments are necessary and are usually all that is possible (NA77).
The need to justify activities that result in occupational doses has
traditionally been a part of guidance for radiation protection, even
though it has seldom been possible to give it direct regulatory
implementation. In the 1960 guidance the FRC said: "There should not be
any man-made radiation exposure without the expectation of benefit
resulting from that exposure" and "It is basic that exposure to radiation
should result from a real determination of its necessity" (Fe60). Other
advisory bodies have used language which has essentially the same
meaning. In its most recent revision of international guidance (1977) the
ICRP said "...no practice shall be adopted unless its introduction
produces a positive net benefit," (IP77) and in slightly different form
the NCRP, in a recent (1975) statement of position, said "...all exposures
should be kept to a practical minimum;...this...involves value judgments
based upon perception of compensatory benefits commensurate with risks,
78
-------
preferably in the form of realistic numerical estimates of both benefits
and risks from activities involving radiation and alternative means to the
same benefits" (NC75).
This principle is adopted in these proposals as Recommendation 1 in a
simple form: "All occupational exposure should be justified by the net
benefit of the activity causing the exposure. The justification should
include comparable consideration of alternatives not requiring radiation
exposure." We offer no specific advice on how costs, risks, and benefits,
which are frequently incommensurate or unquantifiable, should be handled
so as to show that this judgment has been properly reached for specific
activities. It is perhaps useful to observe, however, that throughout
history men and societies have formed risk-benefit judgments, with their
usefulness usually depending upon the amount of accurate knowledge
available. Since more is known about radiation now than in previous
decades, the prospect is that these judgments can now be better made than
before.
The preceding discussion has implicity focused on the need to justify
entire activities, such as the construction and operation of a facility,
or instituting a practice involving radiation exposure of workers.
However, this principle is often most useful at a different level, that of
detailed regulation of facilities and direct supervision of workers.
Decisions about whether or not particular tasks involving exposure to
otherwise justified activities should be carried out (such as inspecting
control systems or acquiring specific experimental data) require judg-
ments on justification which may, in the aggregate, be as significant for
reducing exposure as those justifying the basic activities these tasks
support.
79
-------
B. Optimization of the Protection of Workers
When it has been determined that an activity requiring exposure of
workers is justified, the next step traditionally required is to reduce
the risks to levels that are "as low as is reasonably achievable"
(commonly designated by the acronym "ALARA"). This process is typically
carried out in two different ways. First, it is applied to the
engineering design of facilities so as to lower exposure of workers as far
as is economically justified. Second, it is applied to actual operations;
that is, work practices are designed and supervised to minimize exposure
of workers. Both of these applications of ALARA are encompassed by
Recommendations 2 and 3, which apply to collective and individual
exposures, respectively. The Minimum Radiation Protection Requirements of
Recommendation 4 give more specific guidance on means for insuring that
ALARA is implemented for various levels of worker exposure. These minimum
requirements, which encompass education of workers on health risks and on
radiation protection measures, provision of radiation protection
supervision, monitoring of exposures, and limitation of lifetime dose, are
discussed in Chapter V.
The optimization of radiation protection of workers may sometimes
involve the choice between minimizing collective dose to all of the
workers involved in an activity on the one hand and minimizing dose to the
most exposed individual on the other. In such cases, minimization of
collective dose should generally take precedence, unless the limits
permitted by maximum allowed annual doses to workers may be exceeded, or
large lifetime doses to individuals would be incurred. Such a procedure
80
-------
will minimize the total harm from radiation while preserving the
protection afforded workers against excessive individual risk.
C. Limitation of Risk to Individual Workers and their Descendants
The above requirements are not sufficient by themselves. The harm
from exposure to radiation is incurred by workers who, although they
receive the direct benefits of employment, are usually not the principal
beneficiaries of the activities involved. Limits are therefore required
to assure that the maximum risk to every worker is low. These limits are
provided by regulations which are bounded by numerical guides to Federal
agencies for maximum allowed doses. These numerical guides are the
Radiation Protection Guides (RPGs) provided in Recommendation 3.
Recommendation 6 provides for more stringent limits to be established by
regulatory authorities when this is appropriate. Specific values for the
RPGs are discussed in Chapter VI. We describe here the general
considerations which governed their determination.
Two measures of risk are particularly significant to the individual
worker: first, the risk in his specific job to himself and his descen-
dants, and second, the maximum risk allowed, barring accidents. For most
types of harm from radiation, the first of these is proportional to the
average exposure for the job and the second to limits set by regulations
bounded by this Federal radiation protection guidance. An index of
societal interest is the total somatic and genetic risk from all
occupational exposures and, thus, the total harm to society. This depends
on the collective somatic and genetic dose to the entire work force and to
any exposed unborn.
81
-------
In these recommendations we have tried to insure that the two
measures of individual risk referred to above will be no greater than
those from most other common occupational hazards and, to the extent
feasible, that they will be lower. This approach is the same as that
recommended by the ICRP (IP77). We know of no other criteria which
provide a more rational approach to judging the acceptability of a guide
than these, when they are coupled with the first two basic principles for
radiation protection outlined above. We have also tried to design the
guidance so that the total harm to the entire work force and its
descendants will be as small as possible, while still limiting the maximum
harm to individual workers and descendants. Finally, we have estimated
the total harm to the population as a whole and found that it is small.
Assuming experience for the year 1975 is typical for radiation exposures
of workers, and using the risk estimates developed in Chapter III, the
total harm to the population from a constant annual collective dose equal
to that in 1975 is projected to be an increase of about two to five
thousandths of one percent in the annual cancer death rate, and a
comparable rate of increase in the number of liveborn with mutational
effects.
A striking feature of national statistics on occupational exposure is
the large proportion of all potentially exposed workers who receive annual
doses that are less than 500 millirem. This dose is one-tenth of the 1960
guide of 5 rem average dose per year and only four percent of the maximum
of 12 rem permitted in any single year. In 1975, the latest year for
which extensive data are available, 95% of all occupationally exposed
workers were in this group. Furthermore, all but six of 25 individual
categories (see Table 2) have average annual doses of less than one half
82
-------
of this value. These exceptions are nuclear waste workers, industrial
radiographers, licensed and state registered source manufacturers, nuclear
power reactor workers, and workers in fuel fabrication and reprocessing.
Three major groups of workers - all those in medicine, government, and
education - include no job category with an average annual dose greater
than 250 millirem.
These statistics appear to testify to the success of radiation
protection under the 1960 guidance. The typical risks in all occupations
which involve radiation exposure appear to be small, both absolutely and
in relation to other occupational risks (see Chapter VI). On the other
hand, in many cases these doses are low because people in many jobs
naturally have little exposure. And in all of these occupations the
existing guides permit far higher doses than those commonly received.
These statistics lead to two obvious questions: 1) Should the radiation
protection guides be so much higher than the demonstrated need for
exposure of the vast majority of workers? and 2) To what extent are the
infrequent doses that are above a few hundred millirem really necessary?
Regarding the first question, we believe that the present guides,
which permit doses from 5 to 12 rem in a single year, do not, by
themselves, sufficiently protect most of the radiation work force. The
1960 guidance is, in effect, designed to control doses to the few percent
of the work force whose work requires high exposures. The annual limits
for most workers could be reduced to lower values if suitable provision is
made for occasional higher exposures which are justified.
Detailed data on the extent of the need for annual doses above a few
hundred millirem are not available for the entire work force, although
83
-------
many individual cases of exposure in this range exist (En80). We believe
that adequate justification for some such exposures exists, and that the
guides should provide for this as long as a reasonably low upper limit on
the maximum allowable risk is maintained.
Given the above conclusions, Federal radiation protection guides
could take several forms. One alternative is to specify different guides
for different occupations. However, special studies for each occupational
exposure situation are required to do this well, and reliable information
for determining what maximum exposures are justified in specific
occupations is most appropriately obtained by the regulatory agencies.
Another alternative is to specify increasingly stringent protection
requirements for a set of successively higher ranges of dose, within a
basic upper limit which permits occasionally necessary higher than usual
doses. Such a system should discourage higher doses except when they are
well justified. Regulatory agencies should also then develop
supplementary lower limits for specific types of workers, based on
detailed studies as required, whenever this is appropriate. We believe
this is a more reasonable form for general Federal guidance than direct
specification of different Federal guides for different occupations. It
places the responsibility for such detailed decisions for particular types
of workers where it belongs, in the regulatory agencies who are directly
involved in the specifics of working conditions. We have adopted this
approach in formulating Recommendations 3, 4, and 6, since it
simultaneously avoids the permissiveness of a single high limit that is
only occasionally justified, and the arbitrariness of imposing the lower
limits appropriate for most jobs on the few that are justified exceptions.
