Click for next page ( 76


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 75
4 Role of Exposure and Dose or Risk Assessments in Developing Radiation Standards This chapter discusses the use of exposure and dose or risk assessments in providing a technical basis for standards for radionuclides in the environment. It does not discuss in detail the current approaches to exposure and dose or risk assessments of radionuclides, such as their use in demonstrating compliance with standards. The standards for radionuclides in the environment discussed in this report are directed only at protection of humans. According to available data and modeling, radiation standards that would provide acceptable protection of humans would ensure that other species are not put at unreasonable risk. although individual members of a species occasionally might be harmed (IDEA 1992; ICRP 1991~. Risks to biota other than humans and the issue of whether separate standards are needed for their protection are not considered in this study. The chapter begins with a general discussion of the risk-assessment process for carcinogens and the application of the process to radionuclides. That is followed by a discussion of the calculational elements of dose or risk assessments for radionuclides in the environment and the use of such assessments in developing standards. The chapter ends with the committee's views on suitable approaches to exposure and dose or risk assessments for purposes of developing standards for radionuclides in the environment. RISK ASSESSMENT OF CARCINOGENS As described, for example, by the National Research Council (1983) and the Office of Science and Technology Policy (OSTP 1985) the process of 75

OCR for page 75
76 ROLE OF ASSESSMENTS IN DEVELOPING STANDARDS risk assessment of carcinogens, including radionuclides, generally consists of the following four components: Hazard identification. Estimation of dose-response relationship. . Exposure assessment. Risk characterization. Brief descriptions of these components and the usual approaches to addressing them for radionuclides are presented in the following sections. Hazard Identification Hazard identification for carcinogens entails a qualitative evaluation of the data bearing on an agent's ability to produce carcinogenic effects and the relevance of this information for humans. This component of the risk- assessment process is particularly important for chemical carcinogens because data on possible carcinogenic effects on humans often are lacking. However, hazard identification is not an important concern for radionuclides, including naturally occulting radionuclides found in TENORM, because widespread epidemiologic data have shown conclusively that ionizing radiation can cause cancer in humans (for example, National Research Council 1990; 1988~. Noncarcinogenic health effects also are potentially important for some radionuclides that can be found in TENORM. Particularly for uranium, chemical toxicity in the kidney after exposures above some threshold is clearly demonstrated by a large amount of animal and human data (for example, Leggett 1989~. For uranium, an important issue is whether current radiation standards for the public would prevent chemical toxicity in the kidney. Resolution of that issue depends on a number of factors including: the assumed threshold for chemical toxicity in the kidney; selection of an appropriate safety factor below the assumed threshold for protection of public health, including protection of unusually sensitive populations (such as infants and children); and the relationship between kidney burden and radiation dose from uranium. On the basis of consideration of those factors and current data and models, Kocher (1989) concluded that chemical toxicity generally should be considered in developing health-protection standards for the public with respect to ingestion and inhalation of soluble or insoluble uranium. For example, the Environmental Protection Agency has based a proposed standard for uranium in drinking water on prevention of chemical toxicity in the kidney, as well as limitation of radiation dose (see chapter 7~. However, the chemical toxicity of natural

