nit** iUttJ Orfica of tn* Administrator $AS-RAC-5?-C2S
rwironminti! Prstsitiofi Sri*nct Advisafy taard July 1937
j.fiay Washingien, DC 20U6O
\ '
/ EPA Report of the
Drinking Water Subcommittee of the
Radiation Advisory Committee
Review of the
Office of Drinking Water's
Assessment of
Radionuclides in Drinking Water
and Four Draft Criteria Documents
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON D C. '20460
July 21, 1987 S&B-MC-87-G35
Honorable Lee M. Thomas
Administrator
U.S. Environmental Protection. Agency
401 M Street, S.W,
Washington, D.C. 20460
Dear Mr. Thomas?
The Radiation Advisory Committee of the Science Advisory Board has
completed its review of the Office of Drinking Water's evaluation of
radionuclides in drinking water and four draft criteria documents that
support this evaluation, including Man-Made Radionuclide Occurrence,
Radium, Radon and Uranium.
At the request of the Office of Drinking Water, the Canrnittee addressed
four issues; the weighting factors to be used in effective dose equivalent
calculations, the chemical toxicity and radiotoxicity of uranium, the
linearity'of the dose-response curve for naturally occurring radionuclides,
and the appropriate use of the relative and absolute risk models.
The Subcommittee concludes that the Office o£ Drinking Pfeter has
developed scientifically conprehensive assessiaent documents. The basis
of the risk estimates developed and the text describing it, however,
should be as precise, clear aad consistent as possible. The documents
should include-a clearer exposition of the basis for the risk estimates used,
the concept of effective dose- equivalent and the weighting factors employed,
More care and consistency in the use of quantities and units is recanmended.
Discussion of these issues is presented in the attached report.
The EPA has adopted the effective dose equivalent approach, which
is a useful and applicable methodology in itost circumstances involving
radionuclides in water. A notable exception is radium, where the direct
hiroan information differs very significantly from risks derived from the
effective dose 'equivalent concept*
-------
— rt _
The Committee appreciates the opportunity to present its scientific
views and is prepared to provide additional assistance that the Agency
requests, We request that the l^ency officially respond to the attached
report, indicating which of the recommendations the Office of Drinking
Water plans to accept or reject.
[Till
Chairman
fiat ion Advisory Coroittee
Norton Nelson
Chairman
Science Advisory Board
-------
Review
of
The Office Of Drinking water's Assessnent Of
Radionuclides In Drinking Water
and
Four Draft Criteria Documents:
Man-Made Radiocuclide Occurrence
Uranium
Radium
Radon
by the
Drinking Water Subcommittee
Radiation Advisory Committee
Science Mvisory Board
U. S. Environmental Protection Agency
Juiy 1987
-------
11
NOTICE
Tnis report has been written as a part of the activities of the
Science &3visory Board, a public advisory group providing extramural
scientific information and advice to the Administrator and other
officials of the Environmental Protection Agency. The Board.is struc-
tured to "provide a balanced expert assessment of scientific issues re-
lated to problems facing the Agency. This report has not been reviewed
for approval by the Agency andr hence, the contents of the report do
not necessarily represent the views and policies of the Environmental
Protection Agency, nor of other agencies in the Executive Branch of the
Federal governmentf nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
Ill
U.S. Environmental Protection Agency
Science Advisory Board
Radiation Advisory -Committee
Drinking Water Subeorraittee
Roster
Chairman
Dr. warren Sinclair, President, National Council on Radiation
Protection and Measurements, 7910 Wocdraont Avenue, Suite 1016
Bethesda, Maryland 20814
Members/Consultants
Dr. Seymour Abrahamson, Professor .of Zoology and Genetics,
Departinent of Zoology, University of Wisconsin
Madison, Wisconsin 53706
Dr. John Barley, Box M268, Betoken, New Jersey 07030
Or. Andrew Stehney, Building 203, Argonne National Laborarory
9700 South Cass Avenue, Argonne, Illinois 60439
Executive Secretary
Ms. Kathleen W. Conway, Deputy Director, Science Advisory Board
Radiation Mvisory Cotmittee, 401 M Street, S.W, (A-101F)
Washington, D, C. 20460
Staf£ Secretary
Ms. Dorothy M. Clark, Staff Secretary, Science Mvisory Board
. Radiation Advisory Committee, 401 M Street, S.W. (A-101P)
Washington, D..C. 20460
-------
IV
Table Of Contents
Page
I. Introduction 1
II* Responses to the Four Issues Raised by 1
the Office of Drinking Water
1. Weighting Factors 1
2. uranium Radiotoxic Effects and Chemical Toxicity 2
3. Dose-Response Curve 2
4. Relative and Absolute Risk Models 2
lit. Comments on the Office of Drinking Water's Draft 3
Criteria Documents
A. General Comments 3
B. Technical and Editorial Comments 3
1. Radionuclides in Drinking Water 3
2, Man-Made Radionuclides 4
3. Uranium 5
4. Radium 5
5. . Radon 6
Appendix Is Request for Science Advisory Board Review
-------
I. Introduction
On January 15* 1986 EPA's Office -of Drinking Water (QOW) officially
requested the Science Advisory Board (SAB) to conduct a review of a number
of scientific issues related to its evaluation of radionuclides in drinking
water, the SAB Executive Committee considered and accepted this request
and authorized its Radiation Advisory Committee to carry out the review.
