Chapter 31 The Validity of Risk Assessments for Lung Cancer Induced by Radon Daughters F. Steinhäusler Division of Biophysics, University of Salzburg, A-5020 Salzburg, Austria Available input data for the risk assessment from low level radon daughter (Rn-d)exposure are mostly either of low quality, partially contradicting or simply "guesstimates". Therefore at present only the upper limit of this risk can be estimated. Results of epi demiological studies amongst miners are associated with large uncertainties with regard to the assess ment of past radiation exposure, lung cancer diagnos tic and/or classification and synergistic effects due to smoking and dust exposure. An alternative approach uses dosimetric modelling for Rn-d inhalation to ob tain Rn-d exposure-dose conversion factors. Large un certainty is caused by individual variability due the influence of life style, physical and biological pa rameters. It is concluded that for "normal" indoor Rn-d exposure the resulting risk is neglegible compa red to other risks "accepted" by society. Lung cancer due to inhalation of radon decay products (Rn-d) represents the single most significant risk from all natural and non-made radiation sources. There is growing concern about its impact on society for occupationally and non-occupationally exposed persons. Increased lung cancer incidence amongst miners has already led to the request by labour unions for a lowering of the presently recommended Rn-d exposure limits (Williams, 1985). Trends in energy conservation efforts and recycling of unsuitable industrial wastes as construction material (Steinhàusler and Pohl, 1983) indicate an increase of a Rn-d induced lung cancer risk for members of the public. In view of the significant socio-economic consequences of these issues it is important to ensure that any input data used for risk assessments are subjected to stringent quality control. Such data can be derived from epidemiological studies, experimental animal inhalation research and theroretical dosimetric modelling. All three data pools are associated with considerable uncertainties. In this paper the following aspects are critically reviewed with 0097-6156/87/0331-0430$06.00/0 © 1987 American Chemical Society
31.
regard Rn-d
to
the
431
The Validity of Risk Assessments
STEINHÀUSLER
applicability
to
the
risk
estimation
for
low-level
exposure: -
limitations gical
-
suitability from
Data
of
Risk
Epidemiology.
related
have
extrapolating
inhalation
from
studies
uncertainties
epidemiolo-
of
to
results
obtained
man
presently
used
lung
models.
Assessment
Using
preconditions
of
animal
inherent
Needed f o r
Rn-d
studies
the
to
be
epidemiological
approach
the
met
establish
a
in
order
to
following dose-effect
relationship: a)
individual have
to
doses
be
due
determined
to
Rn-d inhalation
with
a
high
degree
in of
the
exposed
group
r e l i a b i l i t y
("dose-
data"); b)
in
order
gens
to
discriminate
other
should
than
Rn-d,
be
identified
control
population
from
lung
ideally which
is
a
cancer
cases
specific
induced
by
type Rn-d
due of
to
carcino-
lung
only
cancer ("effect-
data"); c)
a the
exposed
except
for
Individual
and
occupational of
as
its
dose
external is
of
risk
from
period
this
between
up
to
over
period
f i f t y and
is
continous latent
about low
using
even
(e.g. of
inhaled from
10
(occupational
mSv/WLM
C r i t i c a l important free. cancer the
diagnosis
population
assessment gens
in
the
of
of and
on
a
exposure
well
the
latent cellular
expression than
as
ten
a
years
characteristics As
the
latent
lifetime
risk,
lengthening
are
converted The
choice
physiological as
a l l
lifetime
of
the
to
dose
of
the
lifetime.
1983).
on
by
less
the
to
This
exposure,
through
1984). to
1983).
the
smoking
lead
particle
physical
size).
parameters
characteristics
Mean
(non-occupational
values
exposure)
range
to
about
exposure). of
the
data
overlooked
c l a s s i f i c a t i o n , under
on
and
lung
cancer
assumed as
comprises
medical
or
and
exposure
is
equally
largely
demographical
investigation
characteristics
atmosphere.
from
a l l non-
resulting
inducing
individual
"effect-data"
group
smoking
the
as
levels The
a l . ,
on and
Rn-d
influenced
on
et
depends
5 mSv/WLM
to
aspects,
data
Pohl,
level
macroscopic
(OECD/NEA,
(e.g.
frequently
group
is
may
data
volume)
analysis
but
This
about
work.
e.g.
exposure
and
varying
proportional of
comparable a l l
occupational
low
varies
Radford
value
minute
of
or
period
factors
aerosol
to
Rn-d exposure
individual
typically
at
home
exposure
excess
numerical
respiratory
the
in
case
depending
Rn-d
conversion
appropriate
Rn-d
inversely
Subsequently, by
to
1879;
level
period,
the
subsequent
years,
Hesse,
is
in
radiological
(Steinhausler
in
latent
which
exposure.
contributing
school,
the The
defined
requires
i n i t i a l
and
cancer.
