Diseases, Populations, and Other Issues of Public Concern
In oral and written testimony presented to the committee at or in response to its information-gathering meetings, members of the public voiced their concerns and raised questions about diseases or conditions for which individuals in one or more of the populations designated in the Radiation Exposure Compensation Act (RECA) of 1990, and as later amended, are not entitled to compensation. These included some cancers and benign neoplasms, and several diseases with known or suspect autoimmune basis, and we have reviewed the literature on those conditions. The public also drew the committee’s attention to groups of people that may have been occupationally or environmentally at risk for radiation exposure from nuclear-weapons development programs but who are ineligible for RECA compensation because they do not otherwise meet certain eligibility criteria. Such groups include
US civilians residing or working in areas overseas that are not designated by RECA but that may have been contaminated by fallout from US atmospheric nuclear tests conducted in the region.
People living in the vicinity of mill tailings or in dwellings built with mill tailing or other mill or mine residues.
People who were at risk for radiation exposure fell outside RECA-designated periods or who were ineligible based on their failure to meet the relevant residence requirement in the defined time interval between exposure and disease.
The committee reviewed epidemiologic and other pertinent scientific and clinical literature to identify new information about the specific diseases of inter-
est, including their characteristics in the general or “nonexposed” population and their known or suspected causes. We looked particularly for new information about relationships between exposure to ionizing radiation and other potential hazards during activities similar to those experienced by the designated RECA population and the individual diseases and conditions identified by the public. We reviewed new information about the effects of in utero exposure to radiation and the psychologic consequences of exposure or suspected exposure to radiation, and reviewed information related to the potential for exposure of other populations not eligible for RECA compensation but with circumstances and potential for exposure that may have been similar to those of RECA-eligible populations, such as those who were in utero and at risk during the nuclear-testing period. We also address other issues regarding eligibility for RECA compensation which the public brought to the committee’s attention. These include concerns in two areas about claimants failed attempts to use affidavits in establishing proof of eligibility.
Our findings are reported here for diseases not currently compensable under RECA for some RECA-eligible populations and for specific populations that are not listed among those eligible for compensation. The diseases are categorized as malignant (or cancerous) and nonmalignant (or noncancerous).
Members of the public drew the committee’s attention to an apparent inconsistency in compensation under RECA: all types of leukemia except chronic lymphocytic leukemia (CLL) are compensable as radiogenic diseases for downwinders and onsite participants but not for uranium miners, millers, or ore transporters. Non CLL leukemia is compensable in a number of other exposed populations.
Leukemia in Uranium Miners
As reported in Chapter 4, a strong association between radiation and all types of leukemia other than CLL has been identified in several populations exposed to external penetrating radiation (gamma and x rays) at doses generally greater than 200 mSv and at high dose rates. However, as noted in Chapter 4, uranium miners are exposed primarily to alpha-particle radiation from inhaled radon. No exposure pathway has been established whereby immature blood cells are exposed to radiation from radon daughters suspended in air. However, a pathway has been proposed on the basis of the transfer of radon gas from the pulmonary region of the lung into the blood and from the blood to fat cells distributed in the bone marrow (Eatough and Henshaw, 1993). Allen and
colleagues (1995) explored that proposed pathway using marrow fat content in a sample of 20 human ribs. They concluded that the bone marrow fat fraction is the important variable related to the alpha-particle radiation dose from radon in fat. Beta particles from lead-214 (214Pb) and bismuth-214 (214Bi) have sufficient range to deliver a dose of radiation to neighboring bone marrow cells, and this is the proposed pathway of the induction of leukemia in underground uranium miners.