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM 77 uranium appears to be potentially important only if the limit on radiation dose from exposure to uranium is greater than about 0.25 mSv (25 mrem) per year and if the dose from uranium results primarily from ingestion and inhalation rather than external exposure (Kocher 1989~. The chemical toxicity of naturally occurring radionuclides found in TENORM and its effect on setting radiation standards is not considered further in this report. Estimation of Dose-Response Relationship Estimation of the dose-response relationship for carcinogens is the process of estimating the magnitude, or an upper bound on the magnitude, of the carcinogenic effect of any given dose. For radiation exposure, best estimates of the carcinogenic effect of a given dose normally are emphasized. Estimation of the carcinogenic response to exposure to particular radionuclides is greatly facilitated by the generally held view (which is based on observation and modeling) that the absorbed dose in tissue is the fundamental physical quantity that determines the response to any exposure. The absorbed energy in tissue depends only on the radiation type and its energy (NCRP 1993a; ICRP 1991), not on the source of the radiation. Therefore, in contrast with the current situation for chemical carcinogens, animal or human studies to estimate the dose-response relationship for exposure to every radionuclide of potential concern are not required. Rather, the dose-response relationships developed for different types of radiation (such as photons, electrons, and alpha particles) can be applied to exposures to any radionuclide once the energies and intensities of all the kinds of radiation emitted in the decay of the radionuclide are known. Current estimates of the dose-response relationships for radiation exposure obtained, for example, by the National Research Council's Committee on the Biological Effects of Ionizing Radiations (National Research Council 1990; 1988) are based on studies of human populations that received radiation doses considerably above environmental levels. The most important study groups include the Japanese atomic-bomb survivors, who received high doses from photon exposures of all organs and tissues of the body, and various groups of miners, who received high doses to the lung from exposure to alpha particles emitted by the short-lived decay products of radon. Applying the dose-response relationships for radiation exposure estimated by such groups as the BEIR committee to the considerably lower doses of concern in controlling routine exposures of the public requires assumptions about extrapolation to doses and dose rates beyond the range of direct observation (ICRP 1991~. For purposes of risk management, the carcinogenic response to environmental radiation generally is assumed to be a

OCR for page 75
78 ROLE OF ASSESSMENTS IN DEVELOPING STANDARDS linear function of dose without a threshold. Although the assumption has often been challenged, the consensus among regulatory authorities and expert organizations is that there is insufficient evidence to support a departure from the linear, no-threshold dose-response relationship for regulatory purposes. This issue has not been investigated in the present study of standards for TENORM. Exposure Assessment Exposure assessment is the process of identifying a group or groups of individuals who might be at risk and describing for the~on the basis of assumptions, observations, or modeling- the various routes (pathways) of exposure and their magnitude, frequency, and duration. Exposure assessment is essentially the same for radionuclides and hazardous chemicals except that external (direct) exposure to penetrating radiation (such as photons) is an important pathway for radionuclides but not for hazardous chemicals. The process of exposure assessment for radionuclides in the environment is considered in more detail later in this chapter. Risk Characterization Risk characterization is the process of combining the information obtained from hazard identification, estimation of the dose-response relationship, and exposure assessment to describe the carcinogenic risk associated with expected or assumed human exposures to the agent of interest. Risk characterization for radiation exposure is relatively straightforward and is almost always expressed in terms of the numerical probability of cancer induction in an individual or the number of cancers in a population over a defined time period. Some approaches to risk characterization for radiation exposure focus on fatal cancer as the end point of concern, whereas others focus on cancer incidence; the difference between the two is not important for most organs and tissues at risk (ICRP 1991~. Other issues in risk characterization include the risks that might be experienced by particularly susceptible populations and alternative approaches to expressing risk. Those issues have not been considered in this study. ELEMENTS OF RADIATION RISK ASSESSMENT This section discusses the calculational elements that normally make up an assessment (estimation) of risk posed by exposure to radionuclides in the environment. It does not apply to risk assessments for exposure to radon and its short-lived decay products in air. As noted in chapter 1, risk assessments of