(Rarting
Rn-d
chronic the
transformation pulmonary
at
be
investigation
exposure
exposed
e.g.
to
sources
importance
always
l i f e ,
of
assessment
radiation
is
stages
has under
lack
internal
particular
man
group
erroron
information
the to
data
lung on
quantitative
other
carcino-
432
RADON AND ITS DECAY PRODUCTS
A l l of
Members Rn-d a
this
has to
"spontaneous" of
this
be seen a g a i n s t
lung
cancer
population
levels
which
varies
relatively
small
city
Animal
Inhalation
partially logical
Although
exposure
i . e .
the
can occur
laboratory
conditions.
possibility
of
either
at
carcinogens
dust.
In
this
different
lung
carcinogens
achieving
true
randomization
epidemiology dence
data.
least lung
Since
However, studies for
for
humans
a)
i n
even
to
the following of
levels human
should to
should human
diesel
controlled to
the atmos-
fumes
or
ore
superposition
of
possibility
of
The
a further
advantage
lung
over
cancer
Rn-d exposed
i n c i -
animals
i s
constraints,
s t a t i s t i c a l l y to
conditions
significant
use the information studies
should
and
at such
numbers
associated with
similar other
studies,
magnitude
than
physiologically
from
risk
animal
assessments
be met:
and non-occupational
carcinogens
be
control to Rn-d,
offers
exposure
financial
dose-effect
a
least
low Rn-d levels.
be able
be o f
i n
at
under
study
unbiased
of
radiobiological
occupational
exposure
at
the uncertainties i n
of
synergistic
by
can provide
order
extrapolations
b)
only
of
even
Rn-d epidemio-
exposed
levels
type
represents
range
a homogenous
s t i l l
e . g . smoke,
obtaining number
Rn-d related
view
of
overcome
alternative
of
a wide
magnitude
human
can be a d d r e s s e d .
the
induction i n
Rn-d,
limited
experiments
cancer
than
this
or
the question
respect
theoretically
inhalation on
with
of
i s
population.
1983).
to
low environmental
pheric
to
of
experiments
i t s e l f
Furthermore
other
a l . ,
inherent
background
general
exposed
identification group
simultaneous
manner
et
Such
problems
control
increasing
the
by an order
(Steinhàusler
Experiments.
the
i n
are themselves typically
the major
studies,
group. this
one o f
a mostly
cases
high
animal
to
low
dose
Rn-d exposure
as those
occuring
i n
conditions;
Rn-d (e.g.
r e a l i s t i c ,
tobacco
simulating
smoke) those
of
exposure
conditions; c)
dosimetric
models
developed
should
adapted
for
into d)
be
account
radiation animals
Dosimetric enable the
should
atmosphere.
As
describe
bronchial (e.g.
or
for
methods cellular
hits
dosimetric needed
also
to
c e l l
Rn-d risk
to
and
of
Rn-d from
concepts values
for
particles. the
i n
test
dosimetry
i s
Rn-d deposited Rn-d
have
i n the
deposition developed
for
tracheo-
dose
calculations
Carlo
t h e random The r e s u l t s
following
i n
to
been
(e.g.
U s i n g Monte
provide
-geometry;
Rn-d concentration
microdosimetric
assessment:
taking
humans.
dose
account
tract
used,
mechanisms
an inhomogeneous
alpha
should
i n
measured
or
respiratory
animal
induction
objective observed
layer).
possible
by deposited
that
of
test
physiology
mean d o s e
region)
calculations
for
t h e human
dosimetric
regional
basal
lung
from
consequence
pulmonary i s
the
different
either the
i t
of
tract a
the lung,
i n
The main
assessment
for
specific
and cancer
be s i m i l a r
respiratory
within to
differences
sensitivity
Studies.
the
the
calculation nature of
of such
information
31.