According to a recent report by the National Research Council Committee on Biological Effects of Ionizing Radiation, BEIR VI (NRC, 1999), no significant excess of leukemia cases has been found among mining cohorts. As discussed in Chapter 5, the most definitive study of cancers other than the lung was the pooled study of 11 mining cohorts reported by Darby and colleagues (Darby et al., 1995). They found that the relative risk (RR) of all lung cancer deaths combined (n = 1,179) was close to expected (RR = 1.01, 95% CI = 0.95-1.07). For leukemia, there was a statistically significant increase in the RR for miners employed in the mines for less than 10 years (21 deaths; RR = 1.93, 95% CI = 1.19-2.95), but there was no increase in the RR for workers employed more than 10 years (48 deaths; RR = 0.99, 95% CI = 0.73-1.31). The overall RR was not significantly increased (69 deaths; RR = 1.16, 95% CI = 0.90-1.47). Darby et al. (1995) thus concluded that protection standards for radon exposure should continue to focus on lung cancer alone. Mortality from leukemia in radon-exposed mining populations has also been summarized (UNSCEAR, 2000). The trend in mortality from leukemia with dose (mean Working Level Month [WLM] = 155) was not statistically significant. Roscoe (1997) reported a nonsignificant increase in the RR of leukemia mortality in uranium miners (13 deaths; standardized mortality ratio [SMR] = 1.6, 95% CI = 0.8-2.7).
In addition to the studies of uranium miners, several ecologic studies have been done of the relationship between leukemia and indoor radon. Their results have been equivocal; some indicate a significant association, and others do not show a relationship. A review of 19 ecologic studies, six miner studies, and eight case-control studies (Laurier et al., 2001) concluded that the overall epidemiologic results do not provide evidence of an association between indoor radon exposure and leukemia.
Leukemia in Millers and Ore Transporters
Uranium millers and ore transporters were at risk for exposure primarily to soluble uranium dusts of low radioactivity, most of which is rapidly cleared from the body. Thus, the probability of radiation-induced leukemia was low in those populations. Pinkerton et al. (2004) show fewer than the expected number of deaths among the cohort of Colorado Plateau uranium mill workers from all types of leukemia, including CLL, compared with the US general population (5 deaths, SMR = 0.66, 95% CI = 0.21-1.53). The committee is unaware of any studies of
leukemia in ore transporters and assumes that their risks would be similar to or less than those of uranium millers.
Conclusion On the basis of the epidemiologic evidence, the committee concludes that the uranium miner and miller populations now identified by RECA are not at significantly greater risk of dying from leukemia than males of similar age in the US general population. The leukemia risk to ore transporters is assumed to be similar to or less than the risk to millers. Thus, there is no epidemiologic basis for designating leukemia as a RECA-compensable disease for those populations.
The terms myelodysplasia and myelodysplastic syndromes (MDS) identify an imprecisely defined group of clonal hematologic disorders characterized by abnormal appearance of the bone marrow hematopoietic progenitor cells, ineffective hematopoiesis, peripheral blood cytopenias, and frequent evolution into the complete clinical picture of acute myeloid leukemia (AML). The disorders, which include aplastic anemia, may occur de novo without apparent cause at any age but mostly at the age of 60-80 years. Approximately 7,000-12,000 new cases of MDS are diagnosed each year in the United States. Secondary forms may occur at any age after the use of intensive chemotherapy with alkylating agents or the combination of chemotherapy and radiation therapy. The frequent progression of both de novo and secondary or therapy-induced MDS to clinically apparent AML suggests that the syndrome represents various steps in the continuous process of malignant transformation to leukemia. The natural progression of MDS syndrome is such that when the number of typical immature blast cells increases above a defined point in the peripheral blood, the diagnosis is changed from MDS to AML.
Leukemia studies in the Japanese atomic-bomb survivors have identified cases of MDS, but results of autopsy-based case studies could not be converted into a dose-response projection (Finch, 2004).
Conclusion The committee concludes that radiation exposures of downwinders and onsite participants are likely to be well below the doses at which MDS has been noted in prior studies and that MDS itself should not be included as an additional compensable disease under RECA. If leukemia is diagnosed, it may be compensable on those grounds based on PC/AS criteria adopted by RECA.