84
-------
V. MINIMUM RADIATION PROTECTION REQUIREMENTS
In Chapter IV we concluded that the most appropriate guidance for
occupational radiation protection includes a set of dose ranges within a
basic upper limit, with increasingly strict requirements as the doses
increase. We have proposed Minimum Radiation Protection Requirements for
three such ranges in Recommendation 4. These requirements include:
1) education of workers about the risks to health from radiation and about
radiation protection requirements and practices; 2) supervision of
radiation protection, including the justification and optimization of
exposure; 3) monitoring and recording of worker exposure; and 4) limiting
lifetime exposure. We discuss each of these in turn.
A. Education of Workers
Workers have been told more about the dangers of radiation than about
many other occupational hazards. However, most of them do not know the
most recent quantitative estimates of radiation risks, or what they are
based on. They should be told. The discussion and numerical evaluations
of risks in this report are examples of what is appropriate for this
purpose. It is clearly not acceptable to inform a worker of the dose
limits and leave the impression that doses below these limits are "safe"
or "negligible." Workers must understand that most risks from radiation
are assumed to be proportional to the dose and understand the size of
85
-------
their risks. Since risks to the unborn are greatest from exposures in
utero, female workers and those who supervise them should be specifically
informed about risks to the unborn. Up-to-date knowledge of radiation
risks should provide a significant incentive in any program for reducing
doses to workers. As a corollary, workers should have uninhibited access
to records of their exposure so that they can assess their own risks.
Education on radiation protection requirements and practices must be
tailored to the needs of different kinds of work and workers - for
example, welders in nuclear facilities and dental technicians have
obviously different protection needs; and female workers and their
supervisors in any kind of work should be well-informed regarding
>
protection measures to reduce exposure of the unborn. As a starting
point, all workers should be fully informed of the basic radiation
protection principles and guides set forth in Federal guidance. Education
of workers is basic to effective radiation protection and is therefore a
minimum requirement for all three ranges of exposure.
B. Radiation Protection Supervision
Supervising radiation protection means assisting and guiding managers
in deciding whether exposures of workers are justified and radiation
protection is optimized (ALARA), as well as supervising day-to-day
protection of workers. We have distinguished three levels of supervision,
depending on the dose.
In Range A, which extends up to one-tenth of the RPGs, the number of
workers is large (95% of the work force) and doses to individuals are
small. However, the collective dose is larger than for either of the
other ranges - almost half that of the entire work force. Clearly, it
86
-------
would be impractical and unreasonable to provide professional radiation
protection supervisors for this large number of workers. However, because
of the large collective dose, careful generic assessments of the
justification of exposure and of the optimization of radiation protection
measures and practices should be carried out whenever practical. These
assessments should influence, for example, design of facilities, such as
diagnostic x-ray installations; regulation of the packaging and handling
of radioactive materials for transportation; regulation of the design of
electronic products, such as diagnostic x-ray machines; and specification
of minimum training or licensing requirements and work practices for the
use of radiation equipment and radioactive materials.
In Range B, which encompasses intermediate doses above one tenth but
below three tenths of the maximum allowable dose levels, professional
radiation protection services should be available in the work place. At
these dose levels, which involve less than 5% of all workers, the risks to
individual workers are large enough so that on-the-job radiation
protection supervision is usually justified. Furthermore, workers in this
dose range are usually involved in a wide variety of specialized work
situations that are not amenable to generic treatment for radiation
protection.
We recognize that in some work places the numbers of workers may be
so small that provision of professional radiation protection services
could be burdensome, so that some flexibility will be needed in applying
this requirement. Such supervision may, in a few casesj have to be
provided on a part-time consulting basis, or a few workers may have to
acquire professional radiation protection training.
However, in the vast majority of hospital, industrial, and laboratory
situations such professional protection services should be available on a
87
-------
full-time basis. This is essential to provide the detailed attention to
radiation protection - including justifying exposure and optimizing
protection - that is required to insure that exposure of workers is
minimized in this dose range. It is also essential that supervisors have
the authority and access to management required to carry out these
functions effectively.
The highest dose range, Range C, which extends upward from three-
tenths of to the full maximum allowed dose, involves less than two percent
of all workers. However, it is these workers who are theoretically able
to accumulate lifetime doses large enough to pose substantial risks to
themselves and their descendants. These workers also tend to work in
situations involving high dose rates and a high potential for accidental
overexposures, so that vigilant care is needed. As in Range B these
exposures should be properly justified and radiation protection
optimized. Beyond this, for those tasks which may make a substantial
contribution to doses in this range, supervision of radiation protection
should be provided on a task-by-task basis - both before and during the
work. This does not mean that radiation protection personnel should
necessarily be located in high exposure areas during the work - that would
not usually keep collective doses ALARA - but that they should maintain
effective control over individual exposures of workers.
C. Monitoring and Record Keeping
An important element of control of occupational exposures is adequate
monitoring and maintenance of records. In Range A, monitoring of the work
place and, as appropriate, monitoring of individual workers should be
88
-------
carried out to the extent necessary to assure that doses are ALARA and are
within the range. In many cases monitoring of all workers in Range A work
situations will not be necessary.
All workers who may receive doses exceeding one-tenth of the RPGs
(that is, doses in Ranges B and C) should be individually monitored and
their doses recorded. Although we have not included a requirement for
maintenance of lifetime records for all Range B exposures, this practice
is strongly encouraged when it is feasible. In Range C, monitoring
results should also be recorded for individual high-dose tasks, as an aid
to maintaining doses ALARA and to provide a basis for review of these work
situations.
D. Lifetime Dose
As discussed below in Chapter VI, in order to achieve the objective
of limiting maximum lifetime risks to a value comparable to average risks
from other occupational hazards, a two- to three-fold reduction of the
maximum lifetime dose permitted by an RP6 of five rems per year is
required. This could be accomplished either by lowering the annual RPGs
or limiting the total lifetime dose.
The first approach has the advantage of simplicity. However, in
order to achieve a significant lowering of potential lifetime risk it
would be necessary to reduce the present average limit of five rem per
year to a significantly lower value. If such a reduction occurred, it
appears that some beneficial activities would be prohibited, that a
significant increase in collective dose would occur, or that unreasonable
costs would be incurred in certain subcategories of the work force
(At80,Do79,HA80).
89
-------
The second approach could be achieved through maintaining lifetime
dose records for all workers, or by keeping only those records required a)
to limit the number of years the annual dose of a worker may exceed some
specified value (significantly lower than the RPGs) or b) limit the
lifetime sum of annual doses which exceed the same value.
The maintenance of lifetime dose records for the entire work force
would be a major undertaking. A requirement of this magnitude does not
appear to be reasonable to protect the very small fraction of the work
force that may receive large lifetime doses, if more reasonable approaches
are available. Further, very few of the already small fraction of workers
receiving annual doses of a few rem or more can be expected to continue at
such dose rates for a working lifetime.
The remaining alternatives avoid the above disadvantages for the
small penalty of not counting annual doses that are less than some
relatively small fraction of the maximum annual dose. In view of
limitations on the accuracy of dosimetry, as well as uncertainties in risk
estimates, we do not believe this penalty is significant. The differences
between alternatives a) and b) include some possible administrative
simplicity for the former and some increased accuracy (and possible
usefulness for epidemiological studies) for the latter.
We have recommended that once a worker incurs a dose in Range C all
subsequent yearly doses in both Ranges B and C be kept in a lifetime
record and that the accumulation of doses by individual workers be managed
so as to avoid allowing this accumulated lifetime dose to exceed 100 rem.
This is roughly equivalent to a lifetime limit on average dose of two rems
per year. It would reduce maximum lifetime risk of death from radiation
exposure to a level at worst comparable with average risk of accidental
90
-------
death in the major occupational categories of agriculture, construction,
and mining.
In the case of older workers, most somatic risks associated with each
increment of dose are probably less than those for younger workers (see
Figure 13) and genetic risks are usually no longer present. However,
because of the highly individual nature of these considerations and
because we do not know age-specific cancer risks well enough, we have made
no age-specific recommendations.
As a general rule, workers who have accumulated an occupational dose
in excess of 100 rem under the 1960 guidance should not incur Range C
exposures. They should be assigned to duties for which the annual
exposure is in Range A. This new guidance, however, should be introduced
with discretion, taking into consideration the economic well-being and the
preference of the individuals concerned. According to currently accepted
radiation-risk models, the risk associated with the dose received in any
year is in addition to and independent of the risk from previously
received doses. The regulator, employer, and the worker should evaluate
the potential incremental radiation risk in relation to available
alternatives.
91
-------
VI. RADIATION PROTECTION GUIDES FOR MAXIMUM ALLOWED DOSES
The 1960 radiation protection guides for limiting occupational
whole-body and gonadal annual dose are 3 rem in 13 weeks and an
accumulated dose of 5 rem times the number of years beyond age 18 (that
is, 5(N-18) rem, where N is the worker's age in years). Two annual limits
may be inferred from these guides: (1) a maximum dose of 12 rem in any
one year; and (2) a maximum average annual dose of 5 rem over an entire
lifetime, starting from age 18.
We estimate the harm associated with recent exposure experienced
under these guides below, first for lethal and nonlethal cancers, next for
effects on the unborn, and finally for a variety of less serious risks.
