&EPA
United States
Environmental Protection
Agency
Office of
Radiation Programs
Washington, D.C. 20460
EPA 520/1-87-030
November 1987
Radiation
Report of EPA/DOE
Roundtable Discussion
of Radon Research Needs
-------
Report of EPA/DOE Roundtable Discussion of Radon Research Needs
April 17-18, 1986
Held at the Environmental Measurements Laboratory
Nev York, New York
Report Prepared by
Philip K. Hopke
Institute for Environmental Studies
University of Illinois at Urbana-Champaign
Urbana, Illinois 61801
-------
Introduction
Expanded research programs into the potential impacts on public
health by naturally occurring radon and its decay products are being
developed in the United States. This increased effort is in part a response
to the discovery of unusually high levels of indoor airborne radioactivity
found in houses in the Reading Prong area of Pennsylvania and New Jersey.
In several houses in this region, the radon decay product concentrations
were substantially above values that are permitted to be present in active
uranium mines. Other surveys in the U.S. and elsewhere are discovering
that many houses have higher than anticipated radon and decay product
concentrations although generally not to the extremes seen in the Reading
Prong region. Thus, the possibility exists that a significant fraction of
the United States population is being exposed to concentrations of radon
and decay products that are currently considered to pose an increased
health risk.
During the past several years, there have been a number of research
efforts to determine the concentration of radioactivity in indoor air, to
relate that activity to the properties of the structure and its geological
setting, to determine the fundamental physical and chemical properties of
radon and its decay products, and their impacts on human health as well
as determining ways to mitigate against the exposure of persons living in
houses with initially high radon levels. A major symposium during which
the results of a number of these research efforts were presented was held
on April 14-16, 1986 at the 191st National Meeting of the American Chemical
Society. Since a large fraction of the researchers in the area of radon
and its decay products were presenting the results of their studies at
this meeting, a roundtable discussion was organized at the Department of
Energy's Environmental Measurements Laboratory to review the status of
our understanding of the properties and effects of radon and its progeny
and to review the scope and directions of the research programs being
planned by the U.S. Environmental Protection Agency and the U.S. Department
of Energy. The purpose of this report is to summarize those discussions.
-------
Current Research Areas
This report cannot fully document the full state of knowledge regarding
radon and its decay products. The volume to be published containing the
ACS Symposium papers (1) will provide considerable information as will
the forthcoming book edited by Nazaroff and Nero (2). The proceedings of
a symposium held in Maastricht, the Netherlands in March 1985 have recently
been published (3) as have the proceedings of a specialty conference on
indoor radon organized by the Air Pollution Control Association in February
1986 (4). The objective here is to examine a limited number of topics
related to the occurrence of high levels of radon in indoor air, the
chemical and physical properties of radon and its decay products, the
health effects of this airborne radioactivity, and the adequacy of the
currently planned research programs to resolve the unknown or uncertain
areas of knowledge critical to evaluating the health risks to the general
population.
Occurrence of Elevated Radon Concentrations
It is now generally agreed that the principal source of indoor radon
is the soil under a structure. The radon in the soil gas flows into the
building because of a pressure differential between the inside of the
building and the soil. In particular locations, there may be increased
indoor radon from radon dissolved in the potable water brought into the
house. There is also a small contribution from building materials although
in the case of certain materials such as phosphogypsum, there can be a
substantial contribution to the indoor radon levels. However, the most
common source of indoor radon is the soil gas under the structure. The
input of radon to the building depends in part on the radium concentration
in the soil, the permeability of the soil and the total volume of soil
gas that can flow into the house. The ability to predict where high levels
of indoor radon are likely to occur would permit the focussing of monitoring
and mitigation efforts on only those areas that pose a substantial threat
to the inhabitants. Thus, the first topic of discussion was the status of
our ability to make such predictions based on geological or geographical
-------
considerations. A number of surveys have been made in Europe, Canada,
Japan, and to a limited extent in the United States. Thus, the question
is how well we can utilize the results of these surveys to predict the
incidence of elevated indoor radon levels.
The factors that govern the observed levels are related to the
structural properties of the building as well as the soil characteristics.
Thus, it is important to consider the understanding of both the physical
properties of the building and the soil that drive the radon infiltration
process. The recent work by the Indoor Air Quality and Ventilation Program
at Lawrence Berkeley Laboratory mapping pressure fields and investigating
migration rates using inert tracers is starting to build our knowledge of
the process. During the past several years, there has been the development
of a qualitative appreciation of the factors that contribute to radon in
the soil and to ingress of radon into houses. However, we are now just
beginning to be able to put that understanding into a framework that may
lead to a predictive capability. This understanding is being developed
through the use of experimental houses that allow the development of
improved building science. These kinds of studies of the relationships
between design, construction and infiltration rates need to be continued.
Similarly there are similar research needs in soil science. Although
it is known that radium content and soil permeability are important factors
for radon availability, it is not possible on a large scale such as the
United States as a whole to dependably predict where the high concentrations
are likely to be found. It may be possible to predict regions of higher
potential radium content or soil permeability based on information such
as the NURE maps or from data from the Soil Conservation Service, and
thus identify areas that might be suspect as potential higher radon areas.
There is a need to test whether the available data are sufficient to make
such predictions. The recent study of Spokane, WA, reported at the Symposium
(5) has found high indoor radon in this area of high permeability soil.
Thus, in some areas the surficial geology may be sufficiently homogeneous
that a lognormal distribution of radon values could be expected. However,
other areas are so geologically heterogeneous that it would be impossible
-------
to predict that the distribution of values would conform to a single
distribution. Other tests of the ability to identify high risk areas
based on soil properties are needed and appropriate in order to test
hypotheses regarding the geological properties and housing characteristics
that might be expected to result in high radon levels.
From the results to date from a variety of surveys, it appears unlikely
that it will possible to make predictions for any particular building
since nominally similar buildings on nominally similar soils have indoor
radon concentrations differing by orders of magnitude. It may be that
further research into both building and soil science will reduce the
uncertainty of prediction, but such improvements are not likely in the
immediate future to provide building by building predictions.
The results of surveys generally report lognormal distributions and
a variety of summary statistics including the median, the geometric standard
deviation and the arithmetic mean. However, different groups of building
may result in quite different distributions. For example, the national
survey in the United Kingdom resulted in one distribution while a local
survey in Cornwall resulted in another lognormal distribution but at a
substantially higher median value. The discussion suggested that reporting
the parameters of a lognormal distribution appears to be the best
representation of the findings in terms of a continuous probability
distribution function to estimate the number of buildings with radon
exceeding a given level. It was suggested that for exposure considerations,
it is the arithmetic mean that is the critical parameter since the
population-weighted average exposure will be the arithmetic mean. Thus,
an arithmetic mean is required for any risk assessment even though the
geometric mean and geometric standard deviation are reasonable to
characterize the distribution.