The Validity of Risk Assessments
STEINHÀUSLER
a)
which
biological
within by b)
the
alpha
which
have
dose
the
Available
of
Risk
mining
studies
shortcomings individual
have
of
of
" . . . t h i s
a
analysis
estimate
was
cause is
present
not
about
a)
of
the et
activity,
Rn-d
exposure-
a l . .
the
individual ments
and
high
effect
the
a
been of
carried
out
the
later
to
to
the
inherent
unavailable
caveats
stage,
concerning
e.g.
for
the
relationship: not
offered
demonstrate in
States, period
of
severe
due
stated
many
as
a
basis
the
that
lung
for
1976)
low
levels
cancer
of
incidence
1967)
of
to
extrapolate
observation
to
at
lifetime
1983) to
draw
the
any
about
dose
uncertainty
in
and
effect
exposure
and
1973) associated with
into
three with
exposure,
job
conclusions
considerable
grouped
of
of
considered appropriate
associated
uncertainties
at
increase
uncertainties
Rn-d
unattached
dose..."(Royal Commission,
United
not
authors
mostly
emphatically
unit
truncated
can be
size,
significantly
they
dose-effect
error..."(Snihs,
general
particle
aware
data
designed...
possible
studies
f u l l y
their
the
uncertainties
b)
physical
The
detectable
1 0 0 WLM d u e
s t a t i s t i c a l In
age,
research has
most
therefore
r i s k . . . " ( M u l l e r
gical
cells)
probability
dose.
Therefore
per
not
any
from is
(e.g.
investigations,
of
is
risk
rate..."(Congress
below
hit
individual
epidemiologic
quantitative
of
study
exposure
" . . . i t
the
populations.
data.
misuse
deduction
" . . . i t
epithelial
highest
(e.g.
on
factor)
to
been
their
exposure
potential
"...the
the
Assessment
Hitherto,
amongst
original
parameters
exposure
for
Epidemiology.
any
influence
disequilibrium
conversion
the
parameters
most
physical
fraction,
mainly
bifurcations, have
conversion;
which
Data
(e.g. tract
particles.
physiological
sex) c)
targets
respiratory
433
these
epidemiolo-
categories: the
mostly
due
reconstruction
of
to
measure-
the
lack
of
past
mobility diagnosis
and
c l a s s i f i c a t i o n
of
lung
cancer
data c)
uncertainty other
In
order
to
Rn-d
between
"lung
since
these
years
ago,
this
available
have
to
using
e.g. Low
exposure
i s
the
on
indirect
level
from
most
important
not
carried
studies
exposure
concerned,
are
with of
ore
Rn-d exposure-effect
indoor
a in
for
of
than
past
the
a l l
a
the
individual
sufficient Rn-d
studies:
to
provided
press).
pure
induced
these
fifteen be
and
thirty
by
any
degree
Therefore
exposure
of of
they
levels,
content.
studies
situation, available
know
out
cannot
(Steinhausler, estimation
to
Rn-d
atmosphere.
rather
weakness
information
uranium
between
inhaled
carcinogen Rn-d and
particular were
essential
inference
i s
it
accuracy
effects
the
causal relationship
retrospective
and
rely
general
" is
in
suspected
measurements
r e l i a b i l i t y
the
This
a
the
cancer
exposure.
the
synergistic present
establish
association effect
about
carcinogens
at
that
least
miners
may as
in
be
far CSSR,
relevant as US,
to
lifetime Canada,
434
RADON AND ITS DECAY PRODUCTS
Newfoundland can
be
study miners
significant
due
to
As
only
for
these
compared
to lung
exposure
to to
derived Rn-d
cytological
regarded
since
observed
as
of
non-miners
as
In
the
cases
noted
s t a t i s t i c a l l y
associated
period.
members
the
Rn-d exposure
number
than
amongst
male
study,
i t
Norwegian
incidence
non-mining
uncertainties
expected
i s
of
the public,
methods i s
quality
It
with
the
assessments
case
even
of
the US-
exceeded
controls since
higher
i . e .
have
there
the
that
i s
an
frequency
of
by a factor
a l l
a
these
strong
lung
quantification
there
this i n
i s
lung
large
2
to
3
assumed
Rn-d
association
with
of
i s
lung
any other cancer
initiated
argued
with of
and c l a s s i f i c a t i o n
like
the
this
small
medical
Rn-d
for
the
Rn-d induced
type
frequency
However,
procedures
supposedly
questionable
c e l l
low
carcinogens,
that
c e l l .
by
h i s t o l o g i c a l -
by other
repeatedly
an unusual of
induced
with
are
s c i e n t i f i -
a
review
and data
used
induced
"effect"
following
reasons
press):
a high
cancer
degree
tissue
differences
cancer
cancer
frequency
aspects
just
distinguished
has been
high
control
that
diagnosis
present, be
associated
i . e .