Where possible, comparisons to comparable occupational hazards are made.
This leads to our conclusions for the RPGs proposed in Recommendation 3.
A. Cancer Risks From Whole-Body Exposure
1. Fatal Radiation-Induced Cancer
a. Lifetime Risks
Estimated lifetime risks of death from radiation-induced
cancer were shown in Figure 8 (see Chapter III) for uniform annual doses
of up to 5 rem per year over a working lifetime. Table 5 showed the
average levels of risk estimated for the entire radiation work force and
for its major components in 1975. The maximum lifetime risk of death from
radiation-induced cancer allowed under the 1960 guide was estimated to
fall between 3 and 6 in a hundred.
92
-------
As an aid to placing these lifetime risks in perspective we have
compared them to the risks of death from on-the-job accidents currently
encountered in various industries in the United States. A comparison to
risk of death from other carcinogenic agents in occupational environments
would also be relevant, but adequate data for such a comparison are not
available. In any case, comparison of radiation risk to risk of
accidental death alone is conservative, since other carcinogenic risks
would increase the total risk of death from causes other than radiation.
We have omitted radiation risks to workers from normal background
radiation, from medical exposures, and from diagnostic x rays that are
required as a condition of employment.
Table 10 lists the average annual risk of death from on-the-job
accidents in various broad groups of occupations (NS73-75). Within any of
these groups of occupations, individuals in different jobs obviously face
different risks, varying from much less than the average value to values
which are several times higher than the risk to the average worker.
Numerical estimates are not available for the distribution of these risks
by specific job assignment. Consequently, we could calculate only the
average lifetime occupational risk. This is defined here as the average
lifetime probability of death from an on-the-job accident faced by an
18-year-old about to enter employment in an occupation in which he or she
will be exposed to its average risk annually until age 65, unless death
occurs earlier.
A premature cancer death attributed to radiation is not equivalent,
in a number of respects, to a premature accidental death. For example,
93
-------
Table 10. Annual Risk of Accidental Death in U.S. Industries (NS73-75)
Deaths per 100,000 Workers
Indus try/Year 1971 1972 1973 1974 1975 1976 1977
Trade
Manufacturing
Service Industries
Government
Transportation and
Public Utilities
Agriculture
Construction
Mining, Quarrying
7
10
12
13
36
66
71
100
7
9
10
13
36
61
70
117
7
8
10
13
35
61
71
117
6
8
10
13
34
54
63
71
6
8
9
12
33
58
61
63
6
9
9
11
31
54
57
63
6
9
8
11
33
53
60
63
All Industry Average 18 17 17 15 15 14 14
94
-------
the estimated average number of years of life lost is 12 to 18 years for a
cancer death due to radiation, whereas it is approximately 35 years for
accidental deaths, under the above assumptions (Bu80). A more in-depth
analysis would undoubtedly reveal additional differences, e.g., hospital
costs, suffering, or impact on others, that could be greater or less than
in the radiation case. Because we lack information on such other factors,
the following comparisons were made on the basis of frequency of incidence
(risk) and reduction in life expectancy only. In order to make these
comparisons, annual accident rates were converted to lifetime risks and
loss of life expectancy using a life table analysis (Co80).
Lifetime risks from radiation exposure are compared to lifetime risks
of accidental death in major U.S. industries in Figure 11. As shown in
the figure, the risk associated with continuous exposure over a working
lifetime to the average dose to the 1975 radiation workforce (0.12 rem) is
lower than the average lifetime risk of death due to accidents in retail
and wholesale trades, the safest occupational group. The range of
lifetime fatal cancer risk to the radiation workers with the highest
average annual dose (0.92 rem for nuclear waste disposal workers with
measurable doses) brackets the average accident risk for all occupations.
Although data are not available for a comparison of maximum risk of
cancer death from radiation with maximum risk of death from accidents, a
comparison of maximum allowed radiation risks under the 1960 guide with
average accident risk is possible and provides some insight, since it
represents the upper bound to the permissible range of radiation risk.
As shown in Figure 11, the maximum currently allowed lifetime risk of
lethal cancer from radiation ranges from equal to about two and a half
95
-------
.07
.06
.05
RADIATION RISK
(Relative Risk Model),
.04
K
3
•X.
u>
in
111
in
0
.03
.02
.01
RADIATION RISK
(Absolute Risk Model)
AVERAGE 1975 DOSE
TO ALL WORKERS
RISK OF ACCIDENTAL
DEATH BY INDUSTRY
1 2 3
ANNUAL DOSE EQUIVALENT (REM)
Figure 11. Lifetime risk of death due to radiogenic cancer by annual dose
level for two risk models compared to average occupational risks
of accidental death. It is assumed that the dose level to
radiation workers and accidental death rates of workers /in other
industries remain constant from age 18 to 65. The average exposure
of various groups of radiation workers is given in Table 2.
96
-------
times the average risk of death due to accidents in mining and quarrying,
construction, or agriculture, the three highest risk industries listed.
These comparisons are in terms of the number of premature deaths.
Loss of life expectancy due to premature death may also be used for
comparison. As noted above, the average number of years of life lost for
a radiation-induced cancer death is only one-half to one-third that for a
job-related accidental death. On the other hand, the effects on others
that are associated with premature loss of life of a worker are not
related in any unique or simple way to the number of years of life lost.
We therefore do not make any judgment on the relative merit of comparisons
based on chance of premature death versus those based on loss of life
expectancy, but present both.
Estimated losses of life expectancy from exposure of radiation
workers and from accidental deaths of workers in other industries are
shown in Figure 12. Radiation workers in all job categories are estimated
to experience a smaller average loss of life expectancy than that due to
accidental death for the average U.S. worker. Moreover, even though an
individual receiving a maximum allowable lifetime whole-body dose of 5 rem
per year from age 18 to 65 is subject to a loss of life expectancy which
exceeds the average due to accidental death for all workers, this maximum
loss is still smaller than the average loss of life expectancy for workers
in the three highest risk occupations listed (mining and quarrying,
construction, and agriculture).
We draw two conclusions from the above observations. First, based on
experience for the past 15 years, the risk of death from radiation-induced
cancer for the average worker is low in comparison with risks of
97
-------
AVERAGE LOSS OF LIFE
EXPECTANCY FROM
ACCIDENTAL DEATH BY
INDUSTRY
CO
I
o
LLJ
a.
X
LU
HI
O
a
DC
UJ
RADIATION RISK
(Relative Risk Model)
AVERAGE 1975 DOSE
TO ALL WORKERS
RADIATION RISK
(Absolute Risk Model)
4 ~
2 r_ -V—
1 2 3
ANNUAL DOSE EQUIVALENT (REM)
Figure 12. Average reduction in life expectancy due to radiogenic cancer
by annual dose level for two models compared to average
occupational reduction in life expectancy. It is assumed that
the dose level to radiation workers and accident death rates to
workers in other industries remain constant from age 18 to 65.
The average exposure of various groups of workers is given in
Table 2.
98
-------
accidental death in other occupations. For this reason we do not find it
necessary or justified to lower the whole-body Radiation Protection Guide
below 5 rem to provide greater protection from radiation-induced fatal
cancer to the work force, taken as a whole. However, a worker who
received the maximum allowed annual dose every year throughout a working
lifetime could accumulate a lifetime risk higher than that of average
workers in the three highest risk major occupational categories not
normally exposed to radiation - mining and quarrying, construction, and
agriculture. We believe that lifetime doses to radiation workers can
normally be maintained at risks that are below the average for these three
high risk occupations. This would be accomplished by maintaining lifetime
doses at less than 100 rem, as proposed under Recommendation 4.
b. Age Dependence of Risk and the 3 Rem Quarterly Guide
The 1960 radiation protection guidance for the whole-body is
that the accumulated dose to a worker not exceed five times the number of
years beyond age 18; that is, 5(N-18) rem, where N is the worker's age in
years. Since the only limitation on the rate of dose accumulation is the
guide specifying a maximum of 3 rem in 13 weeks, a worker may receive as
much as 12 rem in any one year if he does not exceed the total specified
by the 5(N-18) guide. The implications of lifetime accumulation of the
maximum dose permitted under the 5(N-18) guide were discussed above.
The risk associated with the flexibility in the guides permitting
maximum doses in any one year of 12 rem depends on the individual's age at
exposure. The age dependence of the risk of cancer death is shown in
Figure 13 for a single dose of 12 rem, for the two risk models we use.
99
-------
0.004
r
cc
UJ
o
z
0.003
O
UJ
O
\
9
cc
p
iu
D
Q
0.002
UJ
O
\
S2 0.001
10
ABSOLUTE RISK MODEL
RELATIVE RISK MODEL
20
30
40
50
60
70
AGE AT EXPOSURE
Figure 13. The risk of death from radiogenic cancer due to a single
dose of 12 rems, versus age at exposure. It is assumed that
the latency period is independent of age.