Another source of indoor radon is radon in water and a point of
discussion was the current level of understanding on the contribution of
water to airborne radon concentrations for a given radon concentration in
solution. Calculations have been completed at LBL which included the amount
-------
of water use, the actual release efficiency for radon from water, the
volume of houses and the distribution of ventilation rates. Distributions
of values were estimated for each of these factors and then they were
combined to yield a distribution of indoor radon values per unit water
use resulting from radon in water. The arithmetic average of this
distribution is quite close to the 10"^ ratio of airborne to waterborne
radon that has been observed by a number of investigators. The geometric
standard deviation for the distribution is of the order of 2 resulting
from the distribution of ventilation rates. Applying this distribution to
the distribution of waterborne radon that has been measured, the estimated
contribution to radon concentrations can be estimated.
For those houses using public water supplies derived from surface water,
there is essentially no contribution to indoor radon. For houses using
public supplies derived from wells, the average contribution to indoor
radon is approximately 3%. For private wells the values may be much higher
and not generally known. In some circumstances the contribution from well
water may be comparable to the radon present due to soil. There have been
some building materials used in Europe and the United States that yield
high indoor radon levels such as the Grand Junction tailings concrete.
The consensus was that normal building materials are important contributors
to the average radon levels particularly when soil gas radon concentrations
are low, but generally are not responsible for high indoor radon levels.
Building materials may also be more important in multilevel, multifamily
housing where the upper floors are relatively isolated from soil gas
influences.
In order to assess the population exposure, it will be necessary to
make measurements in individual houses since it is not currently possible
to predict the indoor radon concentrations. There have been a variety of
prior survey programs. There arises a sampling strategy question of whether
the measurements should assess the average dose to the general population
or find the high radon houses where a more acute threat to the residents
may exist. These objectives require different survey designs. At the
present time, it is not certain how to identify the high radon level
-------
houses. As previously discussed, there exist areas where it might be
expected that high radon level houses exist and sampling strategies are being
developed. With the effort that has been made to determine areas of near-
surface uranium deposits, it is likely that any potential Reading Prong
area can be identified from such data. Areas of high soil permeability
may also be identifiable from soil maps, but it may prove to be much more
difficult to be certain in these cases. However, there is now a need to
test these concepts to determine their efficacy in making predictions
regarding the likelihood of elevated radon and hence a greater need for
monitoring in that region of high indoor radon potential.
Properties of Radon and Progeny in Indoor Air
Several parameters are needed to describe the concentrations of radon
decay products, their relationship to the radon concentration and the
amount of activity attached to the ambient particulate matter. These
parameters include the working level, WL, or the potential alpha energy
concentration, PAEC, the equilibrium factor, F, and the "unattached"
fraction, f, or the "unattached" fraction of alpha potential energy,
fpOt. In the reports presented at the symposium, a variety of values
were presented for F in contrast to the use of 0.50 as the typical F
value. For example, the average value in the U.K. assessment was 0.35
(6) while in Norway, a value of 0.50 was used (7). In Germany a value of
0.30 has recently been reported (8). Since most of the large scale
monitoring efforts only measure radon, the exposure to the decay products
is estimated from the F value and higher or lower values result in larger
or smaller WL or PAEC values. One of the key factors that affects the F
value is the particle concentration since the F value increases as the
particle concentration increases.
The typical indoor environment in the United States generally includes
smokers and about half of the U.S. homes cook with gas stoves. There are
then major indoor particle sources in the normal U.S. indoor environment.
The apparently decreasing values seen in some of these recent European
studies may result from decreasing indoor particle concentrations. It may
-------
also come from measurements in the bedroom where length of occupancy is
large and there are relatively lower particle levels. As the particle
concentrations decrease, the amount of "unattached" activity increases.
Since some of. the dose models put increased weight on this activity in
terms of dose deposited in the lungs, the effects of increasing f and
decreasing F may counterbalance one another. Vanmarcke (9) presents dose
calculations based on measured F and f values and the dose model of James
and Birchall (10) indicating just such a result. Thus, in effect, the radon
concentration becomes a sufficient measure of exposure to permit dose
calculations.
However, we are then in effect adopting as certain the validity and
accuracy of this particular dosimetric model. There are others for which
this result does not hold. For example the NCRP (11) developed a guideline
based on a particular F value and changes in the value would directly
affect that guideline value. It is likely that we will discover later
that none of the now existing dose models fully account for all of the
important phenomena. Thus, although it is attractive to think that only
an integrated radon measurement is needed to assess exposure, it would
seem wiser to find an appropriate long term integrating measurement of
the decay product concentrations and the size distributions of the activity.
Then, as health effects are identified in the population, the
exposure/dose/effects relationships can be evaluated both from the radon
concentration and the decay product levels. McLaughlin (12) has reported
an effort to develop such a monitoring system for PAEC, and it would
appear that a system that yielded long term integrated measurement of
both radon and PAEC levels would be a valuable addition to our monitoring
techniques.
Physical and Chemical Properties of the Radon Progeny
Reports presented during the Symposium (13-15) have shown new techniques
for measuring the properties of the radon decay products particularly the
activity size distribution of the "unattached" fraction and for interpreting
the ultrafine particle size activity in terms of classical cluster formation
-------
theory. There is interest in these highly diffusional particles from the
viewpoint of understanding the atmospheric chemistry of the radioactivity
immediately following the radon decay. They may also be important in the
dose models depending on whether the dose to a specific portion of the
bronchial tree is being assessed. All of these ultrafine particles will
deposit in the bronchial region. If the dose models only consider the
average dose to the bronchi, then where they specifically deposit is
irrelevant. However, if the dose to a particular generation is being
considered, then there will be a considerable difference in deposition
between 1 and 5 run particles. Thus, there is a critical need for
measurements of the activity size distributions for chemical investigations
of the indoor atmosphere and such measurements may be important for assessing
the dose depending on which dosimetric model is employed.
The improvements in the measurement of the ultrafine fraction and
the initial results of classical cluster formation theory suggest that
significant improvements in our understanding of the physical and chemical
behavior of the radon progeny are possible and that careful experimental
and theoretical studies of these problems have a high probability of
success. It is not too early to begin to consider the development of an
indoor air quality model involving a variety of reactions initiated by
radiolysis, photolysis and combustion processes. In addition to the
chemistry, it will be necessary to include the airflow, ventilation, and
infiltration patterns within a building. It will take a substantial effort
to add the additional knowledge that will be necessary before such a
model becomes a useful tool in understanding indoor air quality, but it
would be worthwhile to begin such an effort as part of the overall research
program on radon and its decay products. It will also be important for
this effort to move forward in conjunction with those researchers who are
studying the variety of other indoor air quality questions. Radon and its
decay products will interact with the other materials present in the
indoor air and examining the radon-related questions in isolation from
the other pollutants will fail to produce the needed level of understanding
of indoor air quality.
-------
There is agreement that the generally used measurement methods for
"unattached" activity are quite crude. Although there are differences in
the different dose models, all of them give additional importance to this
highly diffusional mode and the ability to better define precisely what
is meant by the "unattached" activity was considered to be a useful goal
for additional study. It is not yet clear what specific size value would
best represent the upper limit of the "unattached" activity. However,
the reporting of specific size cut-offs (50% collection points) is strongly
encouraged for all "unattached" fraction measurements.
