5
Radiation Dose Assessment

INFORMATION ON THE PATHWAYS through which radionuclides are transported to the public from a contaminated site provides the starting point for calculating the dose to individuals and for estimating health risk. This chapter assumes that ambient exposures from a overhead plume or from radionuclides deposited on the ground have been determined and that the concentrations of radionuclides in air and water in the vicinity of the exposed population have been estimated. It is further assumed that both have been described in quantitative terms suitable to the needs of dose estimation. This chapter describes the process of dose estimation, the different types of dose assessment that can be undertaken, and the uncertainties involved.

SOURCES OF EXPOSURE

Three sources of radiation exposure that must be taken into account in estimating the dose either to representative or to specific persons are: (a) external exposure due to submersion in contaminated air or due to radiation from an overhead plume or from radionuclides deposited on the ground, (b) the inhalation of radionuclide-contaminated air, or (c) the ingestion of radionuclides in water and foodstuffs.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 40
5 Radiation Dose Assessment INFORMATION ON THE PATHWAYS through which radionuclides are transported to the public from a contaminated site provides the starting point for calculating the dose to individuals and for estimating health risk. This chapter assumes that ambient exposures from a overhead plume or from radionuclides deposited on the ground have been determined and that the concentrations of radionuclides in air and water in the vicinity of the exposed population have been estimated. It is further assumed that both have been described in quantitative terms suitable to the needs of dose estimation. This chapter describes the process of dose estimation, the different types of dose assessment that can be undertaken, and the uncertainties involved. SOURCES OF EXPOSURE Three sources of radiation exposure that must be taken into account in estimating the dose either to representative or to specific persons are: (a) external exposure due to submersion in contaminated air or due to radiation from an overhead plume or from radionuclides deposited on the ground, (b) the inhalation of radionuclide-contaminated air, or (c) the ingestion of radionuclides in water and foodstuffs.

OCR for page 40
Ambient Exposure External exposure to ambient radiation may be a small contributor to the total dose of a member of the public in many dose reconstructions. However, external exposures can be significant for noble gases such as 41Ar or 133Xe or long-lived radionuclides that are deposited on the ground. This exposure may be the easiest to estimate, albeit with some uncertainty, if suitable environmental measurements were made in the open and in typical shelters. Absorbed doses from external sources have been calculated for an extensive list of radionuclides and tabulated in Federal Guidance Report 12 (Eckerman and Ryman 1993). These coefficients are based on detailed scattering calculations from the distributed source to the organs and tissues in a mathematical phantom. Since the coefficients are specific for each radionuclide, it is essential to consider the ingrowth of decay products. For example, an assessment of the external exposure from 137Cs must consider the ingrowth of 137mBa. Inhalation Exposure Contamination of the air with radionuclides can be described in several ways, including the concentration of radionuclides in the inhaled air, the size of the inhaled particles, and the solubility of the particles that contain the nuclides. Calculation of the dose to the whole body or to a specific organ requires further information: the estimated duration of daily exposure (in hours), the breathing characteristics of that individual, a model of the respiratory tract that allows calculation of the amount of airborne particles deposited in the airways, the physical and metabolic characteristics of the deposited radionuclide, and the size of the individual. Aerosol particles are usually assumed to be lognormally distributed in terms of their aerodynamic diameter and hence can be characterized by their geometric mean and standard deviation. The models currently used for determining the deposition of airborne contaminants in the respiratory tract are being revised (ICRP 1995). The new versions will be more detailed and will define the fate of a deposited radionuclide through modeling the clearance characteristics of the particular inhaled compound. If a particle is insoluble in body fluids, it is assumed to clear slowly from the location in the respiratory tract (primarily the lung) where it is deposited. If the particle is soluble, it will be cleared from the lung more quickly. Once cleared to the blood or the gastrointestinal tract, the radionuclides can translocate to other organs before finally being excreted from the body. This process can take several months or longer. The chemistry of the compound containing the radionuclide strongly influences the biodistribution of the radionuclide and