The latter Ccnwiittee formed a Drinking Water Subcommittee to fulfill the
Executive Committee's charge*
The Subctnimittee subsequently received the following documents £rcm ODW:
o Radionuclides in Drinking Water (for publication in the Federal
Register)
o Criteria Document on Man-Made Radionuclides
o Criteria Document for Uranium
o Criteria Document for Radium
o Criteria Document for Radon
o Office of Drinking Water's January 15, 1986 Memorandum, "Request
for Review of Scientific Basis of Proposed Reccnraended Maximom
Contaminant Levels for Radionuclides in Drinking Water11 (see
Appendix I) that includes four issues for review.
The members of the Subcommittee examined these documents according
to individual assignments based on their expertise. They prepared written
cements on frost o£ them prior to the open meeting in Washington on
September 25-26, 1986. Dr. Abrahamson was unable to attend this mseting.
The Subcommittee developed its report in three parts. These include:
o Sesponses to the four issues raised by the Office of
Drinking water.
o General comments on the documents and the review as a whole
o Technical and editorial comments on each of technical documents
Individual inembers of the Subconmittee have also prepared detailed page-by-
page editorial comments that have already been forwarded to the Office of
Drinking Water.
II. Responses To The Four Issues Raised By TheOffice Of Drinking Water
Issue fl. "In the calculation ofc the.effective dose equivalent, should
the weighting factors developed by the ICRP or those developed by EPA based
.on BEIR III be used?"
The Subcownittee concludes that the ICRP weighting factors should be
used in their original form because they are well established, and widely
-------
published and used. Furthermore, they have not been invalidated in any
way, and they include a component for genetic risk. Also, they form the
basis of ICRP 30, Annual Limits of Intake of Badionuclides by Ptorkers,
and these are used directly by EPA,
The EPA weighting factors, although in some respects more traceable
to original data than are those of the ICRP, are not as well established or
widely known, and they have not been thoroughly published and justified.
They do not include a component for genetic risk. Furthermore, their
application would not alter the results obtained for drinking water in a
significant way, except perhaps for bone. However, the Subcommittee notes
that ICRP weighting factors express a result for radium more consistent
with human data. It is not clear why more than one set of weightir^g
factors is necessary.
Issue _#_2. "Are the radiotoxic health effects of ingested uranium
sufficiently well substantiated to be used as the basis of a regulation?
Should uranium be in category A (of EPA's risk assessment guidelines for
cancer) and on what basis? Do you agree with the non-radiotoxicity
health assessment?"
The Subcommittee considers that the radiotoxic health affects of uranium
(renal effects) are not sufficiently well substantiated to form the
scientific basis of regulation, but uranium should be in Category A on a
presumed (but not demonstrated) likelihood of carcinogenesis. The Subcom-
mittee agrees with the non-radiotoxicity health assessment and concludes
that chemical toxicity should constitute the scientific basis used for
regulation.
Issue t3« "is the dose-response curve for naturally occurring
radionuclides linear?"
The Subcommittee agrees that linearity for naturally occurring
alpha emitters is appropriate in the dose range of interest.
Issue t4. "Should the relative or absolute risk model value be used?"
The Subcommittee believes there are no data enabling a clear
scientific choice of relative and absolute risk models for most cancers
but the reccranendation made in the October 28, 1985 report by the Radiation
Advisory Committee in its review of low-level radioactive waste disposal
standards, namely that the risk model considered most appropriate for the
specific disease be applied rather than averaging, is supported by the
Subcommittee. Thus, the absolute risk model should be used for leukema
and bone cancer -and the relative risk model for other sites identified
as associated with radiation induction. The Subcommittee notes that
averaging could be conducted in itiany different ways, and introduce new
complexities.