In houses where air cleaning measures are in use, the effect of such
cleaning will be the reduction in the radon progeny concentrations. However,
the particle concentration will also be reduced leading to a substantial
increase in the "unattached" fraction. Since this activity contributes
much more dose per unit concentration than the attached activity, reduction
in the decay product level may not lead to much reduction in the dose to
the occupants. In these cases, it is important to measure the radon and
estimate the exposure using the much larger fpOt value.
Dose Models
The objective of dose models is to relate the measured or estimated
concentration of radon and/or its decay products to the energy actually
deposited in the tissue of the respiratory tract. There are thus a number
of issues that have been discussed that are deeply interrelated to the
dose models including the equilibrium fraction and the amount and size
distribution of the "unattached" fraction. Dr. B.S. Cohen (16) presented
new results on the deposition of particles down to 0.04 /*m diameter in
lung casts that indicate that the predictive model previously employed to
estimate the deposition underestimates the deposition velocities by a
factor of 2. The results do show that a diffusional framework is generally
correct. However, at the Reynolds numbers typical of respiratory tract
flow which are in neither the laminar nor the turbulent flow regimes,
further theoretical and experimental studies of the details of the
depositional processes are needed.
-------
10
There are distinct differences in the incorporation of the bronchial
deposition in the dose models. As previously discussed, the James-Birchall
model assumes an average dose to the bronchial region. The Jacob!-Eisfeld
model (10) and NCRP (17) consider the differential dose to the segmental
bronchi that depends on the preferential deposition of the unattached
activity in the early generations. In these later models, the dose is
dependent on both F and fpOti while in the James-Birchall, the dose is
relatively insensitive to these almost compensating factors. The results
of these differences yield different doses depending on the particle size
distribution and the amount of airborne particulate matter present. There
are also problems in assessing the dose because of the changing size
distributions of hygroscopic particles. This effect is not as critical to
dose estimation for accumulation mode particles nor in the average dose
approaches, but can cause factor of 2 changes in the segmental bronchial
dose for a factor of 3 increase in size of a 1 run particle. Thus, it is
important to also have an understanding of the behavior of radon daughter-
bearing particles under high humidity conditions.
To then examine the risk from the deposited activity with any model,
it is necessary to know the "unattached" fraction, the activity median
diameter, and a weighting factor for dose equivalent to bronchial tissue.
For the James-Birchall model and a choice of fpot of 5%, a median diameter
of 0.12 /im, and a weighting factor of 0.06, the effective dose equivalent
is about 15 mSv per WLM which is a factor of 3 higher than NEA (10)
recommends or other corganizations have incorporated in their estimation
of the fraction of total radiation dose that an average person receives from
radon decay product exposure. If the particle size doubles because of
its hygroscopic nature, this dose equivalent reduces to 10 mSv/WLM.
However, there is experimental evidence that the hygroscopic increase in
size is actually closer to a factor of 4 than 2. Thus, it may be important
to distinguish hygroscopic from hydrophobic particles although it would be
difficult to incorporate such differences into the models on a generalized
basis. It was suggested that in spite of all these considerations, there
is really a fair degree of uniformity to the calculated dose and that it
-------
11
only varies by a factor of 2 over the various lung generations unless
there is an extremely high "unattached" fraction.
A question was raised as to the use of microdoslinetrie methods over
the commonly applied absorbed dose calculational models. Microdosimetry
says that at low doses, the absorbed dose is not meaningful in determining
the biological effects. It is necessary to know the number of cells that
are hit and the probability that a hit will induce a change in the cell
leading to cancer induction. Most cells will not be affected but those
that are may receive 50 to 60 rads. It thus may be more useful for the
estimation of risk factors to consider microdosimetric methods. However,
the dose-response functions that we use are a function of the type of
dosimetry used to obtain them. It is important not to use microdosimetry
to estimate the dosage to a given cell and then use dose-response curves
developed on the basis of uniform dose to estimate the risk of malignancy
development. It was generally felt that the traditional dosimetric
approaches will yield reasonable estimates of the risks since consistent
dose-response curves have been developed within this framework.
In evaluating the effects of radon decay products, it is not yet
clear whether a relative risk or an absolute risk model applies. Some
data seem to favor an absolute risk while other results support a relative
risk approach. However, any of the models have to consider such factors
as latency interval, age at which cancer appears, etc. Recent unpublished
results on U.S. uranium miners (smokers and nonsmokers) seem to suggest
that the risk of cancer induction decreases with time after leaving the
high exposure mining environment. Thus, neither absolute and relative risk
models apply directly and both have to be modified in order to match the
available data on radon decay product induced effects.
There was general agreement that current measurement programs to
quantify health risks should focus on radon and not the decay products since
reliable monitors, calibration, and quality control procedures are not
yet available for long term, integrated measurements of the decay products
as are the available track-etch detectors for radon. Until there is an
-------
12
inexpensive, efficient and logistically manageable method for decay product
measurements integrated over periods of up to a year, radon measurements
must be used to assess exposure. The decay product level cannot be high
unless there is high radon present so the use of radon monitoring will
identify the population at greatest risk. The details of the actual exposure
of the individuals to the daughters can now only be estimated. Thus, an
integrating potential alpha energy concentration detector would be an
extremely useful development for assessing population exposure in
epidemiological studies and in testing dosimetric models.
A monitoring question that has not been fully addressed is that of
thoron and its decay products. It has been common to assume that because
its short half-life results in a sufficiently short diffusion length, thoron
does not penetrate into houses in any significant amount. However, a
recent report by Schery (18) suggests that thoron may contribute a
significant amount in terms of potential alpha energy concentration, and
more long-term monitoring of the impact of thoron and thoron decay products
will be necessary in order to fully assess exposure to airborne
radioactivity.
DOE/OHER Radon Research Plans
The objective of the discussion was to review the general directions
that the Office of Health and Environmental Research of the U.S. Department
of Energy has developed for radon research over the next five years. The
overview of these plans has been presented by Lowder (19). The goal of
the program is to develop the quantitative data and principles that will
allow accurate assessment of radon exposure and associated lung cancer
risk under environmental conditions. OHER has a particular concern with
regard to public health risk associated with possible future trends in
indoor radon exposure related to the application of advanced energy
conservation technology in new housing. Thus, a comprehensive research
program addressing a variety of basic physical, chemical and biological
processes is needed to provide the basis for the needed risk assessments.
-------
13
Radon Availability and Transport within Houses
The first goal of the research program is the development of a model
for radon availability and transport into the indoor environment. Such a
model would permit the identification of areas where houses would have a
high risk of having elevated radon levels and the evaluation of potential
control technologies for reducing the radon entry rates. It would be
desirable to develop a set of variables that can be readily measured in a
house and using those values, be able to predict the radon levels in that
particular structure. These diagnostic tools would also permit the
development of a cost-effective mitigation strategy for high exposure
situations. The question would then be what are the key areas of research
that are needed to move forward toward this goal.
