UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                      WASHINGTON, O.C. 20460


                                    EPA-SAB-LTR-91-001

                                                        OFFICE OF
       22,  1991                                       THE ADMINISTRATOR


Honorable William K. Reilly
Administrator
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC  20460

     Re:  Letter  Report  on  Radon Risk  Estimates   for  General
          Population and Smokers, Non-Smokers, and Children

Dear Mr. Reilly:

     The Radiation Advisory Committee of the Science Advisory  Board
has reviewed the February  15,  1990 draft document "Estimation of
Risks from indoor ladon Exposure."  In his memorandum of February
17,  1990,   the  Director  of  the  Office  of  Radiation Programs
requested  that  the Science Advisory  Board,  "assess whether the
scientific  radon risk assessment  information  is  appropriately
summarized."   Dr.  Fuskin of  the  Office  of  Radiation Programs
briefed the  Committee  on the specifics  of  the risk estimates on
February 18  and was available  to discuss the document with the
Committee at its public meeting May 17-18, 1990.

     This document,  prepared by office of Radiation Programs staff,
addresses  risk  to  the  general population as  well as to the sub-
populations of smokers, non-smokers and  children.  For  estimating
the risk of  radon exposure to the general population, the Office
of Radiation Programs essentially followed the earlier (June  1988)
recommendation of the Radiation Advisory Committee that the average
of the  projection  values from the  National  Academy of Sciences'
model developed by.the Board on .the Effects of Ionizing Radiation
(BEIR  IV)   and  the  International  Commission   on  Radiological
Protection's ICRP 50 Model be used (of,  letter to Honorable Lee M.
Thomas of  September  9,  1988?  SAB-RAC-88-Q42),    Since there has
been no new scientific data reported since 1988, the Committee sees
no reason to alter that recommendation.

     The BEIR IV model in its projection assumes a decreased excess
relative risk with time since exposure,  a decreased  risk with age
of the individual,  and a minimum latency period  of 5 years.

-------
     The ICRP  50  model uses a constant excess  relative risk but
assigns a three-fold higher risk for exposure to  children and young
adults below the  age  of 20 years.  Although there is admittedly
limited evidence,  the ICRP felt it prudent to assign a higher risk
coefficient  to children  and young  adults   than  to more  mature
individuals.   The ICRP 50 model  uses  a minimum  latency period of
10 years.

     Both  the  BEIR  V and  ICRP  50  models are  relative  risk
projection models based on data from studies  of underground mining
populations.   Despite their differences,  the two  models project
quite  similar  estimates  of  cancer  fatalities for  a lifetime
exposure to a given concentration of radon.

     The  Office  of  Radiation  Programs  has  introduced  minor
modifications to the two models!   In the case of the ICRP 50 model
ORP  applied the same risk  coefficient to  adult members  of the
general  population   as  had   been  determined   for  the  mining
populations from which the risk projection model was derived; the
original ICRP  50  had  reduced  this  risk coefficient by 20%  when
applied  to  the general population.   In the BEIR IV  model, the
Office of Radiation Programs corrected the  risk  coefficient for
the  average background radon  level.   The  results of  these two
modifications  of  the  models  are to  increase the  lifetime  risk
estimate by the former,  and decrease the  risk estimate  by the
latter,* the average value  of the  two estimates remains essentially
the same, however.

     The Committee does not  object  to the modifications that the
Office of Radiation Programs has introduced to the models. That the
BEIR IV did not adjust for  the average background  level of radon
might be viewed as a minor shortcoming of that model.   The Office
of Radiation Programs adjustment of the risk coefficient  in the
ICRP 50 model is based on the opinion that ICRP's 20% lowering of
the  risk coefficient, to  miners . .when,- applied to  the  general
population  is   not  warranted.    In this  regard,  the  Office  of
Radiation Programs is  following the recommendation of the BEIR IV
report. The conversion of  exposure to  dose for the two populations
is fraught with great  uncertainties and some  of the known modifying
factors are likely to cancel each other.

     Overall, the lifetime risk projections  for any of the models
fall within the uncertainty range for those  of the others and,  as
already pointed out,  the "fine-tuning H of the models done by the
Office of Radiation Programs did  not  change  the average value  of
the estimates.

-------
     The average life-time risk  projection value for the general
public derived and used by the Office of Radiation Programs is 360
excess lung cancers per 10  person-working level months (W1M).  For
an average U.s, residential exposure of 0.2S WUt per year and for
a population of 240 million,  this translates to 21,600 deaths per
year of 14% of all  lung cancer deaths in the U.S.A.  The Committee
finds this value acceptable and as good as can be achieved at the
present time.  It  would be highly desirable,  however/ to present
this risk  estimate with an uncertainty range/  such  as the range
from the ICRP report as described  in  Section F of  the document
being reviewed.

