UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D C 20460
July 19, 1388
O f F I C g OF
THE AQMlNlSTB A TOO
SAB-EHC-88-037
Honorable Lee M. Thomas
Administrator
U.S. Envorinmental Protection Agency
401 M street, s.w.
Washington, D.c. 20460
Subject: Science Advisory Board's review of the SELENIUM health
criteria document
Dear Mr. Thomas:
The Metals Subcommittee of the Science Advisory Board's
Environmental Health Committee has completed its review of the
Drinking Water Health Criteria Document for Selenium dated
December 1986. The review was conducted January 14-15, 1988 at
the St. James Hotel in Washington, D.C. The Subcommittee made
two recommendations,* viz, that more information was needed in the
document on the studies on which the standard is based and that
the DWEL should be 200 mierograms/L based on an uncertainty
factor of 10.
The conclusion in the criteria document that the Drinking
Water Equivalent Level (DWEL) should be 107 micrograms/L is based
mainly on the Yang et al (1983) study and the National Academy of
Sciences (NAS) 1980 conclusion that the safe and adequate range
of daily selenium intake is 50-200 micrograms. One justification
for setting the DWEL at this level is that drinking an average of
2 liters per day would provide close to the upper limit of 200
micrograms per day. To arrive at this conclusion it was
necessary to set the uncertainty factor at 15, a somewhat unusual
value but the only one that would allow bath the use of the Yang
et al study and also give an allowable intake close to the upper
limit of the NAS recommendation.
The report properly focuses on the selenium dilemma; the
apparently narrow margin between minimal daily intakes required
to maintain health and intakes associated with toxicity. The
primary source of data for establishing the DWEL is the report of
Yang et al (1983), The subcommittee recommends that this report
be discussed at greater length because of the reliance placed
upon it. For example, the authors discuss two areas labelled as
high selenium areas. In one, selenosis was common in livestock
and toxic signs were apparent in humans. In the other, selenosis
-------
has never been reported. In the tables, however, the first area
is seen to be associated with hair and blood levels about eight
times higher that measured in the second. ' If 3,2 mg/day of
selenium is estimated as the LOAEL, based on the first area, then
400 ug/day would represent a level below the NOEL* To be
consistent, then/ the language in the document should reflect the
concept that the National Academy of Scieces figures may
overestimate the uncertainty margin. Morever, even in high
selenium areas, drinking water seems to be an insignificant
contribution to total intake. If Yang et al are correct in their
assertion that the ratio of toxic to minimal blood levels is
about 16, a figure not too deviant from the HAS value, then the
uncertainty factor could be set to 5-10 without compromising
safety and without jeopardizing a possibly useful source of this
essential element. Also, the 70 kg standard EPA human is
probably not representative of the Chinese population, another
margin in the uncertainty calculation,
Because the Yang et al study and the NAS recommendation play
such an important role in the development of the DWEL, it is
recommended that more detail be provided regarding the studies.
The Yang et al study is not clearly described in the criteria
document and it is not clear what the minimum daily intake of
3200 microgra»s/day (LOEL) used in deriving the DWEL refers to.
Does the study deal with individuals or geographic areas? How
many people were involved? How was selenium intake measured?
Also more information is needed on how the NAS arrived at its
recommendation of 50-200 microgram daily intake value.
A second recommendation is that the USEPA consider raising
the DWEL to 160 micrograms/L by using an uncertainty factor of 10
instead of IS. Sounding off to one significant figure could lead
to a DWIL of 200 micrograms/L. The reasons for this are as
follows:
1. In the Yang et al study the LOEL was 3200 micrograms per
day and no lower values have been reported.
2. A statement on page, VI-30 that chronic human ingestion
of 200 micrograms per day may be related to increased incidence
of symptoms is not supported by any of the data presented other
than the NAS report.
3. A daily intake of 214 micrograms (107 mierograms/L x 2
L/day) is too close to the value needed for human nutrition (50-
200 micrograms per day) and too far from the lowest value at
which human symptoms occur,
4. There is little or no evidence for proposed human health
effects such as cancer or teratogenic effects. Selenium
is not regarded as a human carcinogen.
5. Uncertainty factors should reflect likely beneficial
effects as well as harmful effects. There is limited evidence
that selenium deficiency is related to cancer and perhaps
cardiovascular disease.
-------
Sincerely,
Norton Nelson, Chairman
Executive Committee
Richard A. Griesemer, chairman
Environmental Health Committee
Bernard Weiss, chairman
Metals Subcommittee
-------
,uv,'l !"«*iii.c,!l .3 - j r.jCc011-_ *. -jt'j;i(_'
•Science Aavisory
Environmental Health
S ubc oitii i t c e e
Dr. Bernard Weiss [Chair j, Professor, Division of Toxicology, P.O.
RBB, University ot Rochester, School of heaieine, Rochester, WY 14642
(716)
Dr. Konaio Wyzga [ V ice-chair J , Electric Power Research Institute, 3412
Hi llview Avenue, P.O. Box 1041, Palo Alto, California 94303 (415) 855-2577
Dr. Thomas Clarkson, Processor ana Head, Division of Toxicology, University
ot Rochester, School ot iieaicine, Post 01 f ice Box KBB, Rochester, New York
14642 (716) 275-3911
L)r. Gary Diamond, Assistant Professor of Pharmacology, Universicy ot
Rochester School of Ueaicine, Rochester, hew York 14&42 (716) 275-5250
Ur, Phil Enter! ine [hNVls^MlENTAL hiALTh CUtlillTELj, Department ot Biostaciscics,
t/raduate School of Public health. University of Pitcsburg, 130 Desoto Street,
Fitcsburgh, FA l^bl (412)
Dr. Robert uoyer, Uepartruent ot Pathology, .health Sciences Centre,' University
of Western Ontario, London, Ontario, Canada N6A5C1
\jc, liarvin Kuschner, Dean, School of Medicine, Health Science Center,
Level 4, State University on JNew York., Scony Brook, Net«r York 11794
(316)
Dr« Brooke T, tiossman, Department ot Pathology, The tniversicy of Vermont,
Medical Alumni Building, Burlington, Vermont U5405-00&B (802) 656-2210
Dr. Gunter Oberdoerster, Associate Professor, Kadiacion Biology ana
Bioptiysics Division, University ot Rochester, School ot Uedicine, 40U
Avenue, Rochester, N,Y. 14642 (716) 275-38U4
Dr. i'. William Sunderman, Professor of Laboratory hedicine and Pharmacology
and Heaa, Department of Laboratory liedicine, Universicy of Connecticut
health Center, Room C 2021, fanning t on , Connecticut 06032 (203) 674-232B
Executive Secretary
Ur» Richard Cothern, Executive Secretary, Environmental health Committee,
science Advisory Board [A-1Q1FJ, U.S. Environmental Protection Agency,
Washington, D.C. 2046U (2U2) 382-2552
------- |