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM 79 radon usually are based directly on epidemiologic data without the need to estimate dose from a given exposure and risk per unit dose. There is some overlap between this discussion and the general discussion of risk assessment for carcinogens above. However, we are discussing radiation risk assessment separately mainly because the approaches to estimating radiation risks were developed long before the general risk- assessment process for carcinogens was formally laid out. Radiation risk assessment usually includes estimates of dose from a given exposure, which are generally not part of risk assessments of chemical carcinogens. An exposure and dose or risk assessment of radionuclides in the environment addresses the relationship between the concentrations of radionuclides in various media (air, water, soil, or other materials) at particular locations and the resulting radiation doses or risks to exposed individuals or populations. Knowledge of this relationship can be used either to derive estimates of dose or risk corresponding to given concentrations of radionuclides in the environment or to derive estimates of concentrations of radionuclides in the environment corresponding to a given dose or risk. As shown in figure 4.1, an assessment of risk corresponding to given concentrations of radionuclides in the environment, or vice versa, consists of three elements: An assessment of internal and external exposures, which provides estimates of intakes of radionuclides by ingestion and inhalation, and estimates of exposures to external radiation per unit concentration of particular radionuclides in environmental media. An estimate of radiation dose, given the estimates of radionuclide intakes and external exposures. . An estimate of risk, given the estimate of dose from internal and external exposures. As noted previously, the measure of risk posed by radiation exposure normally is assumed to be cancer mortality, although cancer incidence (morbidity) is calculated in some cases. In many environmental-radiation assessments, the desired end point for the calculations is dose, rather than cancer risk, especially when an assessment is used to develop a radiation standard expressed in terms of dose or to demonstrate compliance with a dose standard. When dose is the desired end point, only the first two elements listed above are included in the assessment. Furthermore, as noted previously, dose often is calculated as an intermediate step even when risk is the desired end point for the assessment.

OCR for page 75
80 e - e "I-~ DIAL ~ ~ : IllllllI1 I F ELIDE T-IO r e : I ~OSU~ 1 -I "OL-1C+ LOGIC DOSE DOS BLOC 1 EuNG . . . __ ~ ~I~l!~1 I. 1 PC . =e CODE J Sag SS~T uNG CONCEN ON ~ ~n ~ Sal F T o o 1 ~ 1 ~ = o = = : = ~ : 8 o = : ~ 1 ~ File 4.1. Ill s ~ risk, ~se, Boat ~ come-on of =~o~cli~s ~ eat Ida. Mica example ~ ~ Cow.

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM 81 The approach to estimating dose and risk for given internal and external exposures (the second and third elements of a risk assessment listed above) often is straightforward. Estimates of committed doses from internal exposure per unit activity intakes of radionuclides by ingestion and inhalation can be obtained, for example, from federal guidance (Eckerman and others 1988~. Similarly, for radionuclides dispersed in air, water, or soil, estimates of external dose rates per unit activity concentration are given in federal guidance (Eckerman and Ryman 1993~. Estimates of dose from internal and external exposure then can be converted to estimates of risk by using the nominal fatal- cancer risk per unit dose discussed in chapter 7. These simplified approaches to estimating dose and risk generally are adequate for purposes of controlling radiation exposures (that is, in developing standards and in demonstrating compliance with standards), but they might not be adequate for estimating doses and risks for real exposure situations (see chapter 11~. The approach to estimating internal and external exposures per unit concentration of radionuclides in various environmental media (the first element of a risk assessment) generally is more complicated than the approach to dose and risk estimation described above, especially for internal exposure. Estimates of ingestion intakes of radionuclides can require estimates of transfers of radionuclides through various terrestrial or aquatic food-chain pathways to humans, including transfers into food crops, meat, milk, and fish; and estimates of inhalation intakes can require estimates of transfers of radionuclides from various environmental media, such as soil, into the air. Exposure assessment also requires estimates of so-called usage factors, including intakes of contaminated water, air, soil, foodstuffs, or other materials by humans or livestock and residence times and shielding factors for external exposure. The general approach to exposure assessment for radionuclides in the environment is depicted in figure 4.2. The source compartments can include water, air, soil, or other materials (such as contaminated buildings or equipment), and the exposure compartments can include the source compartments plus various foodstuffs or other materials into which radioactivity is transferred. The particular exposure pathways that should be considered in any assessment depend on such factors as Me characteristics of the site, the source compartments of concern, the physical and chemical forms of radionuclides, and the assumptions about living habits of exposed individuals or populations. Examples of important exposure pathways for different source compartments are as follows. For contaminated soil, important exposure pathways generally include: Consumption of vegetables, fruits, and grains grown in the contaminated soil.