     There is an important relationship  in lung cancer causation
between cigarette  smoking and exposure  to radon.   In  order to
establish a perspective on the combined risk of exposure to radon
and smoking, the Office of Radiation Programs should qualify the
overall population risk from  exposure to radon in terms of smoking
and indicate  the comparative  risks to smokers  and  non-smokers.
The Office of Radiation Programs should provide advice that there
is an  increased  risk of lung cancer from  continuing exposure to
elevated concentrations of indoor radon, and  also  that this risk
is greatly increased by smoking because of the near-multiplicative
interaction between radon exposure and smoking.  Such advice would
allow smokers to recognize that their overall risk may be greatly
reduced by the  combination  of  cessation of  smoking  and  radon
reduction.

     In regard to  the  assumed  greater risk for radon exposure in
children/  direct evidence is lacking.  However/ follow-up data on
atomic bomb survivors  in Japan indicate  that individuals exposed
as children may  be a higher relative risk for radiation-induced
cancer.  In addition,  dosimetric   modeling suggests  that  for a
given exposure to radon and radon daughters, children will receive
a higher radiation  dose to the bronchial epithelium than do adults.
     One final comment is offered in regard to the present Office
of Radiation Programs' estimate for risk from radon exposure.  The
office of  Radiation Programs should continue to  review its risk
estimate as new information is gained.  The Committee understands
that the Office of Radiation Programs has commissioned a National
Academy of Sciences/National Research Council study of whether any
correction should be required when using epidemiological data from
underground uranium  and  iron miners for the estimation of radon
risk to  the general  U.S,  population.   When  the results  of the
National Academy of Sciences/National Research Council study become
available, they should be reviewed,  as changes in the current risk
assessment may be warranted.

-------
     in conclusion, the Committee believes the office of Radiation
Programs has adequately treated the available  information on lung
cancer risk from radon and arrived at an estimate for lifetime rislc
for the general population which is as good as can be achieved at
the present time.  The committee believes the Office of Radiation
Programs should qualify the overall population rislc  from exposure
to radon in terms of smoking and indicate the  comparative risks to
smokers and non-smokers.  The Committee also believes it is prudent
to assign a higher risk  coefficient  to children and young adults
than to  more mature  individuals  even if  supportive evidence is
limited in regard to this point.   The Office of Radiation Programs
should continue  to review its risk  estimate  and risk assessment
methodology as additional data and technical consensus documents
become available.

     We appreciate the opportunity to present our advice concerning
this radon document and would appreciate receiving a written
response which addresses our recommendations,

                         sincerely,
                         Ra.yftond--CTTjoehr, Chairman
                         Executive Committee
                         Science Advisory Board

                      • i ^i"_^     ^"s..

                     ^^•^^^^^^^^^^^L.^^^
                         Oddvar P. Nygsar«, Chafrnian
                         Radiation Advisory Committee
                         Science Advisory Board
cc:  ORP

-------
ADDENDUM TO LITTER REPORT NUMBERS EPA-SAB-LTR-91-Q01

     Ret  Letter Report on Radon Risk Estimates for General
          Population and Smokers, Non-Smokers, and Children

     The review by the Radiation Advisory Committee of the "Radon
Risk Estimates for the General Population and Smoker, Non-smokers
and Children" was conducted during the spring of 1990.  The
letter report was essentially completed on May 18 and represents
the committee's recommendations based on material available to it
at that time.  Although the Committee does not want to alter its
endorsement of the Agency's risk estimates, it wishes, at this
time, to expand on its comment in regard to the 3-fold greater
risk of lung cancer, assigned to individuals exposed as children
of young adults below the age of 20 years, since this current
risk estimate derives from the SAB 1988 recommendation that the
Agency should average the risks projected by the two radon risk
models proposed by the BEIR IV committee and the ICRP-50 report,
but use the ICRP-50 model for exposures to children and young
adults.  As already pointed out, the only supportive evidence for
a higher relative risk for children derives from studies of
Japanese atomic bomb survivors exposed to external gamma and
neutron radiation.  In a recent case control study of Chinese tin
miners, Lubin et al. (Cancer Res. 50, 174-180, 1990)  report that
there is no evidence for a higher risk to workers who were first
exposed to radon under the age of 13 years.  This study, although
not statistically significant by itself, is apparently the first
report of lung cancer in individuals exposed to radon at an early
age.  Additionally, the recently published (1990)  BEIR V report,
the consensus report of the National Research Council's committee
on the Biological Effects of Ionizing Radiation, states (p. 273)
that "The Committee's analysis of respiratory cancer in A-bomb
survivors showed little effect of age at exposure..,.." which
greatly weakens the rationale for the higher risk to children,
referred to above.

     Another area of uncertainty is the conversion of radon
exposure to lung dose, the relationship of this conversion in
underground miners va. the general—population, and whether
children receive the same dose as do adults from a given radon
exposure.  The National Research Council under contract to EPA
has recently completed a review of this particular issue, and the
report should be available to the Agency by the end of December
(1990).

     In the opinion of the Radiation Advisory Committee, the
review by the National Research Council might be a significant
document for the assessment of radon risk to the general US
population, including children, and the committee emphatically
recommends that the Agency give high priority to reviewing this
report with a view to whether its conclusions warrant a re-
assessment of the currently recommended risk coefficients to the
general population based on radon exposures.

-------