(Steinhàusler,
At
cannot
from
cancer
distribution,
of
intra-
and interobserver
classification
i n
t h e numbers
by
variability
pathologists,
classified
as
a
causing particular
type;
the
source
term,
the
result
of
artifacts
biopsy
i n
post-mortem
or
as
examinations
primary
lung
autopsy,
misclassified
are categorized
biopsies; of
i . e .
potentially
a l l 4-0%
also cancer
information.
lung
revealed
margin, generally
cases
exposure smoke.
the
used
on lung
tobacco
c)
for
be
and uncertainties
e.g.
b)
cannot
"safety"
cancer data
level
in
data
were
cancer
the Newfoundland
large
miners
errors
cally
a)
In
levels,
smoke.
Medical
for
relevant
low exposure
cases
lung
the post-1960
studies
members
related
of
these
Czechoslovakian and
cancer
cases
estimates,
the
inherent
amongst
subject
At
the
observed.
additional smokers
the
the
Canadian miners
actually
lung
increase to
1). of
both
exposure
available
case
population.
small
retrospective and
i n
was compared
relatively
(Figure
the more
However,
the general
are
i n
significantly
expected. of
and Norway
seen that
cancer
i s
"small
c e l l
compared
to
are
of
lung
significance
tumors,
e . g .
carcinomas"
i n
anti-mortem
missed
for
crush 25% o f
diagnosis
altogether,
30%
are
false-positive. d)
histo-cytological ciple
because
intermixing carcinoma In
view
of
these
cancer
task,
resulting
Dosimetry. advances to
i s
lung are
fundamental
adeno-,
d i f f i c u l t i e s
and c l a s s i f i c a t i o n
for
the past
cancer
i s
generally
d i f f i c u l t
i n
heterogeneous,
large-
and
small
prini . e . c e l l
frequent.
frequently
extent
of
tumors
squameous,
diagnosis
Dosimetry
over
a large
of
cells
lung
typing
these
i n
data
inhaled
twenty
of
i s
on the WEIBEL A'
Lung
understandable
represents
a
that
d i f f i f u l t
r e l i a b i l i t y .
Rn-d particles
years. 1
low
i t
h a s made
modelling,
-symmetrical
lung
significant
based
i n i t i a l l y
model,
has
been
***)
URANIUM MINERS
(1 - 19)
NON-URANIUM MINERS
(0 - 9 )
[WLM]
POPULATION/ CUMULATIVE EXPOSURE
NO. OF LUNG CANCER CASES
NORWAY
>
Figure 1. Observed vs. expected number of lung cancer cases f o r miners exposed to low-levels of radon daughters.
ALL SMOKING MINERS
.EXPECTED
.OBSERVED
NEWFOUNDLAND
URANIUM MINERS WITH NO PRIOR GOLD MINING EXPERIENCE
n3 URANIUM MINERS FREQUENCY PER 10"
NUMBER OF LUNG CANCER CASES
436
RADON AND ITS DECAY PRODUCTS
refined lungs
to
present
(Weibel,
artefacts
are
dimensional tion)
i s
for a)
results of
Schum,
made
1980).
the information
(e.g. branching
target be
from
Thereby on
angles,
i n
human
potential
vivo
three-
bronchial
i n c l i n a -
for
from
the
lung
dose
conversion
modelling
of
s t i l l
Rn-d exposure
show
to
a
dose
reasons:
geometry: cells
the
below
i n
the bronchial
sensitive
deposited dose
obtained
values
following
basal
i n
the
target
received
region
for
the airways. bronchial
information b)
and
geometry
on thorax-casts
and
maintained.
range
the
based
Yen
minimized
lung
However, large
models
1963;
lung
c e l l s
on the depth
determines from
to
inhaled
distribution
considered
induction
The assumed depth
surface
by these
are generally
cancer
of
this
a large Rn-d.
i s
due to
to
Rn-d
c e l l
layer
extent
the
Physiological
scarce.
unattached/attached Rn-d: The
nature
attached of
and behavior to
controversy,
environmental presence are
of
exposure
particularly gases
t h e most
conversion.