OCR for page 75
82 ROLE OF ASSESSMENTS IN DEVELOPING STANDARDS CONCENTRATIONS | _ IN SOURCE COMPARTMENTS r TRANSPORT AND INTERACTION PROCE SSES CONCENTRATIONS EXTERNAL ~IN EXPOSURE EXPOSURE COMPARTMENTS 1 1 l DOSE TO l I HUMANS | _ INTERNAL EXPOSURE Figure 4.2. Relationships between concentrations of radionuclides in source compartment and resulting doses to humans from internal and external exposure.

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM Consumption of milk and meat obtained from livestock that eat pasture grass grown in the contaminated soil. . Direct consumption of the contaminated soil. Inhalation of radionuclides from contaminated surface soil suspended in the air. . External exposure to the contaminated soil. 83 For contaminated groundwater, additional important pathways of internal exposure generally include: . Direct ingestion of water from the contaminated source. Use of the contaminated source as a water supply for agricultural purposes, such as watering of livestock; irrigation of vegetables, fruits, and grains, and irrigation of pasture grass consumed by livestock. For contaminated surface water, additional important exposure pathways generally include: . The ingestion pathways listed above for contaminated groundwater. . Consumption of aquatic foodstuffs (such as fish) obtained from the contaminated source. . External exposure to the contaminated source during such activities as swimming, boating, and residence along contaminated shorelines. For contaminated air, important exposure pathways generally include: Inhalation and external exposure to airborne radionuclides in the source compartment. Exposure to radionuclides deposited on vegetation and the ground surface, including external exposure and ingestion of radionuclides incorporated into foodstuffs (vegetables, fruits, grains, meat, and milk). . air. Inhalation of deposited radionuclides that are resuspended in the Finally, if the source compartment consists of contaminated structures or equipment, credible scenarios for ingestion, inhalation, and external exposure

OCR for page 75
84 ROLE OF ASSESSMENTS IN DEVELOPING STANDA=S would need to be developed and evaluated on the basis of the characteristics of the particular exposure situation of concern. Most models for terrestrial and aquatic food-chain pathways assume that the concentration of a radionuclide in the exposure comparDnent of interest (such as milk) is a constant multiple of the concentration in the source compartment (such as soil). Thus, food-chain pathway models generally use radionuclide-specif~c transfer factors (also called concentration factors or concentration ratios in some cases). For pathways involving exposure to contaminated soil, the long-term retention of radionuclides in surface soil, which depends on the solid-solution distribution coefficient (Kd) for the radionuclides, also is important. The models and parameter values for the various internal and external exposure pathways of concern in exposure assessments for radionuclides in the environment are discussed in a number of references (for example, IAEA 1996b, 1994; NCRP 1984d; Till and Meyer 1983~. Additional issues are particularly important with respect to TENORM. Soluble radionuclides are more available for biologic uptake than those sorbed on soils or sediments, so the partitioning of radionuclides among these phases- between groundwater and aquifer materials; between river water and suspended or bottom sediments-is important in pathway modeling. A laboratory-derived distribution coefficient, K<3, is typically used, where K = concentration of radionucTide sorbed on sediment concentration of radionucTide remaining in solution . For a given element, the coefficient can be expected to vary with the chemical speciation of the element, the solute chemistry of the water, the mineralogy and surface area of the solids, redox conditions, and pH. For example, the presence of competing ions in solution can decrease the sorption of a radionuclide. That effect was seen in the sorption of 226Ra from a sodium chloride oil-production brine by soils and marsh sediments where the percentage of radium sorbed increased with brine dilution (Landa and Reid 1983~. In groundwater, uranium partitioning to the solid phase can be expected to increase with more-reducing conditions along a Towpath. Laboratory-derived K<3 values should attempt to simulate field conditions at the TENORM site, and generic, literature-derived Kin values should be used with caution. Much of the nutrients locked up in unweathered rock fragments might not be available for plant uptake; the chemical forms of radionuclides in TENORM might greatly influence their environmental mobilities and biologic availabilities. Total radionuclide concentrations in TENORM might not be