There is a need to develop a standardized field measurement procedure
to assess radon availability in soil gas so that inexpensive and easily
used methods could be used and compared to a standard condition of known
radon potential. Laboratory studies may be needed to help determine the
critical variables to measure and how best to control the field conditions
to obtain reproducible measurements of the critical parameters. To
understand the transport of the radon into structures, it is necessary to
consider that transport both in undisturbed and disturbed soils. The
construction of a house does have an impact on the soil structure that
must be explicitly considered and understood. By measuring the radon
transport in soil where a house is to be built and then performing subsequent
studies after construction, it may be possible to develop a model for the
effects of various construction methods and housing designs on the transport
processes in the soil around the structure.
However, at this time the state of knowledge is such that all of the
critical variables may not yet have been identified or the measurement
methods used to quantify those variables may not be providing sufficiently
accurate and/or precise data to permit a full understanding of the problem.
It is currently possible to make a series of detailed measurements on a
few "representative" units of the housing stock. At that point it becomes
-------
14
necessary to generalize those measurements to a much larger number of
housing units and thus good physical models are needed to provide a framework
for making such extrapolations. It is likely that a number of such models
will be needed to cover the range of buildings that are present in the
available housing stock.
In developing and testing such mathematical models, the availability
of "research houses" where various parameters can be controlled and measured
will be invaluable to the testing and validation of the mathematical
models. In order to examine the variety of interactions between design,
construction practices, geography, geology, and climate, it will be necessary
to have a large number of such houses available and to examine the radon
levels and infiltration rates over the full range of indoor environmental
conditions. However, the way in which the occupants use the house will
affect its behavior. Thus, the pattern of heating, cooling, window and
door openings may have a profound effect on the radon infiltration and
somehow these kinds of factors must be included in the development of the
physical and mathematical models if good predictive ability is to be
achieved. These models can then be used to develop a cost-effective
mitigation strategy that will reduce the radon exposure with limited
penalties in energy utilization and occupant comfort.
A point was made in the discussion of these indoor radon questions
that radon is not the sole indoor air quality problem and that it is
necessary to consider it in the context of a broader range of pollutants
including smoke, cooking residues, combustion by-products, materials off-
gases, etc. Particularly when considering methods to mitigate against
radon entry or reduce existing radon concentrations, the effects on the
levels of the other airborne contaminants should be considered. However,
the problem is that the only solutions that address both radon and these
other pollutants are increased general ventilation and air cleaning.
Increased ventilation for high radon houses will probably not work. Air
cleaning will perform well for removing cigarette smoke, pollen, combustion
products, etc., as well as radon progeny. However, as has been noted
before, the increased "unattached" fraction that would be obtained from
-------
15
air cleaning may result in the same dose even though the airborne
radioactivity level has been reduced. The report by Jonassen (20) addresses
this question of how much dose reduction is obtained relative to the
reduction in potential alpha energy concentration. The problem of effects
of air cleaning on dose requires further study and deserves inclusion in
future research plans.
Radon Exposure and Health Risk Assessment
The dose to the critical cells in the respiratory tract are strongly
dependent on the deposition pattern of the inhaled radioactive particles,
which in turn depends on the particle size and charge as well as factors
associated with the individual breathing pattern and respiratory tract
morphology. The accurate estimation of the long-term dose to these cells
requires a full understanding of how the environmental factors not only
affect the concentrations of radon progeny in the air, but also the physical
properties (charge, size, rate of growth, rate of neutralization, coagulation
rates, etc.) of the radon decay products. As discussed extensively above,
a number of developments in instrumentation have recently led to new
information on the behavior of the newly formed Po-218 some of which was
reported at the ACS Symposium (21,22). New data are now available on the
deposition of particles in the bronchial region over a wider range of
particle sizes than had been previously available (16). There have also
been continuing modifications to the models for estimating the dose (23,24).
However, considerable uncertainties exist and there are clearly different
points of view on the risk factors for radon decay product health effects
(25,26). Thus, the accurate assessment of the amount and nature of airborne
radioactivity, its deposition in the respiratory tract, the long-term
dose that the radioactivity deposits in the tissue, and the dose-disease
relationships still require considerable additional research and represent
important research goals for the DOE/OHER program.
The development of a comprehensive research program then needs to
support both basic and applied studies that examine the fundamental physics
and chemistry of the decay products, the formation of radiolytic nuclei,
-------
16
the attachment of such nuclei to the preexisting aerosol, and the effects
of the radon decay on the other components of the indoor atmosphere such
as through the production of free radicals that can react with the gaseous
constituents. At this time there are really only a very few measurements
of the properties of the indoor aerosol and it is not yet possible to
generalize these results to a wider range of homes. Additional measurements
are needed to be certain that the full range of possible indoor conditions
have been examined.
The previously discussed results of the work by Schery (18) suggest
there is also a need for further measurements of thoron and its decay
products. Thoron progeny may play a larger role in indoor exposure than
has been considered to date. The other potential problem from thoron is
its interference on measurements of radon and radon decay products. Thus,
there is reason to make additional thoron measurements to determine if
and where thoron problems may exist. Thus, there are a variety of additional
measurements that are needed to better characterize the indoor airborne
radioactivity. In conjunction with the need for more measurements, there
must also be the continued development, testing, and validation of
measurement methods for radon, the decay products, and the full range of
properties of the particles to which the decay products are attached.
DOE has held a leadership role in such method developments and needs to
continue to do so.
Although the total picture is not yet clear, it appears that the
components are now coming together that may make it feasible to start the
development of indoor air chemistry models analogous to those that have
already been developed for photochemical smog in the outdoor air. However,
this effort will require the integration of a number of diverse studies
to determine the critical gaps in our knowledge and identify the critical
path toward obtaining the missing information. The benefits of having an
integrated indoor air chemistry model that would include radon, its decay
products along with the other components of indoor air could be a valuable
tool in unraveling the complex patterns of health effects caused by indoor
air.
-------
17
A clear problem in terms of applying any air chemistry model developed
to the U.S. housing stock is the lack of any model of that housing stock.
It is necessary to have good representation of the gross characteristics
and contents, energy-related or otherwise, of the U.S. housing stock.
Thus, even if we could adequately model the atmosphere inside a home with
a given set of characteristics, we cannot extrapolate that understanding
to the population as a whole because of this lack of housing stock data.
This problem is important to a variety of needs besides assessment of
radon impacts and is badly needed for a variety of uses related to indoor
air quality, energy utilization, and their interactions.
There are also research needs for better data on deposition of fine
particles in the nasal region. There is very little literature on submicron
particle deposition in this area of the respiratory tract. The work on
bronchial deposition should be extended to even smaller particle sizes.
There is a need for a better deposition model to bring the predictions
into line with the new submicron data. With these results it will be
possible to more fully predict the behavior of the ultrafine particle
mode and thus better understand the role of the "unattached" fraction in
dose deposition.