"free"
This
with
(Busigin
can
regard
be o f
with
et
or atoms,
nuclei,
conditions,
c r i t i c a l
situations
Rn-d ions
condensation
atmospheric
other
one o f
of
atmospheric
i s
to
such
a l .
the
influence humidity
Free
parameters
for
particular
importance
a
large
the
ratio
not
subject
as
1981).
t
i . e .
s t i l l
of and
R n - d atoms
exposure-dose
of
i n
indoor
unattached
to
attached Rn-d. Figure and is
2
shows
the
dosimetric assumed Due
gical
(OECD/NEA,
to
the
and a
situations
Animal have
extensive using
and
studies. out
and
respectively. relevant
to
lifetime
of
Only the
a member
of
from
these
can
be
that
the
the
100 WLM.
lifespan
Rn-d. man lung
as
However, i t
has to
cancer
i s
These the i n
biological,
physiolo-
2 to
indicated effects
the
published
1 2 0 mGy
that
for
can reduce
per
WLM.
most
this
situations
hamsters, from
6^0 to
2
exposure
indoor
range
to
occurring
tumor data
have
order
to
most
exposure
for
up to
rats
out
et about
3
shows
i s
of
between
corrected
for
extrapolate
these
findings
susceptibility species.
and
1984), WLM
during
tumor
to
It
4%
at
shortening
lung
cancer from
Rn-d
Furthermore,
i s the
lung
2% a n d
than
two
100
the
other
the
rats,
Rn-d exposure.
causes
the
most
on
a l . .
experienced
Figure
for
studies
The two
8 0 0 0 WLM ( C h a m e a u d
situation
not been of
be assumed t h a t similar
carried
as a function
rate
result
were
20 to
inhalation
and mice.
5 6 0 WLM ( C r o s s
the public. studies
Rn-d
rats
investigations
the low-level
indoor
obtained
about
from
indoor
levels
from
rate
seen
situation
modelling,
Experimental
on dogs,
comparable
1984)
physiological
exposure
1983).
mean R n - d e x p o s u r e
Lafuma,
the
other
i n
range
compensatory
(OECD/NEA,
carried
two d i f f e r e n t
standardized
various
analysis
10 mGy/WLM f o r
inhalation
been
for
a
used
factors
sensitivity
to
of
parameters
conversion
about
frequently
results though
superposition
exposure 5
of
even 1983).
physical
exposure-dose However,
spread
models,
of
from
rat
to
induced i t
i s
31.
STEINHÀUSLER
437
The Validity of Risk Assessments CALCULATIONS BASED ON: YEH-SCHUM PHYSIOLOGICAL MODEL AND
JACOBI-EISFELD
DOSIMETRIC MODEL WEIBEL - A - PHYSIOLOGICAL MODEL AND JAMES-BIRCHALL
DOSE [mGy/WLMj
DOSIMETRIC MODEL 200
150
100
50
BREATHING RATE RESTING
LIGHT ACTIVITY
jrn3 /hj
H Figure
2.
Mean b r o n c h i a l
1 WLM e x p o s u r e physical librium
1
to
free
a c t i v i t i e s factor
dose
radon
to
basal
cells,
222 daughter
(assuming t y p i c a l
atoms
indoor
standardized f o r
for
different
exposure
with
equi-
F = 0.4)·
TUMOR RATE
%
15
10
100
200
Figure
3.
tion
Rn-d exposure.
of
Primary
300
e p i t h e l i a l
400
lung
500
tumor
600 700 MEAN EXPOSURE |WLMJ
rate
i n rats
as a
func-
RADON AND ITS DECAY PRODUCTS
438
questionable applies far
i t
of
a
Rn-d
i n
whether animals
has been similar
(Cross
models
a l . .