-------
•- 3 -
III*. Cccments on the Office of Drinking Water's Draft Criteria Documents
A. General Comments
1* The Subcommittee believes that the choice of risk estimates in the
documents is riot sufficiently clear* All the numbers used should be clari-
fied in teems of their references and scientific justification, and they
should be consistently applied. Though the BEIR III (1980) and BEIR (1972)
reports are referred to as the source of some estimates, considerable reli-
ance was also placed on two documents—"Radionuclides: Background Information
Document for Final Rules, Volumes I and II" (EPA 520/1-84-022-1 and 2)—
that were not referenced in some criteria documents. The Subcommittee also
has seme concerns about the choice of the risk coefficient, which seemed to
be 200 deaths/100,000 people for 0,1 rem/y for a life time exposure (BEIR I),
This number agrees with a linear linear model result from BEIR III and is
described in one document as a "safe" or prudent assignation for regulatory
purposes. This choice corresponds to about 3 x lQ~Vrem an^ is ni3n compared
with the preferred linear-quadratic estimates in BEIR III (1980) or those in
UNSCEAR 1977.
2, The concept of the effective dose equivalent (of ICRP) should be ex-
plained carefully as it has a central role in the documents. Also, the
documents should clairfy the difference between effective dose equivalent
and dose equivalent.
3, More clarity is needed in presenting equations concerned with quanti-
ties such as absorbed dose, dose equivalent and quality factor, Ihere is
also alternating use between conventional and Standard International (SI)
units. EPA should consider placing conventional units first with the SI
equivalent in brackets, and to use conventional units in tables with a foot-
note on SI equivalents.
4. The organization of some of the documents and tables could be improved,
As far as possible, a logical sequence in the documents should be follo%red
that might take the following forms metabolism and toxicology —^ concentration
—> dositnetry '—> risk. As for the tables in Radionuclides in Drinking
Water, Appendix D, the first column begins with risk nvfltibers, whereas these
should logically be located in the last column.
B. Technical and Editorial Comments
Radionuclides in Drinking Water
1. This document .should present a clearer exposition for the basis
of the risk estimates used, the concept of effective dose equivalent and the
weighting factors .employed. None of these are sufficiently clear and well
developed in the draft reviewed by the Subcommittee. The Subcommittee under-
stands the concept of effective dose equivalent and the use of weighting
factors, but it recommends additional clarification because of the widespread
use of'the documents by scientific and non-scientific personnel.
-------
4 -
2. the Subcommittee supports the Office of Drinking water's adoption
of the effective dose equivalent approach, a useful and applicable methodology
in roost circumstances involving radionuclides in water. A notable exception
is radium where direct human observation -differs significantly from risks
derived from the effective dose equivalent concept. It is essential that the
human data be fully discussed and made clear. Some Modification will also be
necessary for uranium since the SPA radiological risk estimates are based
on radium.
3. Evidence of internal ineonsistenees in some of the tables (in Appendix
D, for example) are noted in the detailed comments and in discussion. All
tabular data should be carefully examined and revised where inconsistencies
occur.
4. Many available analytical procedures are equal or better than those
presented in the document. EPA should consider using performnce specifica-
tions for the procedures rather than prescribing a specific method*
tMan-Made Bad ionucl ides . .
1, The statement on Page 1-2 concerning isotope concentrations that
"can produce a dose equivalent of 4 mrsm/yr whole-body radiation" is likely
to confuse the reader because it actually refers to effective dose equivalent
(IDE), The EDE concept should be introduced here and used in the document to
avoid confusion with "dose equivalent,"
2, The risk parameters in Table Ill-l include some unclear choices; a
latency period of 10 years for lung cancer but 15 years for other tumors of
soft tissues? a 0 of 1 for breast and thyroid, 20 for leukemia, and 10 for
all others. The document should present a rationale for these choices or,
possibly, they should be reconsidered.
3. The tables of radionuclide concentrations are developed on the basis
of risk equal to that from a whole-body dose rate o£ 4 mrem/yr. No rationale
is given for using this basis, only the risk estimate of 8 excess cancers
(lifetime) per 100,000 people. EPA should consider alternatives, such as
5 raran/yr and 10 excess cancers or 1 mrera/yr and 2 excess cancers. The
rationale for choosing 4 i!ireM/yr should be stated.
4. In Table III-8, the comparison with ICRP 30 shows agreement within a
factor of 2 for the beta emitters except for 131-[, some explanation for the
large difference for ^Ij should be given.
5. In the section on analytical procedures, it is not clear whether or
not these methods are recommended or prescribed. The Subcommittee believes
that the principal requirement should be performance because better procedures
-------
_ 5 —
may become available, and merely following a vffitten procedure does not
insure that adequate data will be obtained.