100
-------
The risk declines with increasing age, but, especially for the relative
risk model, maintains a high value throughout most of a normal working
lifetime. In addition to the risks of cancer death shown in Figure 13,
there are also substantial risks to the unborn (both genetic and to the
unborn child) from doses to parents of 12 rem (see Chapter III).
Allowing doses up to 12 rem in any year permits multiple exposures in
any year of certain workers whose skills are in short supply. It does not
permit single 12 rem exposures, since the existing quarterly guide does
not permit doses greater than 3 rem for any single work operation. Thus,
there are no single jobs now performed that require doses from 3 to
12 rem. Annual exposures at this level can be avoided by training
additional workers. Major segments of the U.S. work force have operated
under a 5 rem annual limit for a number of years (Do79), and international
guidance has not contained a 3 rem/quarter limit since 1972. We conclude
that this flexibility should be discontinued, since the risks to
individuals are not sufficiently warranted by demonstrated need.
2. Nonfatal Radiation-Induced Cancers
We assume that the risk of incurring nonfatal cancer is, at most,
equal to that of incurring fatal cancer (see p. 40). For perspective, we
compared this risk to job-related nonfatal injuries and illnesses in
various industries and occupations in the United States. Nonfatal cancers
are different from other types of injuries and illnesses; there is no
completely satisfactory way to compare all types of nonfatal injury.
Nevertheless, some useful insight may be gained from a simple comparison
of time lost over a lifetime due to these causes. As before, statistics
101
-------
for the harm not involving radiation are only available for average
workers, and comparisons with maximum allowable risk of harm from
radiation must be made with care.
Table 11 displays reported statistics for annual incidence rates and
our estimates of average lost time over a working lifetime for nonfatal
occupational injuries and illnesses in the U.S. private sector in 1976.
The private sector includes all but government workers. Nonfatal
occupational injuries include all those requiring medical treatment other
than first aid; occupational illnesses are defined as those associated
with exposure to environmental factors in the workplace. Statistics for
the latter include all identified acute and chronic illnesses possibly
caused by contact with, or inhalation, absorption, or ingestion of such
factors (Bu78).
In an average population of workers, the expected number of years of
working lifetime from age 18 to age 65 is 43.7 years per worker, not
47 years, since some workers will die before reaching age 65. The annual
incidence rates in Table 11 were converted to lifetime values by assuming
that they remain constant over a working lifetime, using the above average
expectation for length of a working lifetime, and by introducing a factor
to convert working days lost to total days of lost lifetime. The
resulting values are shown in the final column.
A recent study of U.S. experience for cancer morbidity indicates that
the average lost time per diagnosed case is 1.8 months (Na79). Applying
this value to the maximum lifetime risk of such cancers from radiation
exposure developed in Chapter III, the average lost time over a lifetime
is estimated to be less than 0.01 years for the case of a worker exposed
102
-------
Table 11. Nonfatal Injuries and Illness in U.S. Industries (Bu78)
Average Annual Rate/Worker
Industry
Agriculture*, forestry
and fishing
Mining
Construction
Manufacturing
Transportation and
public utilities
Wholesale and
retail trades
Finance , insurance ,
and real estate
Service Industries
Entire Private Sector
Number
of Workers
(thousands)
1,000
781
3,564
18,883
4,528
17,628
4,149
14,158
64,960
Total
Cases
.110
.109
.153
.132
.098
.075
.020
.053
.092
Lost
Work-
Day
Cases**
.047
.058
.055
.048
.050
.028
.007
.020
.035
Lost
Work
Days
.833
1.144
1.050
.795
.940
.432
.116
.384
.605
Working
Lifetime
Lost
(Years)
.146
.200
.184
.139
.164
.076
.020
.067
.106
Excludes farms with fewer than 11 employees.
** The number of cases which result in loss of at least one work day.
103
-------
for his entire working lifetime to the maximum annual dose under the 1960
guide. This lost time per lifetime is an order of magnitude smaller than
that estimated for the average of nonradiation causes in the private
sector, as shown by Table 11.
On the basis of this relatively small effect, coupled with the
judgment that nonfatal cancer is ordinarily less severe in its impact than
fatal cancer, we conclude that the protection provided against risk of
fatal cancers is sufficient also for protection against nonfatal cancers.
B. Health Risks to the Unborn
1. Mutational Effects
The current guides for limiting dose to the gonads are identical
to those for the whole body. For a given annual dose, the risk of serious
mutational effects in all of a male worker's descendants combined is be-
lieved to be numerically comparable to his lifetime risk of fatal cancer
(cf Figures 8 and 10 in Chapter III, Section C). The medical severity of
these hereditary defects is usually less than, and, at worst, comparable
to death from cancer. The largest risk to any single generation, that to
first generation children, is about one sixth that to all generations
combined. For these reasons we do not believe that a more restrictive
guide is required for the male gonads than for the whole body. An
argument could be made for increasing the guide for female gonads, since
the sensitivity is much lower than that of male gonads. However, this
would be meaningless; it is unlikely that a woman could receive a higher
gonadal dose without exceeding the limit for whole-body dose.
Gonads include both testes and ovaries,
104
-------
The proposed guide for gonadal dose is therefore identical to that
proposed for the whole body, as was the previous guide. Unlike the ICRP,
we specify this guide separately and do not include gonads in the scheme
proposed below for partial-body exposures, because the risks involved are
of a different nature: the affected individual is not the one exposed to
radiation and the effects include different types of harm.
2. Risks Due to In Utero Exposure
Protection of those not yet born is an already well-
established principle of radiation protection; the purpose of the
guide for gonadal exposure discussed above is to limit mutational
effects in children conceived after the exposure. However, those
already conceived but not yet born, the "unborn," are also at risk.
Their risks are greater, for a given dose, than risks to those not yet
conceived. Current guidance does not contain a dose limitation to
protect the unborn from somatic effects, although such a limit has
been recommended by NCRP for a number of years (NC75, NC77).
The risk of serious harm following in utero exposure demands
careful attention because of the magnitude and diversity of the
effects, because they occur so early in life, and because those who
suffer the harm are involuntarily exposed. These risks are not as
well quantified as those to adults, but available evidence indicates
that at critical periods in the development of the unborn, for the
same dose, the risks may be many times greater than those to adults.
There are several factors which mitigate this situation. First,
the exposure of most workers under annual limits is relatively evenly
distributed over the year, so that only a quarter of a worker's annual
dose is delivered to the unborn during any trimester. Second, the
105
-------
mother's body provides considerable shielding of the unborn for most
types of exposure. For example, shielding factors for the degraded
spectrum of x rays from 50 kev and 1000 kev photon sources are 0.12
and 0.55, respectively (Di7A). Finally, the total period of potential
exposure is small for the unborn compared to that for a worker - a
period of months compared to a working lifetime.
It is difficult to provide protection of the unborn that is
equivalent to that provided adult without affecting the rights of
women to equal job opportunities. This difficulty is compounded
because the critical period for most harm to the unborn occurs soon
after conception - during the second and third month after conception,
when a woman may not know that she is pregnant. It is therefore
essential that women be properly informed of the risks to the unborn
from radiation. In addition, employers should assess their practice
of ALARA in the light of these risks, when exposure of female
employees is possible. Finally, in keeping with basic premises of the
Occupational Health and Safety and the Civil Rights Acts, employers
should assure that protection of the unborn is achieved without loss
of job security or economic penalty to workers.
Based on our assessments of risks described in Chapter III
(section C.3) and the other factors noted above, we believe that total
dose to the unborn should be maintained a factor of ten below the
maximum permitted adult workers in any year. In Recommendation 8 we
propose four alternatives which would, with varying degrees of
certainty, achieve this objective. Each involves a compromise of one
kind or another:
a. Women are encouraged to voluntarily keep total dose to
106
-------
any unborn less than 0.5 rem during any known or suspected
pregnancy.
This alternative relies upon voluntary compliance. It assumes a
woman knows she is pregnant within six weeks of conception, and will,
along with her employer, take appropriate protective action. It
therefore does not guarantee that doses to the unborn during the
critical early stages of pregnancy will be less than 0.5 rem. Equal
job opportunities for women are not directly affected by this
alternative.
b. Women able to bear children are encouraged to
voluntarily avoid job situations involving whole-body dose
rates greater than 0.2 rem per month, and to keep total dose
to the unborn less than 0.5 rem during any known pregnancy.
This alternative adds a voluntary limit on dose rate to woman who
can bear children so as to protect the unborn whose existence is not
yet known. It permits women to hold any job, but encourages women
able to bear children not to take those few jobs which potentially
involve high dose rates. It would provide voluntary protection of the
unborn, during the critical early stages of pregnancy, in addition to
voluntarily limiting the total dose to the unborn.
c. Women able to bear children should be limited to job
situations involving whole-body dose rates less than 0.2 rem
per month. Total dose to the unborn during any known period
of pregnancy should be limited to 0.5 rem.
The third alternative assures protection of all unborn throughout
gestation by making the voluntary requirements of the second manda-
107
-------
tory. It would bar women of child-bearing capacity from those few
jobs which involve high dose rates.
d. The whole-body dose to both male and female workers
should not exceed 0.5 rem during any six month period.