Multiple myeloma is compensated under RECA except in uranium miners, uranium millers, and ore transporters, and the committee was asked to reevaluate
the justification for this exclusion. Multiple myeloma is a malignant disease of the blood characterized by an abnormality in plasma cells and possibly in the bone marrow stroma. An increased incidence of and mortality from multiple myeloma have been reported in the Japanese atomic-bomb survivors (Preston et al., 1994; Pierce et al., 1996). A similar conclusion was reached by the National Research Council BEIR V committee (NRC, 1990).
Nuclear-industry workers exposed to low LET radiations at lower doses and lower dose rates than the atomic-bomb survivors have not shown statistically significant increases although some of the findings have been of borderline statistical significance (Gilbert et al., 1979; Gilbert et al., 1993).
Estimated risks of multiple myeloma among persons exposed to alpha-particle radiation have generally been negative. The most recent update of the Colorado Plateau uranium miners study (Roscoe, 1997) showed a nonsignificant increase in RR of death from multiple myeloma (6 deaths; SMR = 1.8, 95% CI = 0.6-3.8), with no exposure-response trend. Similarly, the pooled study of 11 miner cohorts (Darby et al., 1995) found a nonsignificant RR of death from multiple myeloma (26 deaths; SMR = 1.30, 95% CI = 0.85-1.90). The most recent study of uranium millers (Pinkerton et al., 2004) found the mortality from multiple myeloma to be about equal to that expected in the US population (3 deaths; SMR = 1.02, 95% CI = 0.21-3.00).
Conclusion Multiple myeloma is classified as having very low or no susceptibility to radiation induction. Incidence is related to low LET dose in some studies, and it is compensated in downwinders and onsite participants. However, there is no convincing epidemiologic evidence to warrant inclusion of miners, millers, or ore transporters for RECA compensation from multiple myeloma.
Under RECA, uranium millers and ore transporters are compensated for renal cancer, but miners, downwinder and onsite participants are not (see Table 2.1).
Renal cancer accounts for about 3% of all cancer deaths annually in males in the US general population. It is about 3 times more frequent in men than in women and generally occurs in people in their 60s and 70s. Of all renal cancers in adults, 85-95% are renal cell carcinomas, which develop from the epithelial cells lining the kidney tubules. A strong association has been observed between cigarette-, pipe-, and cigar-smoking and death from renal cancer (Bennington, 1973).
Although the atomic-bomb survivors and some populations exposed medically to penetrating low LET radiation at high doses and high dose rates have been found to be at increased risk for renal cancer (NRC, 1990), susceptibility to induction of renal cancer by such radiation is considered to be low (see Chapter 4, Table 4.1 in Mettler and Upton, 1995).
Renal Cancer in Miners
The most recent update of the Colorado Plateau uranium miners study (Roscoe, 1997) found a deficit of deaths due to renal cancer (SMR = 0.4, 95% CI = 0.05-1.4). In their collaborative analysis of mortality in 11 cohorts of underground miners, Darby et al. (1995) also found no statistically significant deficit of deaths due to renal cancer (44 deaths, 9 deaths; SMR = 0.91, 95% CI = 0.66-1.22).
Studies of mortality among several cohorts of uranium processors in the US nuclear industry whose occupational risk of exposure to uranium was primarily to insoluble uranium compounds with low to high enrichment (< 5% to > 90%) also failed to find any significantly increased risk of renal cancer (Polednak and Frome, 1981; Checkoway et al., 1988; Loomis et al., 1996; Ritz et al., 1999a and 1999b). None of those negative studies provided dose-response estimates for radiation-induced renal cancer.
Conclusion The committee concludes that there is no epidemiologic evidence that indicates that uranium miners are at greater risk for death from renal cancer than men of similar ages in the United States. Therefore, there is no epidemiologic justification for adding renal cancer to the list of cancers for which uranium miners are compensated under RECA.