6. Unexplained Inconsistencies between this document and the documents
on radium and uranium are noted as follows;
a. Hie organ weighting factors from Table III-l of this document
differed from those given in Part VIII of the radium and uranium
documents.
b. For chronic exposure, the radiation dose rates at specified tiroes
times after birth ace used in this document for estimtes of life-
time risks, but the radium and uranium documents use a constant
radiation dose rate throughout life equal to that attained after
70 years of intake.
Uranium
1. There needs to be more emphasis on the factors, or range of factors,
carried over from one section to another. This could be done by tabulation
or sunmarization at the end of appropriate sections.
2. The "-weighting factors attributed to BEIR III should be credited to
EPA instead. The general method of deriving these new factors should be stated
briefly in each document.
3. The uranium document, in particular, is based on secondary sources.
Some references to the primry literature would be helpful.
4, The lack of evidence for carcinogenesis is properly alphasized in
Section VII but needs to be repeated at the top of Page VTII-3.
Radium
1. The organ risk estimates in Tables VIII-1 and VIII-2 are inconsistent
with the Effective Dose Equivalent rates for the organs; it seems evident
that the organ v
-------
- 6 -
5. The section on "acute" health effects (VI.E) is really a general des-
cription of all radiation effects. It seems out of place as a subsection of
radium effects and would be more suitable as an appendix. Also, this section
omits the acute effects that actually have been ascribed to ingestion of large
amounts of radium (e.g., blood dyscrasias and necrosis of the jaw bone) in
early papers on radium dial painters.
Radon
1. There needs to be more enphasis on the factors, or range of factors,
carried over from one section to another. This could be done by tabulation
or sumnarization at the end of appropriate sections.
2, The radon exposure from water derives from its domestic uses, not from
drinking water. This could be emphasized further, possibly by using the tern
"drinking and domestic water supplies."
3. The weighting factors attributed to BEIR III should be credited to EPA
instead. The general method of deriving these new factors should be stated
briefly in each document.
-------
'i UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
? WASHINGTON. D-C. 20*60
JAM i S, |C?5
jni 1 I ^j Iwu** «<-M»i~«; rtp
SUUECTs Request for Review of Scientific Basis'.of Proposed
Recommended Maximum* Contaminant Levels for Radionuclides
in Drinkrng Water
FROWJ Jogstm A ^^jhjvo^^strfector
standards Division, ODW (WB-550)
TO; 'Terry Yosie, Director
Science Advisory ioard (A-1Q1)
*
I request .that the SAB review the scientific basis of
the proposed Recommended Maximum Contaminant Levels (RMCLs}
for radionuclides in drinking water* In addition, I request
that we be placed on the agenda for the SAB meeting in Las
Vegas on January 21-22, 1986 to give an overview of the
proposed RMCLs for radionuclides in drinking water and to
discuss the specific scientific areas upon which we would
like SAB to focus*
The Office of Drinking Water is currently developing the
Revised National Primary Drinking Water Regulations for
Drinking Water. The third phase of this rule-making process
involves .the-development of regulations for radionuclides in
drinking water. We are developing these regulations in two
steps. First, we propose RMCLs, which are non-enforceable
health goals that are set at levels that would result in no
known or anticipated adverse health effects and which allow
an adequate margin of safety. When the 1MCLS have been
published as a final rule, we take the second steo of
proposing a Maximum Contaminant Level («CL). MCLs are enforce-
able standards and must be set as close to RMCLs as is feasibl*
Feasible means'"with the use of the best technology, treatment
techniques and 'other means, which the Administrator finds are
generally available (taking costs into consideration)*.
-------
-2-
A proposal has been developed for RMCLs for radionuclides
in drinking water. The proposal has cleared the red border
review process and will be sent to the Office of Management
and-Budget this month. Publication of the proposed RMCL's
in the Federal Register is anticipated in March 1985. The
final HMCL and the proposed MCL are scheduled to be published
in the Spring of 1987.
Support documents required in the regulatory process
that summarize the occurrence and potential health effects of
radionuclides in drinking water, are available 'for SAB review.
Thesa documents include: health effects criteria "documents on
radium, uranium, radon and man-made radionuelidesj the proceed-
ings of the National Workshop on Radioactivity in Drinking
Water (the May 1985 issue of the Journal Health Physics)?
various scientific publications.
I would like the Science Advisory Board to review the
scientific basis of the prooosed RMCLs and to specifically
•focus on the four questions that are attached. A brief
statement on OOWs current position is provided for each
question that indicates the approach we are now taking. Much
oC the detailed analysis relating to thes* four questions
and other possible questions are contained in the documentation
described above,' We would be happy to provide any or all the
available documentation to SAB members*
In'addition, I would like to encourage the SAB to consider
two specific needs in our development of the rules for the
RMCLs and MCLs. We have an interest in the epidemiology
studies of- indoor air radon levels and their relationship to
lung cancers. Also* we have some concern about the current
data base relating drinking water levels of radon to indoor
air levels.