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM 85 reflective of the biologically-labile pool, and some subsegment of the total could be of greater value in assessing biologic uptake and resulting dose. For example, in assessing the hazard due to inhalation of radon decay products, radon that does not escape the radium-bearing mineral matrix is not a concern. The radon emanation coefficient is the fraction of the radon formed in a radium bearing solid that escapes to the atmosphere. It can vary widely in natural and industrial materials; for example, although 226Ra coprecipitated with barium sulfate generally emanates less than 1% of the 222Rn generated, that coprecipitated with or sorbed on iron hydrous oxide can emanate nearly 100% (Hahn 1936~. For TENORM with high emanation coefficients, radon release and radon decay- product inhalation might be the pathway of concern. For TENORM with low emanation coefficients, whole-body gamma exposure might yield the limiting dose. For the inhalation or ingestion pathway, the solubility of the TENORM particles in bodily fluids (such as lung serum for inhaled particles, or stomach acid) and in the soil and aquatic environments where TENORM resides can exert a major influence on the direct and indirect uptake of radionuclides by humans. In viva solubility can be affected by several factors such as particle size, position of nuclides in less-accessible particle interiors versus those on particle surfaces (for example, heat-volatilized 2~0Pb/2'0Po condensed onto cooling particle surfaces in flue dust of thermal-process phosphate plants and in coal-combustion fly ash), and kinetics. In soil and aquatic environments, radionuclide solubility can vary with pH, presence of complexing or precipitating ions in contact solutions, and redox conditions. Plant uptake of radionuclides differs between different members of a decay series and between species for a given nuclide; the biologic uptake of radium by terrestrial plants has been reviewed by Simon and Ibrahim (1990~. If biologic-uptake data are not available, leaching tests with dilute acids to assess readily dissolvable fractions, or salt solutions to assess ion-exchangeable fractions can be used to provide an index of biologic availability. USE OF DOSE OR RISK ASSESSMENT IN DEVELOPING RADIATION STANDARDS In principle, radiation standards expressed in terms of dose or risk can be developed without the need to consider exposure-pathway and dose or risk assessments for the exposure situations of concern. If a standard is expressed directly in terms of risk, all that is required in developing the standard is a judgment about a limit on acceptable risk. Similarly, if a standard is based only on a judgment about acceptable risk but is expressed in terms of dose as a

OCR for page 75
86 ROLE OF ASSESSMENTS IN DEVELOPING STANDA=S surrogate for risk, the only additional requirement is an assumption about the risk per unit dose. For purposes of developing standards expressed in terms of dose, the risk per unit dose normally is assumed to be independent of the particular radionuclides and exposure pathways of concern. In practice, however, some type of exposure-pathway and dose or risk assessment normally is used in developing radiation standards, even when the standards are expressed only in terms of dose or risk. For example, such assessments are required by the National Environmental Policy Act whenever imposition of a standard would have substantial economic or environmental impacts. More generally, dose or risk assessments are important in developing radiation standards because risks posed by environmental exposures are not directly observable and dose often cannot be measured, especially the dose from internal exposure. Dose or risk assessments are important in developing radiation standards in the following ways. First, if a standard is expressed in terms of dose or risk and is based on a judgment about acceptable risk, as described above and discussed further in chapter 5, a dose or risk assessment for the exposure situations of concern can be used to demonstrate whether compliance with the standard is reasonably achievable. Such demonstrations help in gaining acceptance of the standard by the public and affected parties. Second, if a standard is expressed in terms of dose or risk but is based on a judgment about the achievability of doses or risks for the exposure situations of concern, as described above and discussed further in chapter 5, a dose or risk assessment clearly is required to support the conclusion that the standard is reasonably achievable. In making such a judgment, some type of cost-benefit analysis often is performed in which the costs of achieving various doses or risks are compared with the corresponding benefits in health risks averted in exposed populations (for example, Wolbarst and others 1996; ICRP 1 989b). Third, radiation standards might be expressed in terms of directly measurable quantities, such as concentrations of radionuclides in environmental media or external exposure rates, rather than in terms of dose or risk which cannot be measured. Standards expressed in terms of measurable quantities often are developed to facilitate demonstrations of compliance, especially in the case of standards for naturally occurring radionuclides (see chapter 7~. If such a standard is based on an assumed limit on dose or risk, exposure and dose or risk assessments clearly are required in deriving the limits on measurable quantities. However, an exposure and dose or risk assessment is not necessarily required even in developing environmental standards expressed in terms of directly measurable quantities. Consider, for example, current standards for alpha-emitting radionuclides in drinking water, radioactivity in liquid discharges from particular mines or mills, control and cleanup of residual radioactive