Renal Cancer Among Downwinders and Onsite Participants
In addition to the risk of exposure to 131I, downwinders and onsite participants also were at risk for external exposure to low LET radiation from fallout, albeit at lower levels. A study of the incidence of solid cancers between 1958 and 1987 among 79,972 people in the Life-Span Study (LSS) cohort of atomic-bomb survivors found no significant radiation effect of cancer of the kidney (Thompson et al., 1994). On the basis of 73 incident renal cancer cases (0.8% of all the solid cancer cases ascertained in this study), radiation exposure was estimated to be associated with an excess relative risk (ERR) of 0.71/Sv (95% CI = 0.21-2.25), and an excess attributable risk (EAR) of 0.29/104 person-years Sv (95% CI = 0.50-0.79), with an AR of 15.2% (95% CI = 2.6-41.3). For renal cancer, similar ERRs were obtained by using either attained age or age at exposure in the model.
In a more recent study of mortality from solid cancers between 1950 and 1997 among 86,572 members of the LSS cohort of atomic-bomb survivors, 60% of whom had radiation doses of at least 5 mSv (0.5 rem), Preston et al. (2003) did not specifically report risk estimates for renal cancer mortality, but in an example they gave the estimated risk of renal cancer at the age of 70 years in a woman who was 30 years old at exposure as an ERR of 0.97/ Sv (90% CI = <3-40), and an EAR of 0.14 (90% CI = < −0.1-0.4), with an AR of 14% (90% CI = <3-42).
Probably more pertinent to evaluating the risk of renal cancer among the downwinders and onsite participants in the absence of site-specific cancer studies of these populations are the studies, primarily of mortality, among several cohorts of military personnel who participated onsite in nuclear-weapons tests conducted by the United States and the UK at test sites in the United States and Pacific islands and in New Zealand, respectively (Darby et al., 1993a, and 1993 b; Pearce, 1996; Pearce et al., 1997, Watanabe, 1995a, and 1995b; Johnson et al., 1996; IOM, 1999). In none of those studies was mortality due to renal cancer statistically significantly greater than that in the comparison group.
Conclusion On the basis of current epidemiologic evidence, the downwinders and civilian onsite participants covered by RECA are probably not at increased risk for death from renal cancer.
None of the RECA populations identified by RECA is eligible for compensation for any type of skin cancer. Nonetheless, the committee was asked to reevaluate the justification for this exclusion. Two histologic types of skin cancer are described here: nonmelanoma skin cancer (NMSC) and melanoma.
NMSC is the commonest type of cancer found among Caucasians worldwide, among whom the incidence is increasing because the risk increases with age. NMSC now accounts for an estimated 1 million new cases of skin cancer each year in the United States. The fatality rate is extremely low, and typically neither incidence nor mortality data are reported by cancer registries. NMSC is somewhat more common in men than in women. It comprises two main pathologic conditions: basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), which occur in a ratio of about 4:1. Exposed areas of the face and head are the common sites of both types, but BCC occurs on the trunk and legs more frequently than SCC. Several environmental and host risk factors have been described for both types; the major ones are exposure to ultraviolet (UV) radiation from the sun (which is nonionizing), genetic predisposition, and lack of much skin pigmentation (Schottenfeld and Winawer, 1996).
NMSC is considered to be moderately susceptible to induction by ionizing radiation at high doses and high dose rates, and it was the first type of cancer to be causally associated with exposure to ionizing radiation. Relationships between NMSC and exposure to ionizing radiations have been extensively reviewed by the International Commission on Radiological Protection (ICRP, 1991) and the United Scientific Committee on the Effects of Atomic Radiation (UNSCEAR, 2000). The latent period varies widely and averages just over 20 years. Epidemiologic studies have suggested a synergistic interaction between UV and ionizing radiation that is more than additive (Shore, 1990), but this was not confirmed in a more recent study of skin cancer risk among atomic-bomb survivors (Ron et al., 1998b).
Melanoma or malignant melanoma is a relatively rare type of cancer—about 55,000 new cases (4% of cancers) in US males and females are predicted in 2004 (Jemal et al., 2004)—but its incidence is increasing more rapidly than that of any other kind of cancer around the world, particularly among people of Caucasian origin. It has a much higher fatality rate than NMSC. Epidemiologic studies indicate that exposure to sunlight (UV radiation) is the major causative factor (Schottenfeld and Winawer, 1996).