"it is my understanding that the SAB will be reviewing
the protocol for the epidemiology study in the northeast
involving the Kaine Medical Csnt-sr, the State of Maine/ and
the university of Maine. We would like to add our support to
this study and your review of it since the results will be most
useful to us.
- \
The largest contribution to uncertainty for the estimated
population risk due to radon in drinking water is the numerical
relationship 'between drinking water concentrations and indoor
air concentrations. The. existing data base for this numerical
relationship is somewhat weak. The support of the SAB would be
appreciated in encouraging 3?A's Office of Research and
to conduct A study broadening this -"iata base.
-------
-3-
If you have any questions or desire further information,
please let me know or call Dr. Rick Cothern at (202)382-7584.
ccs Rich Guimond, ORP (ANR-460C)
.Dave Janes, ORP (ANR-460C)
-------
SCIENTIFIC QUESTIONS RELATING TO THE DEVELOPMENT QP THE
REVISED REGULATIONS FOR RA010MUCL1DES IN DRINKING WATER
Question II. In the calculation of the effective dose equivalent,
should the 'weighting factors developed by the ICRP or those
develpped by SPA based on EEIR III be used?
Current OOW Position The ICRP weighting factors relate
to occupational exposure. There are some minor adjustments
needed for these factors to best describe the possible responses
to general environmental exoosure. The BEIR HI based weighting
factors describe the general environmental exposure more
completely. This approach follows that used in QRP's background
document for the development of the Clean Air Act' regulations
and is the same,as that being used in the develooment by QRP
of the low level radioactive waste standard. (See attached
table from the Health Criteria Document on Man-*ade tadionuclirles
that gives some idea of the difference, due to the choice of
different weighting factors)
Question 12 ' Are the radiotoxic health effects of ingested
uranium .sufficiently well substantiated to be used as the
basis of a regulation? Should uranium be in category A and
on what basis? Do you agree with the non"*radiotO3ticity
health -assessment?
Current. QD_W Position' Although direct epidemiology studies
have not bean conducted for uranium, it is known to be a bone
seeker and thus its radiotoxicity would be expected to be
similar, to that .of radium. For this-reason, we feel it should
be in Category A (a known number of carcinogen under SPA guidelines)
and can' be used as the basis of regulation.
. The primary chemically toxic effect of natural uranium
is' on- the kidneys. The possible standard based on this non-
radioactive health effect or Adjusted Acceptable Daily intake
for uranium .is (SO microgra«»s/l as shown below assuming a 70
kg adult .consuming. 2 li.ters of water ner day. The calculation
is: ...
2UDl-=-- .(WQAEL) (animal fi ) (adult weight) _,.
, (safety factor}(water consumption/day)(human f^)
* (1 tng/kq/dav) f0.01)70 kg)
CiQO(2 1 day)(0.5) . -
= 60 micrograms/1 or 40 pCi/1 (rounded off to one
. - .significant fiaure)
Included in the' above determination o'f the AADI for uranium
is a no 'observed adverse, effect level {NQAED of 1
a safety factor of 190 since only animal data is used(
-------
-2-
uptake of 1%? and human uptake of 5%.
Question 13 is-' the dose-response curve for naturally occurring
radionuelides linear?
Current QDW gosition The^BEIK III reports that there is
some 'evidence that the dose-response curve for aloha emitters
may be supralinear* The evidence for radium allows for the
possibility of linear or a quadratic dose-reoonse 'curve. AS
a scientific policy, we have assumed linearity.
*<
Question 14 Should the relative or absolute risk,model
value be used?
Current OOW Position As the ORP does, we average the
values predicted by the relative and absolute risk models. ;
-------
Table 1II-8* Drinking water concentrations corresponding to a
dose of 4 mreitt/yr for selected nuclidas comparing the results
using two different organ risk coefficients; viz, the ICRP 30 set
and the set used in the calculations for the present document
based on BIER III. i
Concentration in PCi/L corresponding
to a dose of 4 mrem/yr using: -. -,
Nu elide
3H
58CO
39Sr
137CS
ICKP 30
89
,000
1^0 _
21(1
680
41
20
74
110
13
2,5
8EIR III/OSEPA
88,000
2200
240
880
43
14
77
• no
39
3.9
* From the Draft Criteria Document for man-made radionuclides
in drinking water.
------- |