OCR for page 40
the transfer rates. Alkaline earth elements such as strontium or radium that are similar in ionic radius and in chemical characteristics to calcium will deposit in the skeleton. Some of the actinide elements (those elements with an atomic number equal to or exceeding actinium's 89) also will deposit in the skeleton, primarily through their interaction with organic molecules before they are mineralized in bone. An example of this is plutonium. The actinides also tend to accumulate in the liver. As a result of this accumulation, measurements of radionuclides in tissue samples may provide a method of dose assessment when long-lived radionuclides are involved. The absorbed dose due to inhaled radionuclides can be calculated by means of an appropriate model that takes into account lung deposition, transpiration, organ distributions, and clearance rates (Loevinger and others 1991). The absorbed dose to an organ is defined as the radiation energy absorbed per unit mass of that organ; it is determined from the fraction of energy absorbed by the organ from radiation emitted by radionuclides (alpha particle, beta particle, gamma or x-ray, etc.) in particular organs or in adjacent organs. The deposition of 100 ergs/gram is one rad; 100 rad is a gray (Gy). The unit of dose equivalent is the rem, which is the absorbed dose in rad multiplied by a modifying factor, or quality factor, to adjust for the biological effectiveness of a particular type of radiation. The SI unit of radiation dose equivalent, the sievert (Sv) equals 100 rem. Many body organs in addition to the lung may be exposed to the radiation emitted by the inhaled radionuclide. Both doses to specific organs and equivalent doses to the body can be estimated. In the decades before 1970, models of the relationships between physiologic processes and dose were simplistic. Nonetheless, they led to radiation protection strategies that were sufficient to limit the potential health hazard to exposed persons. This simpler perspective has evolved more recently into approaches (e.g., ICRP 1989, 1991, NCRP 1993) that translate organ doses into an equivalent dose to the whole body (an effective dose) by weighting the contributions of all organ doses. Committed doses, i.e., the total dose delivered over an extended period of time, typically 50 years following intake, are used for this purpose. However, for the purpose of epidemiologic study, annual organ doses for low-LET and high-LET radiation are preferred to effective doses. For exposures to long-lived radionuclides, the calculation must be a time-dependent one. Ingestion Exposure The procedure for calculating radiation dose from ingested radionuclides is similar to that for calculating the dose from inhalation. Ingestion is defined as the swallowing of radionuclides in water (or other liquids),

OCR for page 40
and in food. Absorption of radionuclides from the gastrointestinal tract to blood is affected by such factors as chemical form and interactions with other food in the gastrointestinal tract. Given such variable absorption, the most practical means of estimating the dose is to use a model with average absorption kinetics and to calculate the dose to organs based on normal physiological processes. Inhaled particles brought out of the lung by ciliary action are also ingested. From that point, the radiation dose is determined as if the radionuclide were ingested, not inhaled. The radiation dose to the gastrointestinal tract is largely due to the radionuclide activity within the contents of the bowel, the critical organ for some major fission products, for example, 103Ru and 106Ru. Within the gastrointestinal tract, the colon will usually receive the largest doses because the residence times are greatest there. Ingesting of insoluble radionuclides that are alpha-emitters (such as plutonium) can result in low radiation doses since the alpha dose to tissues in the wall of the bowel is only a very small fraction of the dose to its contents. The largest dose will be to organs that accumulate and retain the radionuclide. However, the variability in absorption of the ingested radionuclide in the gastrointestinal tract is responsible for the greatest uncertainty in the potential dose. Because radiation guidelines are usually conservative, it is likely that the commonly used absorption factors over-estimate the amount of the radionuclide that is absorbed and hence the organ dose. POTENTIAL CONSEQUENCES OF RADIATION EXPOSURE The biologic effects of ionizing radiation are better understood than are those of exposure to any other potentially harmful element or compound. In large doses, ionizing radiation clearly causes cancer in humans, but fortunately radiation is a relatively weak carcinogen. Earlier in this report it is stated that no measurable increase in the risk of cancer has been observed in the Japanese population exposed to doses below 0.2 Gy (20 rad). Traditionally, radiation protection guidelines are predicated on a linear dose response, which assumes that the harmful effects of radiation are linearly related to the dose and that there is no threshold dose. Most experts believe this assumption is conservative; that is, it overestimates the effects of ionizing radiation at low doses because it ignores the potentially beneficial effects of the body's repair mechanisms. In the past, this probable overestimation of the risk was regarded as a good thing consistent with the still widespread philosophy that it is better to be safe than sorry. This philosophy holds true only when unlimited resources are available to protect the public health and the environment. Once resources are acknowledged to be limited, overestimates of a particular