(Hofmann
and doses
tive
Animal influence
i n
cancer
to
on the lung
results
where
the
(Cross
et
Assessment
Rn-d
risk i s
latent Rn-d
period,
low
phase
up to
The
from
i n
with
more,
in
The studies
of
i s
animal following
compare
other
only
cocarcino-
to
the
i s
effec-
to
i t s
over
do n o t
relative
w i l l i n
with
i s
of
be
the
lung
of
This
to the
necesa l l
epidemiological this
occurs the
important
at
typically
latent
period
20 y e a r s
their
i s
observed
Rn-d exposures
during
used risk
input
the
early
to
assess
model,
i s
often
are
asso-
data
used
uncertainties.
to i n
(BEIR-II,
assess
the
the wide
an order
of
Rn-d exposure
population 1980).
range
of
The
cancer values
(Figure
dosimetry c a n be
the
Further-
lung
magnitude
by epidemiology,
indoor
to
include
the Rn-d exposed
used
i n Rn-d
exposure
to
a l l
model
the
reflected than
and
press).
same r e s u l t
the data
provided to
the
long
10 t o
absolute
follow-up
b y more
regard
i n
after i s
length
f u l f i l l
the
a
fumes
found
lifetime
a few years
risk
because
this
expression.
received
of
v s .
smoke
on the
that
unquantifiable
give
Rn-d inhalation
information
to
probably
mathematical
to
rate; was
on
1982).
exposure
l i k e l y
(Steinhâusler,
complete
differing
i s
diesel
The assessment of
tumor
smoke
indicate
smoke
rate a l . ,
Therefore
a period
importance
models
exposure cancer
low-level
as compared
and often
of
i t
and
alternating
of
which
clearly
ore dust
be s u f f i c i e n t l y
cases.
who most
of
not
induction
simultaneous
and cigarette
do
rate,
from
study
by the assumption
Therefore
uncertainties
from
to
a l l
region
the lung
chronic
studies
i . e .
case
published,
Whilst
by a period
mining
large
both
multiple
so
inhaling
the
c r i t i c a l
fumes
Exposure.
period
rates
choice
ciated groups,
the used
cancer
a
cancer
miners,
risk,
overrated
j u s t i f i e d
results
cancer
Indoor
4-0 y e a r s
uranium
lifetime
risk
lung
exposure
of
the relevance
1 9 7 8 ; Chameaud e t
t
determined
rates.
uranium
high
i n
Non-occupational Rn-d indoor
increased at
for
are truncated
exposure
with
lung
followed
induced
criterion.
i s
diesel
increases
a l .
the follow-up
studies
and animals
neglecting
been
obtained
resulting
largely
sitates
also
effect.
exposure
exposure,
cancer
of
out
( L C ) p e r WLM
demonstrated
i t
i . e .
However,
synergistic
decreasing Risk
man
the dose
have
induction.
Rn-d
contrast
risk
cancer
and dose
carried
induction.
no e f f e c t
i n i t i a l
only
lung
i f
cases,
Rn-d and ore dust,
potentially showed
i n
exposure
study
An a n a l y s i s
1983):
both
between
the animals
Rn-d exposed miners
1982).
studies
of
for
the
inhalation
then
tumor
From
that
Daschil,
effects, i n
lung
concluded
Rn-d
radiation genic
same r e l a t i o n s h i p
magnitude
et
for
the
and man.
and
4 ) ·
animal
summarized
as
follows : 1)
Physical
Rn-d),
parameters
(e.g.
the with
the
most of
Cigarette induction.
of
breathing
smoking
5
of
rate)
conversion
frequently
about
the a i r
ventilation
exposure-dose
exposure 2)
characteristics
unattached
for
inhaled t h e room
(e.g. and
can influence the
occurring
individual
fraction
of
physiological significantly inhabitant,
value
for
indoor
promotor
for
lung
Rn-d
mSv/WLM. may
act
as a
cancer
31.
STEINHÀUSLER
439
The Validity of Risk Assessments
RISK FACTOR CANCER/PERSON* WLM -4 x 10 V Ί I 1
Figure cancer
4. risk.