The final alternative would restrict the exposure of all workers,
male and female, to a level which would protect the unborn at almost
the level of alternative c. It would still subject the unborn to much
greater risk of harm than a worker could incur in the same exposure
period. This alternative preserves equal job opportunity for women at
the cost of causing more total harm to the work force. Studies of
several high exposure activities show that decreasing the dose limits
to this extent would significantly increase the collective dose to
workers, and that some current activities would not be possible
(At80,Do79,HA80). Alternatively, society could avoid this increased
risk by foregoing some high exposure activities, which can be expected
to occur principally in the six job categories identified in Table 2
(Chapter II) that exceed 0.5 rems average dose per year.
None of these alternatives is completely satisfactory; they each
involve either varying degrees of adequacy of protection of the
unborn, some sacrifice of equal job opportunity for women, or causing
more total harm, or foregoing some of the benefits to society from
activities using radiation. We invite public comment on the relative
importance to be attached to each of these factors in formulating
guidance, and on whether or not the guidance should address this
matter now. We would also be happy to receive suggestions for other
alternatives.
108
-------
C. Health Risks from Partial Body Exposure
1. Cancer Risks to Organs and Tissues
The list of specific parts of the body for which guides are
required has evolved over the years as knowledge of radiation effects has
increased. We have reviewed previous choices in the light of current
information, and the recommendations contain both additions and deletions.
We have added breast and lung to the list of specific organs
considered in the 1960 guidance, since these are two of the principal
contributors to the risk of cancer death from radiation. Forearms, feet,
and ankles are now covered by the guides for skin and whole body.
Finally, the parts of the body formerly designated as "blood-forming
organs," "head and trunk," and "bone" are now covered as "red bone
marrow," "whole body," and "bone surfaces," respectively, in keeping with
current ICRP views on appropriate nomenclature (IP77).
Exposure of portions of the body can occur through localized irradia-
tion of extremities (such as hands in glove boxes), or by breathing or
swallowing radioactive materials which then migrate to different organs in
the body. The current guidance limits such exposure through separate
numerical guides for individual parts of the body. However, it does not
consider the sum of the risks of cancer when more than one organ is
irradiated. We propose to take the total risk of cancer death into
account. This is done by first assigning a weight to the dose to each
organ equal to the risk from a dose to that organ divided by the risk from
the same dose to the whole body. We then limit the sum of these weighted
doses. This scheme is similar to that recently adopted by the ICRP
(IP77,IP78). These weights are listed in Table 6 (Chapter III).
109
-------
We used three criteria to choose numerical guidance to limit exposure
of organs or parts of the body: 1) the lifetime risk from exposure should
not exceed that for the whole body, 2) any threshold for non-stochastic
effects should not be exceeded in a working lifetime, and 3) no guide
should be established at a value higher than experience shows is needed.
Proposed Recommendation 3, part b, provides that the sum of the
weighted annual dose equivalents to all organs should not exceed 5 rem,
the guide for exposure of the whole body. This provision, however, only
limits the risk of cancer death and is not sufficient in itself to prevent
large doses to a single organ in which other effects, such as non-lethal
cancers and non-stochastic effects, may cause harm. A supplementary
annual limit of 30 rem to any single organ provides an ample margin of
safety for these other effects and we propose it as an independent
criterion.
We have chosen the limiting annual dose to any single organ to be
30 rem, rather than the internationally-adopted value of 50 rem, because
we do not see a need for adopting a value higher than any now in use in
this country. The risk associated with a 30 rem dose to any of the organs
is equal or less than that of a 5 rem dose to the whole body. Additional
differences from internationally-used values for gonads are discussed
above under the heading "Mutational Effects," and for lens of eye and
hands below under the heading "Other Risks."
It is usually impractical to directly monitor the dose received by a
worker who breathes or swallows radioactive materials, but it is useful to
be able to predict doses that may be received from breathing contaminated
atmospheres or swallowing contaminated materials. To make decisions about
radiation protection of such workers possible it is necessary to calculate
110
-------
for different kinds of radioactive materials the amount which gives the
maximum annual dose allowed by the RPGs. These calculations require
complex models of metabolism and dosimetry. We propose in
Recommendation 5 that these amounts of radioactivity be designated the
"Radioactivity Intake Factors" (RIFs), and that they replace the currently
used "Radioactivity Concentration Guides."
Note 3 to the recommendations specifies the appropriate models for
use in calculating the RIFs. Recent advances in modeling of metabolism
and for dosimetry have resulted in significant changes in the doses
calculated for radioactive materials in the body (IP75,IP79). For most
radioactive materials the changes in the calculated doses due to changes
in the models are considerably larger than the changes in the proposed new
RPGs (Fo79). These new models usually, but not always, reduce allowable
intakes. A summary of the changes due to the new models and to the
proposed new guides is provided for the more significant radionuclides in
the Appendix.
2. Other Risks (Eyes and Skin)
The guidance recommends that, whenever reasonable, the lifetime
dose to any worker be less than 100 rem, a total dose at which no harmful
non-stochastic effects are expected to occur if the whole-body dose in any
one year is 5 rem or less. Threshold doses for non-stochastic effects are
not well known at such low dose rates, but it is likely that these values
are well below the dose at which recognizable damage would occur.
Nevertheless, all workers are unlikely to have the same sensitivity and we
do not believe these limiting doses should be increased since no need for
higher limits has been established.
Ill
-------
The ICRF has very recently decreased its recommendation on- the
limiting annual dose to the eye from 30 to 15 rem (IP80). While adequate
protection against cataracts of the lens of the eye might be provided by a
higher maximum average annual dose than the 5 rem now allowed by U.S.
guidance, no operational difficulty is reported with use of 5 rem as an
annual limit (Ch79). That value is therefore retained in these proposals.
The maximum annual dose for skin of the whole body is maintained at
30 rem, since a need for allowing higher doses has not been demonstrated.
However, the current guide permits 75 rem annual doses to hands and
forearms, or feet and ankles, because of the assumed lower risk when only
these portions of the skin and underlying tissue of these extremities are
involved. We agree that at low dose rates the risk depends in some degree
on the amount of skin and tissue exposed, and that exposure of the
extremities is therefore less dangerous than of the whole body. However,
for forearms, feet, or ankles such a high value is not needed and we
propose that the annual guides for skin and the whole body apply to these
extremities. For the hands a higher value appears to be justified for
work in glove boxes. We propose 50 rem, the limit recommended by the
ICRP.
112
-------
VII. SPECIAL EXPOSURE SITUATIONS
Previous chapters have addressed exposure of adults under normal
conditions of exposure. We address some exceptions below. These include
emergency and accidental exposures, exposure of workers from the
activities of others, exposures for medical purposes (both those that are
job-related and those that are not) and other non-occupational exposures,
exposure of minors, and exposure of underground miners to radon decay
products. There may be special circumstances other than emergencies for
which exposures above the RPGs are justified. In addition, exposure
limits may be required for periods other than one year, the period to
which the RPGs apply, or for situations in which internal and external
exposures are combined. We address each of these special exposure
situations in turn.
Emergency situations are, almost by definition, unique. In Note 4 to
these general recommendations we choose not to provide numerical guides
because of the great variability in the circumstances which may surround
emergencies. Only broad principles can be relied upon to provide useful
general guidance. These are provided by Recommendations 1 and 2.
Additional guidance is also provided by Recommendations 7, 8, and 9 that
may be applicable to some emergency situations. We have also published
specific informal guidance for personnel involved in emergency actions in
the early phase of accidents at nuclear facilities when the airborne plume
is the principal radiation exposure pathway (En75). This guidance is
under review for eventual incorporation into Federal guidance.
113
-------
Accidental exposures may be high enough in some cases to require
medical treatment. This guidance does not address such matters, which
should be handled by medical personnel competent to deal with the acute
effects of radiation exposure. We have not addressed the issue of whether
overdoses in one year should lead to additional restrictions on doses in
future years, including the management of lifetime dose. Such situations
must be dealt with on the merits of each case and under the regulatory
mandate of the responsible Federal agency (Note 5 to the
Recommendations). We do not consider it either practical or reasonable to
prejudge or prescribe general conditions for such situations beyond the
general principles which apply to all radiation exposure contained in
Recommendations 1 and 2.
In some situations workers are exposed to radiation from sources in
locations not under the control of their employer, or due to contamination
from previous use of the premises. In the former case these workers need
not be considered occupationally exposed, since existing laws require the
owners of such sources to maintain doses in areas outside their control to
levels acceptable for the general public. In the latter case workers are
subject to regulations governing occupational exposure established under
this guidance.
The question often arises whether or not exposure for medical
purposes and other non-occupational exposures should be considered in
calculating the doses that workers receive within the guides. If there
were a threshold for risk of health effects from radiation this could be
an important consideration. However, since
we assume that the risk at low doses is proportional to the dose, each
exposure must be justified on its individual merits. For this reason,
114
-------
in Note 1 to the recommendations we exclude medical and other non-
occupational exposure from the assessment of occupational radiation
exposure of workers.