OCR for page 40
risk are ultimately harmful to the public health because funds are diverted from larger risks to protect society from smaller risks. This diversion of funds ultimately will result in greater mortality than would have occurred if resources were spent in proportion to the amount of health benefit that would be achieved. To illustrate the magnitude of the cost of a dose reconstruction project, the U.S. Defense Nuclear Agency in the Nuclear Test Personnel Review (NTPR) over 10 years ago undertook the task of developing radiation dose estimates for about 203,000 military personnel and civilian employees of the U.S. Department of Defense who participated in the atmospheric nuclear weapons tests between 1945 and 1962. Approximately $100 million was required to estimate the doses received by the service personnel. The National Research Council was asked to review the scientific aspects of the NTPR project and in its report suggested that it is far more important to be able to state with a high degree of confidence that the dose is below some selected value than to estimate what the dose is when it is far lower (NRC 1985). In other words, it might be reasonable to establish an effective dose commitment over so many years below which it makes little sense to waste huge sums of money for further quantification. To estimate health risks reliably, the most accurate estimate of dose to members of the public should be obtained. Estimates of the health risk of radiation exposure must be provided to the public in the most understandable terms possible, using a sound methodology and proper quality assurance. This is not simple because risk can be expressed in many ways. It can be stated in relative terms, such as the ratio of the risk in one population (or exposure group) to another or as the excess relative risk (the difference between the observed relative risk and 1, where 1 is the value expected in the absence of an effect). Risk also can be couched in absolute terms as the excess number of occurrences of an effect (for example, cancer deaths or cases) above the number "normally expected"—in the absence of exposure to other unnatural causes, with age and sex distribution and the length of observation taken into account—in the population of interest. "Normally expected" in this context would include exposure to ionizing radiation emanating from the Earth's crust, originating in outer space, or incurred medically. Risk also can be expressed as years of life lost, or as attributable risk, or as the proportion of cases that would not occur if the exposure had not happened. It is not likely that any single measure of risk captures all of the information a member of the public might desire, especially when risk can vary with factors such as age, gender, and time after exposure. Because it is proportional to the spontaneous occurrence of the effect of interest (cancer, for instance), relative risk has limited utility in the ab-

OCR for page 40
sence of knowledge about the spontaneous or baseline rate; with relative risk alone the number of events that might occur cannot be predicted. Risk expressed as excess occurrences of a health effect (deaths or cases) has an immediacy that is readily grasped, but it depends on the spontaneous rate of such events and on the accuracy and completeness of their recognition. The rate of spontaneous cancers and the accuracy of diagnosis differ within a country from region to region, even from city to city, and neither can be estimated with assurance in small populations. These differences in the expression of risk become important in the present context, however, only when one seeks to extrapolate from the available partial-lifetime data to the full lifetime of a population such as that in the vicinity of the Hanford Nuclear Site, or when one seeks to extrapolate risks derived from one population to another one that has very different baseline rates for particular cancers. In the first instance, for example, the choice between the multiplicative or the additive risk projection model can lead to projections that may differ severalfold. In the second instance, the use of the Japanese A-bomb risk estimates can lead to substantial differences in the projection of site-specific risks where the baseline values are very different such as for cancers of the stomach, colon, liver and breast. The following sections describe in general terms the approaches that can be used to assess potential risk that attends a given exposure. After that is a section on the types of uncertainty that can be encountered for such a procedure. The dose assessments have been divided into categories: preliminary, comprehensive, and individual. The preliminary approach is essentially a scoping process, the results of which can indicate the need for a more comprehensive dose assessment. If a dose assessment for a specific individual is desired, the process requires much more detailed information about that individual, including weight, height, lifestyle, and the like. PRELIMINARY DOSE ASSESSMENT The purpose of a preliminary dose assessment is to determine the need for a study of the health effects resulting from the exposure of a population to radiation. A preliminary assessment will normally precede a comprehensive health effects study, but when there is evidence of epidemiologic effects or widespread public concern, the health study can be initiated in lieu of, or in parallel with, the preliminary dose assessment. The features of a preliminary dose assessment can be subdivided into those related to input, method, and output. An accurate estimate of the source term, with time dependence where necessary, is required. Spatial and temporal environmental data are