Published lifetime
risk
factors
for
Rn-d related
lung
440
RADON AND ITS DECAY PRODUCTS
3)
Up to 100 WLM of cumulative exposure, lung cancer risk shows linear dependence on Rn-d exposure; this is followed by a plateau region up to about 500 WLM. Due to the various sources of uncertainties associated with both groups of input data (dose and effect) needed for risk assessment, at present it is only possible to give an upper limit for the pure Rn-d related cancer risk. Using the miner data for this purpose may be overestimating the risk due to indoor exposure, since it includes risks from simulta neous exposure to external gamma radiation, long lived alpha emitters. This effect, however, may be outweighed by the fact that smokers amongst miners are generally twice as frequent as the number of smokers among members of the public. Therefore risk estimates derived from miners may only reflect the risk of male smokers exposed to Rn-d but may not be valid for children and women. Altogether it can be assumed that such a risk factor obtained from miners represents an upper limit for the risk associated with indoor exposure. Another possibility is the use of lung cancer incidence amongst non-smoking members of the public, representing the maximum value that can be attributable to Rn-d only. Because of statistical problems with a rare event, such as lung cancer amongst non-smokers, estimates of annual rates range from about 20 to 300 cases per 10 men depending on the age (Steinhàusler, in press). A third possibility consists of comparing the theoretically calculated lung cancer rate based on risk coefficients derived from miners with the actual cancer occurrence among non-miners, derived from Rn-d exposure assessment in dwellings and using appro priate exposure-dose conversion factors (Steinhàusler et al. 1983; Edling, 1983). From the above it can be concluded that the risk for lung cancer induction from chronic indoor exposure to Rn-d is unlikely to be higher than 1.10~4/mSv. i order to understand the magnitude of this risk it has to be emphasized that man can be exposed to a multitude of different hazardous materials in the indoor atmosphere besides Rn-d, such as formaldehyde, nitrogen dioxide, carbon monoxide, nitrosamines, polyaromatic hydrocarbons, volatile organic compounds, asbestos and pesticides (Gammage and Kaye, 1985). t
n
Literature Cited Busigin,A., Van der Vooren,A.W., Babcoock, J.C. and C.R. Phillips, The Nature of Unattached RaA (Po 218) Particles, Health Phys.40: pp 333-344 (1981). Chameaud,J., Perraud,R., Lafuma,J. and R. Masse, Cancers Induced by Rn-222 in the Rat, In: Proc. of Spec.Meet. on Ass, of Radon and Radon Daughter Exposure and Related Biological Effects (G.F. Clemen te, A.V. Nero, F. Steinhäusler and M. R. Wrenn, eds) p. 198, R.D. Press, Salt Lake City (1982). Chameaud,J., and J. Lafuma, Experimental Study of Cancer Induction with Rn 222. Daughters, Rad.Prot.Dosimetry 7: 385 (1984).
31. STEINHÀUSLER
The Validity of Risk Assessments
441
Congress of the United States,Joint Committee on Atomic Energy, Part I and II (1967). Cross,F.T., Palmer,R.F., Filipy,R.E., Busch,R.H. and B.O. Stuart, Study on the Combined Effects of Smoking and Inhalation of Uranium Ore Dust, Radon Daughters and Diesel Oil Exhaust Fumes in Hamsters and Dogs, Pacific Northwest Laboratory Rep. No. PNL-2744, Richland, National Technical Information Service, Springfield, VA., USA (1978). Cross,F.T., Palmer,R.F., Dagle,G.E., Busch,R.H., and R.L. Buschbom, Influence of Radon Daughter Exposure Rate, Unattached Fraction and Disequilibrium on Occurrence of Lung Tumors, Rad.Prot.Dosimetry 7: 381 (1984). Cross,F.T., Palmer,R.F., Busch,R.H., Filipy,R.E., Dagle,G.E. and B.O. Stuart, Carcinogenic Effects of Radon Daughters, Uranium Ore Dust and Cigarette Smoke in Beagle Dogs, Health Phys. 42: 33 (1982). Edling,C., Lung Cancer and Radon Daughter Exposure in Mines and Dwellings: study no. V. Linköping University, Medical Dissertation No. 157, Dept. of Occup. Med., Linköping, Sweden (1983). Evans,R.D., Harley,J.H., Jacobi,W., McLean,A.S., Mills,W.A. and C.G. Stewart, Estimate of Risk from Environmental Exposure to Radon-222 and its Decay Products, Nature 290: 98 (1981). Gammage,R.B.and S.V. Kaye (eds), Indoor Air and Human Health, Lewis Publ. Inc., Chelsea, MI., USA (1985). Härting,F.H., und W. Hesse, Der Lungenkrebs, die Bergkrankheit in den Schneeberger Gruben, Teil I, Eulenbergs Vierteljahrschriftf., Gerichtliche Medizin und öffentliches Gesundheitswesen, neue Folge 30: 296 (1879). Hofmann,W., and F. Daschil, The Relevance of Animal Models for Radionuclide Inhalation in Man, in Current Concepts in Lung Dosime try, Proc. of the 30th Annual Meeting of the Radiation Research Society at Salt Lake City 1982 (D.R. Fisher, ed) pp 95-102, Utah (1983). International Commission on Radiological Protection, Limits for Inhalation of Radon Daughters by Workers, ICRP Publication, No.32, Pergamon Press (1981). Lundin,F.E., Wagoner,J.K. and V.E. Archer, Radon Daughter Exposure and Respiratory Cancer, Quantitative and Temporal Aspects, Nat. Inst. Occup. Safety and Health/Nat. Inst. Env.Sciences, Joint Monograph No.1, US Dept. of Health, Education and Welfare, Public Health Service (NTIS, No. PB 204871), Washington, D.C. (1971). Masse,R., Histiogenesis of Lung Tumors Induced in Rats by Inhalation of Alpha Emitters: an Overview, in Pulmonary Toxicology of Respirable Particles (C.L. Sanders et al., eds) p. 498, CONF - 791002, National Technical Information Service, Springfield, USA (1980).