Melanoma is generally regarded as nonradiogenic with respect to ionizing radiation; this conclusion is based on limited information (UNSCEAR, 2000). However, followup by Reynolds and Austin (1985) of an earlier finding (Austin et al., 1981) of a 4-fold excess incidence of melanoma among employees at the Lawrence Livermore National Laboratory (LLNL), Livermore, California, found statistically significant increases in the incidence of melanoma among males (21 cases observed, 6.46 cases expected) and females (7 cases observed/1.35 expected). No other increase was found in incidence of groups of cancers recognized as being susceptible to induction by ionizing radiation. A more recent followup study (Austin and Reynolds, 1997) compared 31 melanoma cases diagnosed in LLNL employees between 1969 and 1980 with 110 individually matched controls with respect to established or suspected several risk factors, including occupational exposure history. Several occupational risk factors, including working around radiation sources (odds ratio [OR] = 3.7), were strongly associated with the cases. Multivariate analyses of the risk associations suggested that these factors did not act as confounders to the radiation/melanoma relationship, that is, they were independent risk factors. After adjustment for constitutional and occupational risk factors of interest, the OR for the association with radiation status remained increased (OR = 2.3, 95% CI = 1.0-7.6. However, the finding of increased melanoma risk at LLNL remains equivocal since none of the several epidemiologic studies of workers at other US federal nuclear plants or national laboratory facilities has found increases in the risk of melanoma. A nonsignificant increase in melanoma found among US radiologic technologists employed for > 5 years before 1950 (Freedman et al., 2003) suggests a need for further studies to evaluate the role, if any, of ionizing radiation in its induction.
Skin Cancer in Uranium Miners, Millers, and Ore Transporters
Tomasek et al. (1993) and Denman et al. (2003) suggested that ambient conditions in underground mines might cause cancers at sites other than the lung. Specifically, they suggested that radon in ambient air might be responsible for excess risks of skin cancer and leukemia. The committee has reviewed the pertinent literature to assess the scientific strength of that suggestion.
The risk of NMSC in radon-rich atmospheres is not associated with inhalation or other intake of radioactivity (see Appendix B) but is related to the plate
out of radon daughters on the skin. The dose to the germinal layer of cells in the skin depends on the concentration of radioactivity on the skin surface and the ability of alpha particles from polonium-218 (218Po) and polonium-214 (214Po) to penetrate down to the location of sensitive cells. Previous studies of the process have used a nominal value of 70 µm as the depth of the basal cell layer at the base of the epidermis. Recent measurements have indicated large variation in the depth of those cells, and this could lead to higher radiation doses to the basal cell layer than were previously expected (Eatough, 1997).
The incidence of skin cancer was reported (Sevcova et al., 1978) to be higher in Czech uranium miners than in nonminers in the Czech general population (observed = 28.6 per 10,000 workers, [95% CI = 14.4-51.2], expected = 6.3 per 10,000). Most of the excess was found in miners employed underground for more than 10 years. In all cases, the tumors were surgically removed, and there was no sign of recurrence. In a followup of the same cohort, Sevcova and colleagues (1978) again found a significantly higher incidence of basal cell carcinoma, with an attributable annual risk of 1 per 10,000 workers per year per 1 Sv. Not unexpectedly, because of its low fatality rate, no statistically significant increase in the RR of death from skin cancer was found in either the Czech study (Sevcova et al., 1978) or in the pooled study of 11 uranium-miner cohorts (Darby et al., 1995). Similarly, the most recent update of the Colorado Plateau uranium-miner mortality study (Roscoe, 1997) found a nonsignificant excess of skin cancer deaths among miners compared with the US population (2 deaths, SMR = 2.5, 95% CI = 0.3-9.2). The pooled study by Darby and colleagues (Darby et al., 1995) found a deficit of malignant-melanoma deaths (18 deaths; SMR = 0.92, 95% CI = 0.54-1.45). The committee found no reported risk estimates for skin cancer among uranium millers or ore transporters.