OCR for page 40
needed, both from modeling studies and from measurement programs (modern and historic). Lifestyle and dietary data must be assumed, either from generic sources or from more local sources if available; individual data are not usually necessary at this stage. Depending on the pathways of importance, additional data can be required, to include food distribution and production data, population movements, and countermeasure data in the case of a specific accidental releases. Existing data on intake and external dose pathways will normally be used. Depending on the nature of the assessment, data on doses from other sources (natural, medical) and local background disease incidence might be needed. This early assessment can help to focus further studies. For example, it might be sufficient to consider only individuals of a specific age, or only a limited number of organs, or a specific set of radionuclides. The principal intention of a preliminary assessment is to estimate the doses, and hence the risk of health effects, to an exposed population. For prolonged exposures (over several years), a useful approach can be to estimate the doses received by representative (but hypothetical) individuals. Their estimates can be extended to more extreme groups in the population by considering the sensitivity of the results to the assumption of unusual habits that can give rise either to greater exposure from standard pathways or to exposure from unusual pathways. The sensitivity of the results to variations in all of the parameters thought to be uncertain should be considered. A sensitivity analysis will generally be sufficient for the preliminary assessment, and a full uncertainty analysis should not be required (see Glossary for the distinction between sensitivity and uncertainty analyses). The results to be obtained from a preliminary dose assessment will depend largely on the nature of the specific assessment and on the questions the study is designed to answer. In general, it is more useful to calculate annual dose rates and risks to reference individuals than to calculate risks based on committed effective doses. In addition, the over all dose to and risk of each cohort can be calculated by combining the representative results with the size and age structure of each cohort. With total cohort risk, the estimated total number of health effects can be obtained. The preliminary dose assessment should include a review of any bias that is likely to damage the successful outcome of a more comprehensive study. A related output will be an estimate of the statistical power of the comprehensive study. COMPREHENSIVE DOSE ASSESSMENT In a preliminary assessment, some releases and pathways will be found to be potentially more important than others. A more comprehen-

OCR for page 40
sive study of the releases and pathways will be needed if an epidemiologic study is warranted. A comprehensive dose assessment also will be needed to provide individual specific doses that will reduce uncertainty in the dose estimate from the preliminary assessment. The preferred dose estimate from a comprehensive assessment is the annual absorbed dose to the target organs of the body. For radiation exposures from external sources in the environment, the absorbed dose to body organs increases with decreasing body size; this effect is most pronounced at low photon energies and for absorbed dose to organs located near the middle of the body that are shielded by overlying tissue. Petoussi and others (1991) indicate that, at photon energies greater than 100 keV, absorbed doses in an infant can be about 40% higher than those in an adult male for exposures both from a contaminated ground surface and from submersion in a cloud source. Below 100 keV, the difference could approach a factor of 3 for deeper organs such as the ovaries and colon (Eckerman and Ryman 1993). If the radiation exposure is from inhalation or ingestion, then physiologically based pharmacokinetic (PBPK) models will need to be developed to supplement the ICRP Publication 30 models (ICRP 1979) developed for a ''reference" man for general applicability to radionuclides of various resident times in different organs of the body. Hence, PBPK models, which can be combined with the age-specific mathematical phantoms, can be used to make realistic calculations of absorbed dose as a function of age at the time of exposure and at a specific time after exposure. For radionuclides in bone, the absorbed dose can be delivered for many years after the initial exposure. Thus, intake could be considered daily or weekly and the absorbed dose calculated annually over the lifetime of the individual. As with the preliminary dose assessment, the absorbed doses due to low-LET radiations (x- or gamma rays, beta particles) should not be added to absorbed doses from high-LET radiations (usually alpha particles). The doses from the radiations with different LET values should be listed separately before computing "equivalent doses." The organ dose is especially important when developing doses to compare to site-specific health effects. However, there are circumstances when it is desirable to compare irradiations with different distributions among body tissues or to combine doses from non-uniform irradiation of various organs and tissues. Therefore, it may be necessary to calculate effective doses in addition to organ doses. The applicability of absorbed doses to a particular situation is often influenced by exposure conditions that differ from the assumptions made in a model and by other factors that might alter these conditions at the location of the exposed individual. Any such assumptions must be justified and validated. For example, the radiation exposures inside a resi-

OCR for page 40
dence could be substantially different from those outside because of shielding, in the case of a contaminated ground plane, or because of filtration, in the case of inhalation of contaminated air. Environmental factors also can influence the time-integrated activity that characterizes exposures, and an individual's lifestyle can influence the extent of contact with radionuclides in the environment. For example, the time-integrated activity of a radionuclide in an urban environment can be substantially different from that in a rural area. INDIVIDUAL DOSE ASSESSMENT An individual dose assessment can be used in epidemiologic studies, and it will be needed for persons in the potentially exposed population who are interested in their own exposures. The preferred dose estimate from an individual risk assessment is the annual absorbed organ dose. As with the comprehensive dose assessment, the radiation dose from internal sources as a result of either inhalation or ingestion can use PBPK models. The absorbed doses also must be calculated in terms of the LET of the various particles emitted (the high-LET alpha particles and the low-LET beta particles and photons). The absorbed dose to an exposed individual is influenced by the exposure conditions for that person. As with the comprehensive assessment, environmental factors will influence the time-integrated activity that characterizes exposures, and an individual's lifestyle can influence the extent of contact with radionuclides in the environment. The dependence of the absorbed dose on other factors such as gender, lifestyle, and diet must be considered, and the details applied will be commensurate with the level of detail in the source term and the pathway analysis (Napier 1992). For example, the amount of time spent outdoors could affect the degree of exposure through inhalation or contact with radioactive materials deposited on soil or vegetation; the amount of milk consumed could affect the intake of radioactive iodines. It is unlikely that efforts to assess doses individually will be rewarding unless fairly detailed information is available on those factors that influence the individual dose. UNCERTAINTY Dose reconstruction involves the use of measurements and calculation procedures and scientific judgments made on the basis of available data. Each should be characterized by statements of uncertainty, providing measures of precision and accuracy. Statements of uncertainty must be made in defining the problem and presenting the results of investiga-