Another problem is the determination of the level of long-term exposure
to radon decay products. There have been suggestions of measuring the Pb-
210 concentrations in the skeleton as such an indicator. There was
substantial criticism of this approach at the workshop on the basis that
only a small fraction of the Pb-210 in the body comes from inhalation of
radon decay products. There will also be considerable intake of this
radionuclide in drinking water and food and thus it becomes very difficult
to determine the effects of airborne radon relative to dietary intakes.
There is also inadvertent ingestion of Pb-210 because of its accumulation
on environmental surfaces resulting in hand contamination. Eating hand-
held food may thus enhance the dietary intake to an unknown extent. There
is also an indication that the detection systems for the Pb-210 may not
be sensitive enough to accurately quantify typical exposure levels and may
-------
18
only be useful for high level exposures comparable to uranium miner levels.
There was a report of using urinary levels of Pb-210 as an exposure
indicator. Such measurements in conjunction with dietary monitoring might
provide a useful measure of past exposure and may deserve further study.
Also, Pb-210 deposited in dusts in poorly ventilated areas of a house may
be useful as a measure of past radon levels and this concept also needs
to be examined further. Although there is a real need for assessing past
exposure and there has been some positive evidence for detecting radon
exposure by Pb-210 in laboratory animals, there were very strong reservations
expressed regarding the use of skeletal Pb-210 levels as an integrating
monitor of past radon exposure.
Assessment of Health Risks by Epidemiological Studies
Current estimates of the lung cancer risks from inhalation of radon
decay products are derived from epidemiological studies of underground
miners, primarily uranium miners, in Colorado, Czechoslovakia, Sweden,
and Canada. Current OHER research is limited to case control studies of
uranium miners in New Mexico where dosimetric, medical, and lifestyle
information are superior to previous miner studies, and a study of female
lung cancer cases in Pennsylvania being initiated in October 1986. It is
anticipated that these studies will provide much more reliable data on
the lung cancer risk from radon exposure than prior studies.
It is important to plan epidemiological studies such that they can
answer the critical questions that are being addressed. It is thus essential
to be able to assess both the dose and the response in a sufficiently
well characterized manner so that the resulting statistical analysis will
be valid. It is not always possible with retrospective studies to be able
to choose a representative population, to accurately assess the exposure
of that population to the agent of interest, or to be able to sufficiently
document the potential confounding factors. Thus, a case-controlled study
of lung cancer in women in Pennsylvania outside of the major urban centers
is being initiated in order to assess the risk from radon in a more typical
population and at radon levels to which the general public may be exposed.
-------
19
This study will also be able to provide information on confounding factors
such as active and passive smoking. The study will measure the radon
levels in houses in which the lung cancer patient lived. Year-long track
etch detectors will be used in each house with a pair of detectors to
determine the radon and thoron concentrations. The population in this
area is quite stable and past experience by the Argonne National Laboratory
on radium dial painters in this region suggests that they will be able to
follow the population back in time long enough to provide useful results.
There are other epidemiological studies on-going in Canada and in
Sweden as well as related studies in the U.S. Other populations that
might be considered for studies are workers and/or patients at spa and
natural hot springs where high radon levels are found. There are significant
numbers of individuals involved at fewer locations to be monitored.
Quantitative Model of Lung Cancer Induction
A major emphasis of OHER radiation research program has been the
investigation of the mechanisms of cancer induction following radiation
exposure. The understanding of these mechanisms are necessary for the
determination of the dose-response relationships over the range of radiation
exposure encountered in the environment and the workplace. The current
radon risk estimates are derived from the high exposure level of miners
and the extrapolation to lower doses. In the absence of a well defined
dose-response relationship, the extrapolation leads to considerable
uncertainty in the resulting risk estimates.
The planned program will involve studies at both the cellular and
subcellular level including searches for cellular markers and oncogenes,
kinetic studies of tissue repair processes that lead to metaplasia and
neoplasia, and other efforts to connect the effects of the passage of the
alpha particle to the physico-chemical changes that ultimately manifest
themselves as cancer. Experimental animal studies are needed, particularly
to help elucidate the relationship between radon and smoking in lung cancer
-------
20
induction. There is some evidence that radon decay products primarily
initiate the carcinogenic process while smoking promotes the cancer growth.
Such a difference in function between these factors may significantly
affect the risk estimates. Additional animal studies can help clarify
this distinction as well as provide additional information on the transport,
deposition and clearance of radioactive particles in the respiratory tract.
Another important aspect of animal studies is that the effects of the
radiation exposure rate as well as the total cumulative dose can be
studied. Rate dependent effects are difficult to identify in epidemiological
studies. There was an extended discussion of the difficulty in finding an
adequate animal model for such studies and in extrapolating the results
of controlled exposure of animals to the uncontrolled exposure of humans
to complex mixtures of environmental toxins. Thus, although models have
their place in elucidating mechanisms, the results of such studies may
not directly result in better estimates of the risk of cancer from radon
exposure.
The development of a detailed model of lung cancer induction caused
by radiation exposure is a very major task, and this goal thus will
potentially represent a very large effort with the required commitment of
substantial resources. There are other agencies such as the National
Cancer Institute who also have a mission in this area of cancer initiation
and it will be important to coordinate OHER-sponsored studies to avoid
duplication of effort being sponsored by the variety of interested agencies.
It may be well for OHER to establish more formal links to other such agencies
interested in this problem to provide the necessary program coordination.
EPA/ORP Radon Action Program
The Environmental Protection Agency has a radon-related program that
involves both research and operational aspects. In addition to conducting
the necessary research studies, a primary goal of the program is the
dissemination of technical information to a variety of interested parties
including state and local agencies dealing with public health and/or
environmental issues, industries such as home building and construction,
-------
21
and individuals so that they can take informed action to accurately measure
the radon levels and take appropriate actions to protect public health.
The EPA Radon Action Plan is conceived as a five year program with
yearly review and updating. The primary goal is to reduce and prevent
exposure to radon decay products in order to reduce the risk of lung
cancer to the general population. The plan is striving to achieve this
goal in the context of a non-regulatory program. There is currently no
statutory authority for radon regulation in homes. The program is thus
aimed at developing information and making that information available in
such a manner as to provide the motivation and technical knowledge to
alleviate the problem. The initial activities involve the implementation
of national and state surveys to obtain data on the patterns of radon
levels in homes and to evaluate methods for the prediction of areas where
housing is at risk of having high radon levels, the development of
standardized sampling and analysis protocols for radon and decay product
concentrations, the development, demonstration, and evaluation of mitigation
methods for reducing the radon levels in existing homes, and the development
of new housing construction practices that would reduce the risk of the radon
problems occurring.
Exposure Assessment
The plans for the EPA efforts in the area of exposure assessment
have been described in detail by Magno and Guimond (27). This report
outlines the objectives and planning considerations for the national
survey of indoor radon concentrations to be conducted in the near-term
future as well as EPA's program of quality assurance including measurement
protocols and the measurement proficiency program. The design of the
national survey is currently being planned and will be subjected to review
before its implementation.