The lifetime risk of death from skin cancer after whole-body exposure to ionizing radiation is 2 × 10−4/Sv (ICRP, 1991, p 139). That translates to a risk of 1 × 10−7/WLM. The lifetime risk of lung cancer from inhalation of radon daughters is 2 × 10−4/WLM (ICRP, 1991, p 139). Thus, the risk of death from skin cancer is about 1/2,000 the risk of lung cancer for the same exposure to radon daughters (see Appendix B).
Conclusion The committee found no convincing epidemiologic or dosimetric evidence that uranium miners are at increased risk for death from basal-cell or other type of NMSC. However, the one study that examined the incidence of basal-cell cancer in miners (Sevcova et al., 1978) found an excess incidence among those more heavily exposed to radon decay products. Basal cell cancer is very rarely fatal, is the most common of all cancers, is caused primarily by exposure to sunlight, and causes no change in quality of life; the committee therefore found insufficient evidence to add basal-cell carcinoma to the list of RECA-covered.
Skin Cancer Among Downwinders and Onsite Participants
A study of cancer incidence among the extended LSS cohort of atomic-bomb survivors (n = 79,972) found an increased risk of NMSC in the high-dose group (over 1 Sv [100 rem], mean = 2.22 Sv) compared with the referent (less than 0.01 Sv) group and low dose group (0.01-0.99 Sv, mean = 0.18 Sv); in this study, the ERR for NMSC at 1 Sv was estimated to be 1.0 (95% CI = 0.41-1.89) (Thompson et al., 1994). The more recent study by Ron et al. (1998b) of skin tumor risk among the same cohort of atomic-bomb survivors found a strong nonlinear dose-response relationship for basal-cell carcinoma (80 cases ascertained) (ERR at 1 Sv = 1.9, 90% CI = 0.83-3.3) but not for squamous-cell carcinoma (69 cases) or melanoma (10 cases). As in earlier studies, age at exposure was found to be predictive of skin cancer development; the risk was highest among those exposed when young.
A recent study of NMSC among US radiologic technologists (Yoshinaga et al., 2005) found small increases in risk for basal cell carcinoma but not squamous cell carcinoma among technologists who first worked before 1960 compared with those who first worked after 1960. The effects were greater among those with light complexions compared with those with darker hair and eye color. The findings suggest that further studies are needed to evaluate the role of chronic exposure to low to moderate doses of ionizing radiation in the induction of basal call carcinoma and modification of the risk by pigmentation levels.
Conclusion The committee concludes that currently no epidemiologic or dosimetric evidence that the radiation doses received by either the downwinders or the onsite participants increased their risk of NMSC or melanoma relative to that of the US general population. Thus, there is no epidemiologic justification for adding NMSC or melanoma to the list of compensable diseases under RECA.
Hodgkin’s disease (HD), sometimes called Hodgkin’s lymphoma, is not compensated under RECA and the committee was asked to reevaluate the justification for this exclusion.
HD has not been associated epidemiologically with exposure to ionizing radiation. That finding (Boice, 1992) is supported by more recent reviews (UNSCEAR, 2000), and susceptibility of HD to radiation induction is categorized as very low or absent (Mettler and Upton, 1995).
HD is a rare condition (less than 1% of all new cancers diagnosed in the United States) that is morphologically distinct from non-Hodgkin’s lymphoma, almost never has a leukemic component, and has clinical and histologic features that suggest a chronic infectious process. It is more common in males than females and more common among whites than blacks. It has a bimodal age
distribution with peaks in young adulthood and after the age of 60 years. The overall incidence of HD has been decreasing, but it is increasing in young adults and decreasing more in adults over 40 years old. Survival rates have improved dramatically during the last 3 decades because of improved treatment methods. Various risk factors have been evaluated, including viral infections, particularly those associated with Epstein-Barr virus; occupational exposures to chemicals, particularly those used in woodworking industries and in the form of herbicides and pesticides; genetic factors; and primary immune deficiency (Schottenfeld and Winawer, 1996).