OCR for page 40
tions. One must state the degree of certainty with which one needs to know the answer, the sensitivity of the methods employed, and how the results of the studies are expressed. Statements of uncertainty must be expressed in probabilistic terms. A quantitative estimate of dose uncertainty is important in determining the needed sample size, the achievable precision and statistical power, and the number of required measurements. The feasibility and utility of a study would depend on these considerations. In assessing exposure and absorbed dose, uncertainty should be expressed for physical, biologic, and computational methods. The calculations of uncertainty should be propagated throughout all calculations. Errors associated with physical measurements are likely to be smaller than are errors associated with biologic measurements, because the major contributor to the latter is interpersonal variability. In obtaining measures of propagated errors, procedures for incorporating methods of assessment of uncertainty for physical and biologic results are required. It is helpful to separate uncertainty of knowledge of the state of nature from the act of making decisions. It is often necessary to make decisions even in the face of the uncertainty. This problem is solved by using decision thresholds above which action will be taken. That is, before deciding to carry out a study, decision-makers need first to determine whether the uncertainty that will attend it is acceptable. The dependence of the effects of radiation on dose rate and latent period makes it necessary to take into account the time dependence of doses delivered to the whole body or to specific organs. For instance, previous annual doses are used to assess the risk to individuals in the future. Accumulation of these doses over time will depend on factors such as residence history and yearly dose rates. Direct measurement of radionuclide content in the body and in environmental samples, and individual dose measurements (physical or biodosimetric) should be used for validation of model predictions wherever available. External doses for different groups within a population can be evaluated from dose rate measurements in open areas or from radioactive contamination as determined by sample counting. Internal dose from inhalation and ingestion can be evaluated from radionuclide concentration in air and in food products. If the result of a direct measurement is different from a model prediction, preference must be given to the measured result. SUMMARY AND RECOMMENDATIONS Three levels of dose assessment can be envisaged: a preliminary assessment, a comprehensive assessment, and an individual assessment.

OCR for page 40
The purpose of a preliminary dose assessment is to determine the need for a full-fledged study of the health effects that results from the exposure of a population to ionizing radiation. The more comprehensive dose assessment will be needed if an epidemiologic study is undertaken. Among the aims of a comprehensive assessment are to provide the information necessary to compute individual specific doses and to reduce the uncertainty in the dose estimate arising from a preliminary assessment. Finally, individual dose assessments define doses to particular individuals in the potentially exposed population who may be interested in an evaluation of their potential health risk. The most useful dose quantity in a preliminary dose assessment is the effective dose; in a comprehensive dose assessment or in the computation of an individual dose, the preferred dose estimate will be the annual absorbed organ doses from low-LET and high-LET radiations. However, it should be noted that published intake dose conversion factors are for committed doses. Hence, except for the equilibrium situation with short effective half-life radionuclides, it is more difficult to calculate annual doses for various years following intake. The committee makes several recommendations as follows: All exposures from external sources, inhaled radionuclides, and ingested radionuclides should be considered; when certain pathways or other factors suggest that a particular source term or radionuclide will not contribute substantial dose, reports should explain why these sources or specific radionuclides were not considered in the final estimations. Dose assessment should proceed at three levels: preliminary, comprehensive, and individual dose assessment. Acceptable levels of uncertainty should be defined before a decision is made to carry out a detailed study. A readily available set of intake-to-annual dose conversion factors for long-lived radionuclides should be established. Doses should be expressed as effective doses in a preliminary dose assessment and objective criteria should be used to decide whether it is warranted to embark on a full-fledged study. Dose estimates in a comprehensive dose assessment should be expressed as the annual organ absorbed doses from low-LET and high-LET radiations. Estimation of the effective dose may also be helpful. In an individual dose assessment, the doses should be described separately as the annual organ-absorbed doses from low-LET and high-LET radiations.