In many jobs diagnostic x-ray examinations are a routine part of
periodic or pre-employment physical examinations. Some of these
examinations are a condition of employment and some are not. Federal
radiation protection guidance on use of diagnostic x rays was issued by
the President on February 1, 1978 (En78). These recommendations provide
that, in general, use of such x-ray examinations should be avoided unless
a medical benefit will result to a worker, considering the importance of
the x-ray examination in preventing and diagnosing diseases, the risk from
radiation, and the cost. Although all of the recommendations in that
guidance may be usefully applied to x-ray examinations of workers,
Recommendations 1 through 4 are particularly pertinant. Because this
matter has been addressed by separate Federal guidance, exposure from such
diagnostic x-ray examinations is not included in this guidance for
occupational exposure.
Current Federal guidance provides that occupational exposure of
minors (those below the age of eighteen) be limited to doses one tenth
those allowed older workers. Since no justification has been advanced or
arises out of improved knowledge of health risks for either lowering or
raising this guidance, in Recommendation 7 we propose no change.
No other general types of exposed workers are singled out for special
protection by these recommendations. However, one special class of
workers - underground uranium miners - is already subject to a separate
Federal Guide (En71)(see Note 7 to the recommendations). That guide
limits their exposure to radioactive decay products of radon gas. The
115
-------
Mine Safety and Health Administration regulates exposure of all under-
ground miners in accordance with this guide. We expect to review the
guide for workers exposed to decay products of radon in the future.
Some situations may justify planned exposures exceeding the guides.
The exposure of U.S. astronauts to doses exceeding the present quarterly
limit is a recent example of such justified exposure (Na70).
Recommendation 9 provides for such situations, but requires that the
responsible Federal agency fully consider the reasons for doing so, prior
to any such exposures when possible, and on the public record when that
would not compromise national security.
The time period for which limits have been set has varied widely,
from a daily basis for the first official limit recommended by the ICRP in
1934 (0.2 Roentgens per day) (IP34) to the current combination of a
quarterly limit and the age-dependent annual limit of the 5(N-18) rule.
In many cases the choice of time period can be considered largely a matter
of administrative convenience, since only for potentially pregnant workers
is there an adequate scientific basis on which to limit dose rate for the
range of doses of interest here. In all but this case the proposed guides
are expressed on an annual basis because this is the simplest choice
available. Note 6 to the Recommendations provides that regulatory
agencies may choose other periods for administrative reasons, if these are
implemented in a manner consistent with the intent of the Guidance.
The proposed Guidance for non-uniform doses to the body, such as from
internal exposure to radionuclides, takes into account the additivity of
risk when different organs of the body are exposed. These exposures may
also take place in the presence of uniform external exposure of the whole
body. In keeping with the principle of limiting the sum of all cancer
116
-------
risks (and consistent with current recommendations of the ICRP), the total
risk should not exceed that allowed for external doses (Recommendation
3c). When non-uniform doses are due to intake of radioactive materials
alone this limitation may be satisfied by following the condition on
combined external whole-body doses and intake of radioactive materials
specified in Note 2 to the recommendations.
117
-------
REFERENCES
At80 Atomic Industrial Forum, Inc., National Environmental Studies
Project. Study of the Effects of Reduced Occupational Radiation
Exposure Limits on the Nuclear Power Industry, AIF/NESP-017,
Washington (January 1980), and; Atomic Industrial Forum, Inc. An
Assessment of Engineering Techniques for Reducing Occupational
Radiation Exposure at Operating Nuclear Power Plants, Washington
t February 1980/.
Ba71 Bateman, J.L. Organs of Special Senses. Part I: Eye and
irradiation, in Pathology of Irradiation, Charles C. Berdjis
M.D., Editor, Williams and Wilkins Co., Baltimore.
Be78 Beebe, G.W., H. Kato and C.E. Land. Mortality Experience of
Atomic-Bomb Survivors, 1950-74, Life Span Study Report No. 8.
RaoiatioinTfecmJesearch Foundation, TR 1-77, National Academy
of Sciences, Washington.
B173 Blot, W. and R. Miller. Mental retardation following in utero
exposure to the atomic bombs of Hiroshima and Nagasaki.
Radiology, 106:617.
Bu78 Bureau of Labor Statistics. Cbartbook on Occupational Injuries
and Illnesses in 1976, Report 535, U.S. Department of Labor,
Washington.
Bu80 Bunger, B. M., R. Cook and K. Barrick. Life table methodology
for evaluating radiation risk, an application based on
occupational exposures. To be published in Health Physics.
Ch79 Charles, *' . and P.J. Lindop. Skin and Eye Irradiations.
Example Some Limitations of International Recommendations in
£adiolog.ical Protection, IAEA-SR-36/6, International Atomic
Energy Agency, Vienna.
C172 Clement, A. W., Jr., Miller, C. R., Minx, R. P., and B. Shleien.
Estimates of Ionizing Radiation Doses in the United States,
1960-2000, ORP CSD-72-1, Office of Radiation Programs, U.S.
Environmental Protection Agency, Washington.
Co78 Cook, J.R., B.M. Bunger, and M.K. Barrick. A Computer Code for
Cohort Analysis of Increased Risks of Death, EPA 520/4-78-012,
Office of Radiation Programs, U.S. Environmental Protection
Agency, Washington.
Di73 Diamond, E.L., H. Schmerler and A.M. Lilienfeld. The
relationship of intra-uterine radiation to subsequent mortality
and development of leukemia in children: a prospective study.
Am. J. Epidem., 97:283.
118
-------
Di74 Dillman, L.T. Absorbed gamma dose rate for immersion in a
semi-infinite radioactive cloud. Health Pays. 27 p.571-580.
Do79 U.S. Department of Energy. Study of Anticipated Impact on DOE
Programs from Proposed Reductions to the External Occupational
Radiation Exposure Limit, DOE/EV-0045, Washington (1979J.
En71 U.S. Environmental Protection Agency. Radiation protection
guidance for Federal agencies: underground mining of uranium
ore. Federal Register 36:12921 (July 9, 1971).
En75 U.S. Environmental Protection Agency. Manual of Protective
Action Guides and Protective Actions for Nuclear Incidents,
EPA-520/1-75-001, September1975 (revised, June 1980;, Washington.
En78 U.S. Environmental Protection Agency. Radiation protection
guidance for Federal agencies for diagnostic x rays. Federal
Register 43:4377 (February 1, 1978).
En80 U.S. Environmental Protection Agency. Occupational Exposure to
Ionizing Radiation in the United States: A Comprehensive Summary
for the Year 1975, EPA 520/4-80-001, Office of Radiation
Programs, Washington.
Ev79 Evans H., K. Buckton, G. Hamilton and A. Carothers.
Radiation-induced chromosone aberrations in nuclear-dockyard
workers. Nature, 277:533.
Fe60 Federal Radiation Council. Radiation Protection Guidance for
Federal Agencies, Federal Register 25:4402 (May 18, 1960); and
Report No. 1, Background Materiaf~for the Development of
Radiation Protection Standards, Staff Report of the Federal
Radiation Council, Washington.
Fo79 Ford, M, and S.B. Watson. Updated Recommendations for
Occupational Exposure to Radonuclides Compared to Previous
Values, presented at Twenty-Fourth Annual Meeting of the Health
Physics Society, Phildelaphia (July 8-13, 1979).
Ge81 General Accounting Office. Problems in Assessing the Cancer
Risks of Low-Level Ionizing Radiation Exposure, EMD-81-1,
Washington.
HA80 Harrison, N.T. The Consequences of a Reduction in the
Administratively Applied Maximum Annual Dose Equivalent Level for
an Individual in a Group of Occupational Exposed Workers.
(NRPB-R98), National Radiological Protection Board, United
Kingdom (February 1980)
Ha69 Hagstrom, R.M., S.R. Glasser, A.B. Brill and R.M. Heyssel.
Long-term effects of radioactive iron administered during human
pregnancy. Am, J. Epidenu, 90:1.
119
-------
He67 Heller, C.G. Effects on the germinal epithelium, in
Radiobiological Factors in Manned Space Flight, Publication
#1487, National Academy of Sciences, Washington.
In79 Interagency Task Force on the Health Effects of Ionizing
Radiation. Report of the Work Group on Science (June 1979).
Office of the Secretary, U.S. Department of Health Education and
Welfare, Washington.
IP34 International Commission on Radiological Protection.
International recommendations for x ray and radium protection.
Radiology, 23:682.
IP73 International Commission on Radiological Protection. Publication
#22, Implications of Commission Recommendations that Doses be
Kept as Lov as Readily Achelvable, Pergamon Press, New York.
IP75 International Commission on Radiological Protection. Publication
#23, Reference Man; Anatomical , Physiological and Metabolic
Characteristics, Pergamon
IP77 International Commission on Radiological Protection. Publication
#26; Recommendations of the International Commission on
Radiological-Protection, Pergamon Press, New York.