The stated objective of the survey program is to determine the frequency
distribution of radon in houses on a national basis. A major question in
the development of the survey design is that of how large a population of
-------
22
houses will need to be examined in order to provide the necessary
information. In Table 1 reproduced here from Magno and Guimond (27), the
number of units to be sampled is presented as a function of the point in
the distribution of indoor radon concentrations to be determined and the
precision with which that determination is to be made. Thus, to determine
with 5% relative standard error, the point in the distribution where the
radon concentration is only exceeded in 0.1% of the housing stock of the
United States, it will be necessary to survey 800,000 houses. This analysis
assumes a lognormal distribution with a geometric mean of 33 Bq/m^ and a
geometric standard deviation of 2.8 (28). In the discussion of this
question, there was general agreement that there was little justification
for determining the distributional value to better than 20% relative standard
error and that the point in the distribution to be determined should be
in the range from 0.1% to 1%.
There are several large data sets on radon in houses collected by the
Terradex Corporation for the track-etch detectors they have sold to various
agencies and individuals. Similarly, B.L. Cohen (29) in his symposium
paper reported on a large number of samples that he has measured using
charcoal canisters. However, in both these data bases, the samples are
not accumulated on a planned design basis and in many cases, there is
poor or misleading documentation of the location of the detector in terms
of where the house is and where in the house the detector was placed.
Thus, although there are many data points, it is extremely difficult to
extract meaningful information regarding the distribution of radon in the
entire U.S. housing population. Cohen (30) has recently published a more
systematic survey of 453 houses using 1-yr track-etch detectors that
yielded a lognormal distribution with a geometric mean of 38 Bq/nr and a
geometric standard deviation of 2.36, quite similar to the Nero et al.
(29) results. These distributions probably serve as the best starting
points for the development of a larger survey design.
In the survey, there is consideration of stratifying the survey on
the basis of 7 to 8 geographical regions and between single and multi-
family housing with a bias toward oversampling single family houses and
-------
23
Table 1
Number of Sampling Units for Various Population Percentages
Population Percentages (P)
0.1
1
5
Sample
5% RSE
800,000
80,000
15,000
Size (No.
10% RSE
200,000
20,000
4,000
of Units)
20% RSE
50,000
5,000
1,000
P - Percentage of housing units exceeding known radon concentration
RSE - relative standard error
-------
24
undersampling multi-family housing. The sampling would still be fully
random with the potential for additional oversampling in areas where the
geology/radon relationships are being explored.
In the previous discussion, the problem of the lack of a survey of
the U.S. housing stock was raised. In part the national radon survey will
provide some additional information on the houses that are surveyed for
radon. However, in light of the 1990 Census that will be undertaken, it
may be more useful to enlist the cooperation of the Bureau of the Census
to include some critical questions on the buildings in which people live
to provide a more comprehensive data base on the housing stock.
There may be a problem with obtaining cooperation of the individual
dwelling residents and various approaches have been employed to enlist
people in various survey programs. It may be necessary to offer inducements
to people to join the survey. How to approach the individual homeowner
may be critical in obtaining a sufficient level of participation as to
avoid biasing the sample. Mail solicitations have generally been less
effective than direct personal contact. Random digit dialing may be an
intermediate cost approach that will lead to an acceptable participation
level.
New York State has used this approach in its survey of owner-occupied
single family houses and initially had poor participation levels. They have
reported better responses recently, but there is still a question of the
efficacy of telephone solicitation relative to the door-to-door approach.
In any case, it will be important to make a careful review of the survey
as it is taken to insure that it is not being biased through uneven
participation amongst the various strata in the design.
Another consideration in the survey is whether the occupant and/or
owner should necessarily be given the results of the radon measurement.
There may be legal problems in selling a home where the owner knows there
is a radon problem, but does not disclose it to the potential buyer. In
the Argonne study of Pennsylvania, they are allowing the occupant to
-------
25
request that he/she not be informed of the results. Offering this option
may also help to improve the participation level.
While planning is ongoing for the national survey, an immediate
problem is the identification of high risk areas. This problem is being
approached in two ways: 1) by encouraging and providing assistance for
state surveys to identify high risk areas, and 2) through the development
of predictive methodologies to identify such areas. The EPA will assist
the states with radon analyses with charcoal canisters, with assistance in
survey design and sample selection, and with assistance in data analysis,
management, and interpretation. It is hoped that approximately 15 to 20
states per year could be cooperatively surveyed with the goal of identifying
those areas of high radon level potential.
In the longer term, the EPA is trying to develop a land assessment
program that would provide a general idea of the hazard potential throughout
the United States as well as develop both macro- and microscale models to
evaluate regional areas down to individual sites for their potential
radon hazard. In order to accomplish this objective, EPA is working with
the U.S. Geological Survey to develop predictive models relating the
geological setting of a dwelling with its potential for elevated radon
levels. During field studies of mitigation methods as described in the
next section, there will also be opportunities to gather critical data on
soil characteristics and the effects of housing construction practice on
the levels of radon in the resulting structures.
It is of great interest to the home construction industry to develop
an inexpensive and simple method for evaluating a particular parcel of
land for the potential radon in the house to be constructed and to develop
an inexpensive and simple method to treat the soil during the construction
to minimize the radon in the finished home. These are clearly difficult
tasks and there are not obvious approaches to providing these methods.
-------
26
Mitigation Methods
The mitigation demonstration program is being conducted by the EPA
Office of Research and Development (ORD). The program is aimed at
development of cost-effective mitigation methods. The initial work has
focussed on 18 homes in the Reading Prong area of Pennsylvania and initial
results of that work has recently been reported (31). The study is being
expanded to some additional 30 to 40 homes in Pennsylvania and additional
studies are to be conducted in New York and New Jersey on 50 and 30 homes,
respectively. The program is a joint EPA/state venture, and it is hoped
that results of the studies can be extrapolated to problems in other
areas of the country. The initial step for each house is to identify the
radon source, soil or water. If soil is the source, then understanding
of the house dynamics becomes important and collaborative studies are
being conducted with personnel from Lawrence Berkeley Laboratory to make
measurements similar to those they have made elsewhere (5). Various
mitigation methods from crack sealing to air-to-air heat exchangers can
be tested and evaluated for cost and effectiveness in radon control and
related to the building and soil properties. It is hoped that within a
few years there will be several hundred homes that have had mitigation
methods applied to them so that the experience with a variety of methods
in a number of settings will be available.
A house evaluation program is also planned. This program would be to
identify high radon level houses where the homeowner is participant in
the evaluation and mitigation efforts in conjunction with the EPA and
state government. In this way additional houses could be evaluated,
mitigation methods evaluated and the understanding of the control methods
improved.