442
RADON AND ITS DECAY PRODUCTS Muller, J., Wheeler, W.C., Gentleman, J.F. Suranyi, G. and R. Kusiak, Study of Mortality of Ontario Miners, Part I, Rep. Ontario Ministry of Labour,Ontario Workers' Compensation Board, Atomic Energy Control Board of Canada (1983). OECD - Nuclear Energy Agency, Dosimetry Aspects of Exposure to Radon and Thoron Daughter Products, ΝΕΑ Experts Report ISBN 9264-12520-5, OECD, Paris (1983). Radford,Ε.P., and K.G. St.Clair Renard, Lung Cancer in Swedish Iron Miners Exposed to Low Doses of Radon Daughters, The New England Journal of Medicine, 310:1485(1984). Royal Commission,Report of the Royal Commission on the Health and Safety of Workers in Mines (J.M. Ham, F.R. Hume, C.C. Gray, A.L. Gladstone, J. Beaudry, E.A. Perry, R.P. Riggin, eds), Ministry of the Attorney General, Ontario, Canada (1976).
Snihs,J.O., The Approach to Radon Problems in Non-Uranium Mines in Sweden, in Proc. Third Int. Congr. Prot. Assoc. p 909, (IRPA), U.S.A.E.C. Conf. 73097-P.2. (1973). Steinhäusler,F. and Ε. Pohl, Lung Cancer Risk for Miners and Atomic Bomb Survivors and its Relevance to Indoor Radon Exposure, Radiation Protection Dosimetry: Vol.7, No.1-4: 389-394 (1983). Steinhäusler,F., and Pohl,Ε., Lung Cancer as a Result of Energy Conservation and Waste Recycling, Proc. XI. Regional Congress of IRPA, Vol.1: pp 258-262, Vienna, Austria (1983). Steinhäusler,F., Hofmann,W., Pohl,Ε. and J. Pohl-Rüling, Radiation Exposure of the Respiratory Tract and Associated Carcinogenic Risk due to Inhaled Radon Daughters, Health Physics Vol.45, No.2: 331-337 (1983). Steinhäusler,F., The Epidemiological Evidence for Health Risks, in Radon and its progeny in indoor air, (A. Nero et al., eds), CRC -Press, (in press). United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), Sources and Effects of Ionizing Radiation, Report to the General Assembly with Annexes, United Nations, N.Y., USA, (1977). US-Environmental Protection Agency, Draft Environmental Impact Statement for Remedial Action Standards for Inactive Uranium Proces sing Sites (40 CFR 192), US-EPA Rep. No. EPA 520/4-80-011 (1980). US-National Academy of Sciences, National Research Council BEIRII, The Effects on Populations of Exposure to Low Levels of Ionizing Radiation, Washington, DC, USA (1980). US-National Academy of Sciences, National Research Council BEIR-II: Biological Effects of Ionizing Radiation, Washington, D.C., USA (1972).
31. STEINHÀUSLER
The Validity of Risk Assessments
443
Weibel E.R., Morphometry of the Human Lung, Springer Verlag, Berlin (1963). Williams, L. R., Labour viewpoint, in Proc.Int.Conf. on Occupat. Rad. Safety in Mining (H. Stocker, ed) 29, Canadian Nucl. Assoc., Toronto, Canada (1985). Yeh,H.C., Schum, G.M., Models of the Human Lung Airways and their Application to Inhaled Particle Deposition, Bull.Math.Biol. 42: pp 461-480 (1980). RECEIVED September 9, 1986