IP78 International Commission on Radiological Protection. Publication
#28, Statement from the 1978 Stockholm Meeting of the ICRP,
Pergamon Press, New York.
IP79 International Commission on Radiological Protection. Publication
#30, Limits for Intakes of Radionuclides by Workers, Pergamon
Pr e s s, New York. ""™"™~^"' """""'
IP80 International Commission on Radiological Protection. Statement
from the 1980 Brighton Meeting of the ICRP, to be published,
Pergamon Press, New York.
IU71 International Commission on Radiation Units and Measurements.
Report 19, Radiation Quantities and Units, Washington.
IU73 International Commission on Radiation Units and Measurements.
Supplement to Report 19, Dose Equivalent, Washington.
IU76 International Commission on Radiation Units and Measurements.
Report 25, Conceptual Basis for the Determination of Dose
Equivalent, WashingtonT"^ "~^"""*'
Ka71 Kato, H. Mortality in children exposed to the A-bombs while in
utero. Am. J« Epidenu , 93:435.
Ke78 Kerr, G. Organ Dose Estimate for Japanese Atomic Bomb Survivors,
ORNL 5436, Oak Ridge National Laboratory, Oak Ridge.
120
-------
Ki68 Kinlen, L.J. and E. D. Acheson. Diagnostic irradiation,
congenital malformations and spontaneous abortion. Brit.
Radiol., 41:648. """"""
Mc75 McKusick, V. Meudelian Inheritance in Man; Catalogs of Autosomal
Dominants, Autosomal Recessives, and X-linked Phenotypes, Fourth
Edition, Johns Hopkins University Press, Baltimore.
Me62 Merrian, G.R., Jr. and E.F. Focht. A clinical and experimental
study of the effect of single and divided doses of radiation on
cataract production. Tr. Am. Optlu Soc., 60:35.
Me72 Merrian, G.R., Jr., A. Szechter and E.F. Focht. The effects of
ionizing radiations on the eye. Front. Radiation Tber, One.,
6:346. ——-——
Mi72 Miller, R.W. and W.J. Blot. Small head size following in utero
exposure to atomic radiation. Lancet, 2:784. ~~" """"""""
Mi76 Miller, R.W. and J.J. Mulvihill. Small head size after atomic
irradiation. Teratology, 14:355.
Mo78 Moriyama I. M. and L. Guralnick. Survival Experience of Atomic
Bomb Survivors, Hiroshima and Nagasaki 1951-76. Radiation
Effects Research Foundation, TR 17-78, National Academy of
Sciences, Washington.
NA72 National Academy of Sciences, National Research Council. The
Effects on Populations of Exposure to Low Levels of Ionizing
Radiation. Report of the Advisory Committee on the Biological
Effects of Ionizing Radiations. National Technical Information
Service, P.B. 239 735/AS, Springfield, Virginia.
NA77 National Academy of Sciences, National Research Council.
Considerations of Health Benefit-Cost Analysis for Activities
Involving Ionizing Radiation Exposure and Alternatives. Report
of the Advisory Committee on the Biological Effects of Ionizing
Radiations, U.S. Environmental Protection Agency report
EPA 520/4-77-003, Washington (1977).
NA80 National Academy of Sciences. The Effects on Populations of
Exposure to Low Levels of Ionizing Radiation. Committee on the
Biological Effects of Ionizing Radiations, National Academy
Press, Washington (1980).
Na70 Radiatiou Protection Guides and Constraints for Space Mission and
Vehicle Design Studies Involving Nuclear Systems, Report of the
Radiobiological Advisory Panel of the Committee"on Space
Medicine, Space Science Board, National Academy of Sciences,
National Research Council, Washington (1970).
121
-------
Na73 National Center for Health Statistics. 1970 Vital Statistics of
the United States, 1970, Volume II, Mortality. U.S. Department
orHeaTth, Education and Welfare, Washington.
Na75 National Center for Health Statistics. United States Life
Tables. 1969-71, Volume 1, Number 1. U.S. Department of Health,
Education and Welfare, Washington.
Na79 National Center for Health Statistics, Data from the National
Survey, Series 13-Number 43, The National Nursing Home Survey;
1977 Summary for the United States, U.S. Department of Health,
Education and Welfare, Washington.
NC71 National Council on Radiation Protection and Measurements.
Report 39; Basic Radiation Protection Criteria. Washington.
NC75 National Council on Radiation Protection and Measurements.
Report No. 43; Review of the Current State of Radiation
Protection Philosophy. Washington.
NC77 National Council on Radiation Protection and Measurements.
Report No. 53; Review of NCRP Radiation Dose Limit for Embryo and
Fetus in Occupationally Exposed Women. Washington.
Ne70 The National Environmental Policy Act of 1969, Public Law 91-190,
January 1, 1970.
Ne74 Neel, J., V. H. Kato and W.J. Schull. Mortality in the children
of atomic bomb survivors and controls. Genetics, 76:311.
NS73- Accident Facts. National Safety Council; 1972, 1973, 1974,
75 and 1975 Editions. Washington.
Nu80 U.S. Nuclear Regulatory Commission. Performance Testing of
Personnel Dosimetry Services, NUREG/CR-1064, Washington; and
personal communication, P. Plato.
Op75 Oppenheim, B. E., M.L. Griem and P. Meier. The effects of
diagnostic x-ray exposure on the human fetus: an examination of
the evidence. Radiology, 114:529.
Re70 Reorganization Plan No. 3 of 1970. Federal Register 35:15623
(July 9, 1970). ——
Rw74 Rowley, M., D. Leach, G. Warner and C. Heller. Effects of graded
doses of ionizing radiation on the human testis. Rad. Res.,
59:665.
Un77 United Nations Scientific Committee on the Effects of
Atomic-Radiation. Sources and Effects of Ionizing Radiation,
1977 Report to the General Assembly, publication E.77.IX.I, U.N.
Publications, New York.
122
-------
Up69 Upton, A.C. Radiation Injury Effects, Principles, and
Perspectives, The University of Chicago Press, Chicago.
Wa60 Wadsworth, G.P. and J.G. Bryan. Introduction to Probability and
Random Variables, McGraw-Hill, New York.
WH71 World Health Organization and International Atomic Energy
Agency. Manual on Radiation Haematology, Technical Reports
Series #123, International Atomic Energy Agency, Vienna.
Wo67 Wood, J.W. and K.G. Johnson and Y. Omori. In utero exposure to
the Hiroshima atomic bomb. An evaluation of head size and mental
retardation: twenty years later. Pediatrics, 37:385.
123
-------
APPENDIX A
Non-Linear Dose Responses in Human Populations
Leukemia data from the Life Span Study of Nagasaki survivors is often
cited as an example of a nonlinear dose response in a heterogeneous human
population, and these observations have been generalized to include most
radiogenic cancers from low-LET radiation (Ro74, Ro78). We believe these
data are insufficient to support any broad generalizations, since there
are only five leukemia cases in the Nagasaki Life Span Study in the dose
range between 5 and 100 rad (bone marrow dose) (Be78). As illustrated in
Ge80 and other reports, such a small number of cases has such a large
sampling variability that the observed response is consistent with a
number of possible dose response models, including linear and quadratic.
In this regard, it is of interest to compare the leukemia experience
among those Nagasaki survivors in the Life Span Study, which includes only
23 percent of those exposed at Nagasaki, with that of the larger Nagasaki
Leukemia Registry. This registry contains 23 leukemia cases among those
exposed to between 5 and 100 rad (bone marrow dose) (Be78). The Life Span
Study contains only 5 cases in this dose interval. Since the neutron dose
to bone marrow at Nagasaki was quite low in this dose range (less than 200
mrad), the dose response in both of these samples is mainly due to gamma
(low-LET) radiation. Figure Al (taken from Be78) shows the ratio of
observed-to-expected leukemias in the Nagasaki Leukemia Registry and in
A-l
-------
LU
UJ
LU
LU
LU
LU
3
LU
O.
X
LU
22
20
18
16
14
12
10
8
6
4
2
NAGASAKI
LIFE SPAN STUDY
LEUKEMIA REGISTRY
0 56 112 168 224
BONE MARROW DOSE (yrad)
Figure Al Dose response for leukemia in two samples of
Nagasaki survivors, redrawn from Fig. 13 in
Be78. The Life Span Study results (mortality)
contain 22 excess cases among persons exposed
to more than 5 rads to the bone barrow. The
Leukemia Registry (incidence) contains
86 excess cases among persons exposed to more
than 5 rads to the bone marrow (Be78). In each
sample, the expected number of leukemias is
based on leukemia in low-dose survivors. The
average bone marrow dose for those individuals
is about 2 rads (Life Span Study) and 0.4 rads
(Nagasaki Leukemia Registry)(Be78,Ke78).