Another part of the program is the evaluation of design and construction
practices that would provide new housing construction with a greater
level of protection against high indoor radon levels. This program is
being developed jointly with the Department of Housing and Urban Development
and various states. One other aspect of this design development is the
-------
27
desire on the part of builders and others to have a prescription or
alternative prescriptions for the construction in the building code so
that as long as the house conforms to the code, it can be sold regardless
of the actual levels of radon found. Ericson (32) reported a series of
designs for high radon potential areas that appeared to provide sufficient
protection against radon entry. He also suggested that the house be
designed and built with provision for subsequent radon reduction
modifications to be made if needed at relatively low cost. Some doubts
were expressed that there would be sufficient understanding of the radon
entry problem to permit the development of designs that would provide
absolute protection against future radon problems. However, the concept
of incorporating provision for easy and low cost mitigation was strongly
supported.
It was reported that in Sweden, there is a long term plan by the
national government for reducing the general population exposure over the
next century by a factor of 2. They plan to accomplish this reduction by
insuring that as the housing stock is replaced, the newly constructed
houses are radon resistant and easily mitigated. However, achievement of
this goal does require the cooperation of the local building authorities
to insure that construction practices adhere to this goal.
Finally, the EPA is currently confronted with whether they should
encourage or discourage the use in high radon houses of air cleaning
devices (electrostatic precipitators, HEPA filters, etc.) to remove the
decay products relative to structural fixes to reduce the radon levels.
The air cleaning systems will reduce the working level but increase the
unattached fraction. Jonassen (20) has shown that you can reduce the
overall dose by electrostatic filtration based on particular dose models.
Although these devices do provide some limited improvement in the dose,
they are not the best solutions to the indoor radon problem and their use
may discourage the owner from taking the steps necessary to provide more
effective and more permanent solutions to the problem.
-------
29
Other Research
The EPA is also supporting one epidemiological study. The study
focusses on lung cancer in women living in Maine and New Hampshire and
with the University of Maine performing the study. This study is on-going
and it will also be some time before results are available.
Public Information
Another of EPA's strong interests is in providing the public with
clear and understandable information regarding the nature of radon and
its risks. It wants to provide motivation for the public to cooperate in
their own protection by having measurements made and then mitigating the
problem if one is found to exist. The agency has released pamphlets entitled
"A Citizen's Guide to Radon" and "Radon Reduction Methods". One of the
major problems has been to find an effective way to express the comparative
risk from radon in terms that the general public can understand. The
brochure makes comparisons to cigarette smoking and medical x-ray exposures.
It was pointed out that the brochure does not make clear that only the
lung cancer risks are being compared since there are other disease risks
associated with cigarette smoking. As an alternative, it was suggested
that it may be useful to put the rates into perspective relative to causes
of death other than cancer, for example, fatal accidents in the home or
automobile-related fatality risks.
The point was made that the level of risk that the EPA is using is
at the high end of the range of risk estimates and results in a remedial
action guideline level (4 pCi/1) that is too restrictive and that will
overly alarm the public. Levels previously recommended by health physics
organizations such as the NCRP or the ICRP have been substantially higher
(8 to 10 pCi/1). However, the need for public health protection suggests
that a level should be set at a very conservative value that would then
serve as a goal.
-------
30
An alternative approach might be to introduce the ALARA concept with
the associated costs and risks so that the individual who must make
expenditure decisions can see the trade-offs between costs and risk
reductions. However, it is going to be necessary to express those risks
in comprehensible terms. Risks quoted as 1 in 100 or 1 in 1000 are not
easily understood by the general public while $1500 for a ventilation
system is immediately understood.
A strongly supported suggestion for all forthcoming documents is to
o
present the radon concentrations in Bq/mJ rather than starting the general
public on the archaic unit of pCi/1. It would be best to begin the wide-
spread dissemination of information on radon with the proper standard units
rather than having to reeducate them later. Another suggestion for improving
dissemination of radon information particularly in rural areas is to
enlist the aid of the Agricultural Extension Service. The Ag Extension
Advisors in each county could serve as a repository of pamphlets and
efforts could be made to train some of the statewide extension advisors
in radon related matters to provide consulting services to the county
advisors in informing the public of the problem and ways to deal with it.
Conclusions
It was the consensus of the roundtable discussion that the research
programs outlined and discussed did address the critical scientific issues
related to radon and its decay products. During the past decade there has
been a substantial improvement in our understanding of the behaviour and
effects of indoor radon. However, there remain critical uncertainties
that must be resolved with regard to the occurrence and causes of elevated
radon concentrations in houses, the behaviour of the decay products and
their relationship to indoor air quality, the effects of the decay products
on public health, and the most effective methods for mitigating against high
concentrations and human exposure. There appears to be excellent
coordination and cooperation between the Department of Energy and the
Environmental Protection Agency in developing the various aspects of the
research and development programs. The goals and objectives of the plans
-------
31
presented covered the range of problems that our current state of knowledge
reveal and the types of studies proposed represent a scientifically valid
approach to resolving the existing questions.
Recommendations
As a result of these discussions, the following recommendations were made
with regards to the critical areas of research and development related to
indoor radon that need to be addressed in the near term future.
• Studies of the relationships between design and construction of houses
and infiltration rates need to be continued. Such studies should
include both existing housing and research houses.
• Studies are needed to test whether the available data on high potential
radium content or soil permeability based on information such as the
NURE maps or from data from the Soil Conservation Service are sufficient
to make predictions of areas that might be high radon areas.
• From the results of such studies, a land assessment program should
be developed that would provide a general idea of the hazard potential
throughout the United States as well as develop both macro- and
microscale models to evaluate regional areas down to individual
sites for their potential radon hazard.
• A standardized field measurement procedure to assess radon availability
in soil gas is needed. Laboratory studies are needed to determine
the critical variables to measure and how best to control the field
conditions to obtain reproducible measurements.
• Physical models are needed that would be able to predict the radon
levels in a particular structure for a given radon availability and
could evaluate potential construction or control technologies for
reducing the radon entry rates.
• Studies are needed to find an appropriate, long-term, integrating
measurement of the radon decay product concentrations and the size
distributions of the activity so that from the health effects found
in a population, the complete exposure/dose/effects relationships
can be evaluated.
• There is a critical need for measurements of the activity size
distributions for chemical investigations of the indoor atmosphere.
Such measurements may also be important for assessing the dose depending
on which dosimetric model is employed. Improvements in the methods
for determining activity size distributions particularly for particles
with diameters below 10 nm are needed.
-------
32
Basic and applied studies are needed to examine the fundamental
physics and chemistry of the decay products, the formation of radiolytic
nuclei, the attachment of such nuclei to the preexisting aerosol,
and the effects of the radon decay on the other components of the indoor
atmosphere such as through the,production of free radicals that can
react with the gaseous constituents. The results of such studies can
then be incorporated into a comprehensive indoor air quality model.
Measurements of thoron and its decay products are needed to determine
if thoron progeny are more significant in indoor exposure than has been
considered to date.
Further theoretical and experimental studies of particle deposition
at the Reynolds numbers typical of respiratory tract flow are needed.