OCR for page 75
GUIDELINES FOR EXPOSURE TO TENORM 87 materials at uranium and thorium mill-tailings sites, and indoor radon, all of which are concerned with exposures to naturally occurring radionuclides (see chapter 7~. In each standard, the quantitative criteria are expressed in terms of measurable quantities, and the criteria were developed on the basis of considerations of levels of radioactivity that are reasonably achievable for the exposure situations of concern, given existing background levels and the ability of current technologies to reduce them. However, the cost-benefit analysis used in developing the criteria in each case was performed with respect to levels of radioactivity, rather than with respect to the corresponding doses or risks to individuals or populations. Thus, the values of measurable quantities that were judged reasonably achievable were derived essentially without concern for the resulting doses or risks. Those examples illustrate that an exposure-pathway and dose or risk assessment is required in developing standards expressed in terms of measurable quantities only when the standards are based on a judgment about acceptable dose or risk. Additional discussions on the various ways that standards for TENORM might be expressed and the implications of the different forms of standards for exposure-pathway and dose or risk assessment are presented in chapter 1 1. SUITABLE APPROACHES TO RISK ASSESSMENT IN DEVELOPING STANDARDS As indicated earlier, it is not the purpose of this chapter to discuss in detail the kinds of exposure and dose or risk assessments that might be used in developing a technical basis for radiation standards, especially standards for TENORM. However, the committee offers the following general observations on suitable approaches to risk assessment in developing radiation standards. First, in developing radiation standards, it is appropriate to use stylized methods of exposure and dose or risk assessment for assumed reference conditions, provided that the assumed conditions are reasonably representative of the exposure situations of concern and that the regulations permit the use of alternative and more realistic approaches for specific exposure situations, especially in demonstrating compliance with the standards. The use of stylized methods of exposure, dose, and risk assessment in developing environmental standards for radionuclides is consistent with the approach used in developing secondary limits (limits on intakes of radionuclides or concentrations of radionuclides in air) for control of radiation exposures in the workplace (for example, Eckerman and others 1988~. Second, exposure and dose or risk assessments used in developing standards should be reasonably realistic, particularly the assumptions about

OCR for page 75
88 ROLE OF ASSESSMENTS IN DEVELOPING STANDARDS pathways for internal and external exposure, including transfers of radionuclides among various environmental compartments, transfers through terrestrial and aquatic food chains, and the various usage factors for internal and external exposure pathways. That is, the assumptions should not be intended to greatly overestimate or underestimate actual outcomes for the exposure situations of concern. The need for reasonably realistic assessments is particularly important if standards are based on cost-benef~t analyses with respect to dose or risk. A dose or risk assessment that is unreasonably conservative or nonconservative could lead to standards that either are not reasonably achievable or are not adequately protective of public health.