A-2
-------
the Life Span Study as a function of dose. The increased frequency of
cancer as a function of dose for the larger tumor registry group looks
quite different from the dose response in the Life Span Study. Both data
sets are consistent with a linear response as well as a number of other
possible relationships. In view of the variation between the larger and
smaller samples, we are not sufficiently convinced by the available Life
Span Study data to assume a reduced cancer response for low-LET radiation.
References
Be78 Beebe, G.W., H. Kato and C.E. Land. Mortality Experience of
Atomic-Bomb Survivors, 1950-74, Life Span Study Report No. 8.
Radiation Effects Research Foundation, TR 1-77, National Academy
of Sciences, Washington.
Ge81 General Accounting Office. Problems in Assessing the Cancer
Risks of Low-Level Ionizing Radiation Exposure, EMD-81-1,
Washington.
Ke78 Kerr, G. Organ Dose Estimate for Japanese Atomic Bomb
Survivors, ORNL 5436, Oak Ridge National Laboratory, Oak Ridge.
Ro74 Rossi, H. and A. Kellerer. The validity of risk estimates of
leukemia incidence based on Japanese data. Rad.-Res., 58:131.
Ro78 Rossi, H.H. and C. W. Mays. Leukemia risk from neutrons.
Health Physics, 34:353.
A-3
-------
APPENDIX B
The Radioactivity Intake Factors
Most occupational doses arise from external radiation and are to the
whole body. However in some circumstances air or water containing
radioactive materials can deliver doses to workers. Usually this occurs
through breathing contaminated air. Occasionally it occurs through just
standing in such air. Doses from contaminated water are extremely rare
and almost invariably occur through accidental swallowing.
Doses from contaminated air or water are governed by where the
radioactive materials go once they enter the body, and by how penetrating
of human tissues their radiations are. Internal radiation usually does
not affect the whole body equally and it is necessary to calculate where
the radioactive materials go in the body and which organs and tissues
their radiations penetrate. This depends, in part, on the chemical form
of the particular radionuclide involved and how it is metabolized.
Over the past several decades our understanding of these processes
has grown, and complex models have now been developed to determine the
doses involved (IP79). These models have changed and have improved
significantly since the current guidance was established in 1960 (IP59).
The results of calculations using these models are usually expressed
in terms of the concentration of radioactivity in air or water that a
"standard" man (IP75) would have to breath, stand in, or drink for an
entire year of work to just meet the RPGs.
B-l
-------
Table Bl shows the results of such calculations for radioactive
substances in air for three different cases (Ec80). The table contains
48 examples encompassing the 26 most commonly encountered radionuclides.
The first case is for the models used when the RPGs were established in
1960. The second shows the values obtained using the improved models now
available, but retaining the 1960 RPGs. Of the 43 examples for which 1960
values exist, 23 are reduced, 5 do not change, and 15 are increased by the
new models. The largest reduction is a factor of 17 (Uranium-234 and
Uranium-235, Class Y), and the largest increase a factor of 7
(Strontium-90, Class D).
The last column shows the results for the proposed new guides, using
the new models. Compared to the 1960 values, 21 are reduced, 6 do not
change, and 16 are increased. The largest reduction is a factor of 14
(Thorium-232, Class Y) and the largest increase is a factor of 17
(Strontium-90, Class D). In general, values for alpha emitters are almost
all reduced, and those for beta and gamma emitters more often go up than
down.
It is clear from a more detailed examination of the results that the
models play a far greater role in determing the values than the choice of
which of these two sets of guides is used. We have chosen the "summation
of risk" approach shown in the last column, because it provides a more
complete and consistent basis for risk limitation than the "critical
organ" approach now in use.
B-2
-------
128
Table Bl. Maximum concentration oF selected radionuclides in air I in
mi llicuries/liter)*
Current Guides
Nuclide/ClassD 1960 Models0.
New Models
Proposed New Guides
d
New Models
P-32
Mn-54
Mn-56
Co-58
Co-60
Sr-89
Sr-90
Zr-95
Nb-95
Mo-99
1-125
1-129
1-131
1-133
Cs-134
Cs-137
Ce-144
D
W
D
W
D
W
W
Y
W
Y
D
Y
D
Y
D
W
Y
W
Y
D
Y
D
D
D
D
D
D
W
Y
7( -8) bone
8( -8) lungs
4( -7) liver
4( -8) lungs
8( -7) LLI
5( -7) LLI
8( -7) LLI
5( -8) lungs
3( -7) LLI
9( -9) lungs
3( -8) bone
4( -8) lungs
3C-10) bone
5( -9) lungs
1( -7) whole body
3( -8) lungs
5( -7) whole body
1( -7) lungs
7( -7) kidney
2( -7) LLI
2( -9) thyroid
9( -9) thyroid
3( -8) thyroid
4( -8) whole body
6( -8) whole body
1( -8) liver
6( -9) lungs
9( -8) red marrow
6( -8) lungs
4( -7) red marrow
3( -7) lungs
4( -6) lungs
3( -6) lungs
2( -7) lungs
1( -7) lungs
5( -8) lungs
5( -9) lungs
1( -7) red marrow
2( -8) lungs
2( -9) red marrow
6(-10) lungs
3( -8) bone surface
9( -8) lungs
4( -8) lungs
3( -7) lungs
2( -7) lungs
9( -7) liver
3( -7) LLI
2( -8) thyroid
2( -9) thyroid
1( -8) thyroid
7( -8) thyroid
4( -8) gonads
6( -8) gonads
7( -9) liver
2( -9) lungs
3( -7)
K -7)
4( -7)
3( -7)
5( -6)
5( -6)
3( -7)
2( -7)
5( -8)
8( -9)
3( -7)
4( -8)
5( -9) bone surface
1( -9) lungs
3( -8) bone surface
K -7)
7( -8)
4( -7)
4( -7)
8( -7)
4( -7)
2( -8) thyroid
2( -9) thyroid
K -8) thyroid
7( -8) thyroid
4( -8) gonads
6( -8) gonads
8( -9)
4( -9) lungs
B-3
-------
Table Bl. (Continued)
Current Guides
Nuclide/Classb 1960 Models0 New Models'1
Ra-226
Th-228
Th-232
U-234
U-235
U-238
Pu-238
Pu-239
Am-241
W
W
Y
W
Y
D
W
Y
D
W
Y
D
W
Y
W
Y
W
Y
W
3C-11)
9(-12)
6(-12)
2(-12)
K-ll)
6(-10)
K-10)
5(-10)
K-10)
7(-ll)
K-10)
2(-12)
3(-ll)
2(-12)
4(-ll)
6(-12)
lungs
bone
lungs
bone
lungs
bone
lungs
kidney
lungs
kidney
lungs
bone
lungs
bone
lungs
bone
K-10)
2(-12)
2(-12)
3(-13)
7(-13)
3(-10)
K-10)
6(-12)
3(-10)
K-10)
6(-12)
4(-10)
K-10)
6(-12)
2(-12)
4(-12)
K-12)
4(-12)
K-12)
lungs
bone
lungs
bone
bone
bone
lungs
lungs
bone
lungs
lungs
bone
lungs
lungs
bone
bone
bone
bone
bone
Proposed New Guides
New Models d
surface
surface
surface
surface
surface
surface
surface
surface
surface
surface
surface
2(-10)
2(-12)
5(-12)
3(-13)
7(-13)
3(-10)
2(— 10)
K-ll)
3(-10)
2(-10)
K-ll)
4(-10)
2(-10)
K-ll)
2(-12)
4(-12)
K-12)
4(-12)
K-12)
bone
bone
bone
bone
lungs
bone
lungs
bone
lungs
bone
bone
bone
bone
bone
surface
surface
surface
surface
surface
surface
surface
surface
surface
surface
surface
a Exposure is assumed to continue for one year at the rate of 40 hours per
week. When an organ is listed it is limiting and determines the value
shown. If no organ is listed the value is determined by the sum of risk
to all organs. Read 4(-7) as 4x10-7. LLI means the large lower
intestine.
b The letters D, W, and Y (days, weeks, and years) designating the class of
the material in the first column of the table are rough measures of the
amount of time the material remains in the lungs before elimination.
This is mainly governed by the solubility of the chemical form of the
radioactive material involved.
c ICRP-2 metabolic models, and intake and biological parameters for standard
man (IP59).
d ICRP-30 metabolic models, and intake and biological parameters for
standard man (IP75,IP79).
B-4
-------
References
Ec80 Eckerman, K. Private communication. Oak Ridge National
Laboratory, Oak Ridge.
IP59 International Commission on Radiological Protection.
Publication #2, Report of Committee II on Permissible Dose for
Internal Radiation, Pergamon Press, New York.
IP75 International Commission on Radiological Protection.
Publication #23, Reference Man; Anatomical, Physiological and
Metabolic.Characteristics, Pergamon Press, NewYork.
IP79 International Commission on Radiological Protection.
Publication #30, Limits for Intakes of Radionuclides by Workers,
Pergamon Press, New York.
ft U S. GOVERNMENT PRINTING OFFICE : 1981 337-100/8003
B-5
------- |