There are specific needs for better data on deposition of fine particles
in the nasal region.
Experimental animal studies are needed to help elucidate the
relationship between radon and smoking in lung cancer induction,
provide information on the transport, deposition and clearance of
radioactive particles in the respiratory tract, and examine the
effects of the radiation exposure rate as well as the total cumulative
dose.
Epidemiological studies are required that can assess the
exposure/dose/respose in a human population in a sufficiently well
characterized manner so that the resulting statistical analysis will
be valid.
Investigation of the basic biochemical mechanisms of cancer induction
following radiation exposure is needed for the determination of the
dose-response relationships over the range of radiation exposure
encountered in the environment and the workplace.
There is a need for a national survey to determine the frequency
distribution of radon in houses on a national basis with sufficient
precision as to be able to assess the public health threat posed by
radon and the number of houses in the upper portion of the distribution
that require immediate and effective mitigation efforts.
Further studies are needed in the development of cost-effective
mitigation methods for high radon houses.
Additional work is needed to improve design and construction practices
so that new houses would provide a greater level of protection against
high indoor radon levels.
There is a continuing need to provide the public with clear and
understandable information regarding the nature of radon and its
risks as well as the motivation for the public to cooperate in their
own protection by having measurements made and then mitigating the
problems that are found to exist.
-------
33
References
1. Hopke, P.K. , editor, Radon and Its Decay Products: Occurrence.
Properties and Health Effects . Symposium Series, American Chemical
Society, Washington, D.C. (1987).
2. Nazaroff, W.W. and A.V, Nero, editors, Radon and Its Progeny in
Indoor Air. CRC Press, Boca Raton, Florida (1987).
3. Bosnjakovic, B. , P.H. van Dijkum, M.C. O'Riordan, andJ. Sinnaeve,
editors , Exposure to Enhanced Natural Radiation and its Regulatory
Implications. Science of the Total Environment 45: 1-699 (1985).
5. Sextro, R.G. , B.A. Moed, W.W. Nazaroff, K.L. Revzan, and A.V. Nero,
Investigations of Soil as a Source of Indoor Radon, in Reference 1.
6. Cliff, K.D., A.O. Wrixon, B.M.R. Green, and J.C.H. Miles, Radon and
Its Decay-Product Concentrations in UK Dwellings, in Reference 1.
7. Stranden, E., Rn ^n Norwegian Dwellings, in Reference 1.
8. Urban, M. , A. Wicke, and H. Kiefer, Bestimmung der Strahlenbelastung
der BevSlkerung durch Radon und dessen kurzlebige Zerfallsprodukte
in Wohnhausern und im Freien, Report No. KIK 3805, Kernforschungszentrum
Karlsruhe GmbH, Karlsruhe, West Germany, September 1985.
9. Vanmarcke, H. , A. Janssens, F. Raes, A. Poffijn, P. Berkvens, R. Van
Dingenen, On the Behaviour of Radon Daughters in the Domestic
Environment and Its Effect on the Effective Dose Equivalent, in
Reference 1.
10. NEA Experts Report, Dosimetry Aspects of Exposure to Radon and Thoron
Daughter Products, NEA/OECD, OECD, Paris, France (1983).
11. NCRP, Exposures from the Uranium Series with Emphasis on Radon and
its Daughters, Report No. 77, National Council on Radiation Protection
and Measurements, Bethesda, MD (1984).
12. McLaughlin, J.P., Population Doses from Radon Decay Products in
Ireland, in Reference 1.
13. Holub, R.F. and E.O. Knutson, Measurement of ^18Po Diffusion Coefficient
Spectra Using Multiple Wire Screens, in Reference 1.
14. Kulju, L.M. , K.D. Chu, and P.K. Hopke, The Development of a Mobility
Analyzer for Studying the Neutralization and Particle Producing
Phenomena Related to Radon Progeny, in Reference 1.
15. Raes, F. , A. Jannsens, and H. Vanmarcke, Modeling Size Distributions
of Radon Decay Products in Realistic Environments, in Reference 1.
-------
34
16. Cohen, B.S., Deposition of Ultrafine Particles in the Human Tracheo-
bronchial Tree: A Determinant of the Dose from Radon Daughters, in
Reference 1.
17. NCRP, Evaluation of Occupational and Environmental Exposures to Radon
and Radon Daughters in the United States, Report No. 78, National
Council on Radiation Protection and Measurements, Bethesda, MD (1984).
18. Schery, S.D., Measurements of Airborne 212Pb and 220Rn at Varied
Indoor Locations with the United States, Health Phys. 49: 1061-1067
(1985).
19. Lowder, W.M., Future Directions in U.S. Radon Research, presented to
the 191st National Meeting of the American Chemical Society, New York,
April 1986.
20. Jonassen, N., The Effect of Filtration and Exposure to Electric
Fields on Airborne Radon Progeny, in Reference 1.
21. Chu, K.D., P.K. Hopke, E.O. Knutson, K.W. Tu, and R.F. Holub, The
Induction of an Ultrafine Aerosol by Radon Radiolysis, in Reference
1.
22. Porstendorfer, J., A. Reineking and K.H. Becker, Free Fractions,
Attachment Rates and Plateout Rates of Radon Daughters in Houses, in
Reference 1.
23. James, A.C., A Reconsideration of Cells at Risk and Other Key Factors
in Radon Daughter Dosimetry, in Reference 1.
24. Harley, N.H. and B.S. Cohen, Updating Radon Daughter Bronchial
Dosimetry, in Reference 1.
25. Steinhausler, F., On the Validity of Risk Assessments for Radon
Daughters Induced Lung Cancer, in Reference 1.
26. Kartell, E.A., Critique of Current Lung Dosimetry Models for Radon
Progeny Exposure, in Reference 1.
27. Magno, P.J. and R.J. Guimond, Assessing Exposure to Radon in the
United States: An EPA Perspective, in Reference 1.
28. Nero, A.V., M.B. Schwehr, W.W. Nazaroff, and K.L. Revzan, Distribution
of Airborne 222Radon Concentrations in U.S. Homes, LBL 18274, Lawrence
Berkeley Laboratory, Berkeley, CA (1984).
29. Cohen, B.L., Surveys of Radon Levels in U.S. Homes as a Test of the
Linear-No Threshold Dose-Response Relationship for Radiation
Carcinogenesis, in Reference 1.
30. Cohen, B.L., A National Survey of 222Rn in U.S. Homes and Correlating
Factors, Health Phys. 51:175-183 (1986).
-------
35
31. Henschel, B. and A.G. Scott, The EPA Program to Demonstrate Mitigation
Measures for Indoor Radon: Initial Results, in Proceedings of the APCA
International Specialty Conference on Indoor Radon, SP-54, Air Pollution
Control Association, Pittsburgh, PA (1986).
32. Ericson, S.-O. and H. Schmied, Modified Design in New Construction
Prevents Infiltration of Soil Gas Carrying Radon, in Reference 1.
------- |