Conclusion Epidemiologic studies fail to show any significant association between HD and exposure to ionizing radiation. The committee found no epidemiologic basis for designating Hodgkin’s Disease as a RECA-compensable disease.
Downwinders and onsite participants are eligible for compensation under RECA for colon cancer, but uranium miners, millers, and ore transporters are not and the committee was asked to reevaluate the justification for this exclusion.
Colon or colorectal cancer is the third most common type of cancer and accounts for 10% of all cancer deaths among men and women in the United States (Jemal et al., 2004). Several risk factors have been identified, including genetic predisposition and other familial and hereditary factors, diet, and inflammatory bowel disease (Schottenfeld and Winawer, 1996).
Colon cancer is moderately susceptible to induction by radiation and has been associated with external exposure to low LET radiation at high doses and high dose rates in the atomic-bomb survivors and in some groups of patients treated with radiation for ankylosing spondylitis or for gynecologic conditions (NRC, 1990). Radiation doses to the colon from internally deposited uranium are, however, very low (UNSCEAR, 2001).
Neither uranium miners nor millers have been found to be at increased risk for colon cancer (Roscoe, 1997; Pinkerton et al., 2004). Roscoe found a deficit in intestinal-cancer mortality among Colorado Plateau uranium miners (SMR = 0.7, 95% CI = 0.4-1.2). Similarly, Pinkerton and colleagues found a deficit in colon-cancer mortality among uranium millers on the basis of 12 deaths (SMR = 0.63, 95% CI = 0.33-1.11). Although a statistically significant increase in deaths from colon cancer (24 deaths, SMR = 1.56, P = 0.025) was reported among workers employed in the United States as radium-dial painters before 1930, when large quantities of radium were ingested, the increase was found not to be related to initial radium intake (Stebbings et al., 1984). Studies by Clarke et al. (1996), Voelz et al. (1997), and Wing et al. (2004) have also failed to find any evidence of an increased risk of colon cancer among worker populations exposed to radiation from internally deposited radium or plutonium.
Conclusion There is epidemiologic evidence of a dose-related increase in the relative risk for colon cancer in the Japanese atomic-bomb survivors and in some patient populations externally exposed to low LET radiations. However, there is no evidence of an increased risk of colon cancer in miners, millers, or ore transporters associated with their internal exposure to alpha-particle radiation. Thus, we found no basis for designating colon cancer in these occupational groups as compensable under RECA.
Testicular cancer is another type that is not compensable in any of the populations designated under RECA or under other radiation compensation programs in the United States. Nonetheless, the committee was asked to reevaluate the justification for this exclusion. The susceptibility of testicular cancer to radiation induction is deemed very low or absent. It is generally regarded as being nonradiogenic, and risk estimates of the incidence of or death from this cancer type have typically not been asserted (Preston et al., 1994; UNSCEAR, 2001; Preston et al., 2003).
Testicular cancer, which is rare in the United States (about 1% of all cancers in males), is more common worldwide among whites than blacks or Hispanics. It is due almost exclusively to undifferentiated germ cells remaining in the testis. Its age distribution is bimodal; embryonal carcinoma peaks between the ages of 15 and 35 years, and declines after the age of 40 years, and a small rise after the age of 75 years is reported because of the increasing incidence of testicular lymphoma. Maldescent of a testis (cryptorchidism) is the major risk factor associated with testicular cancer but is unlikely to be the initiating event. Exposure of the mother to exogenous steroids (such as diethylstilbestrol) during pregnancy may play a minor role in the development of testicular cancer (Roth et al., 1992).
Conclusion No epidemiologic evidence suggests that testicular cancer is induced by ionizing radiation. Thus, there is no basis for designating testicular cancer as compensable under RECA.
Prostate cancer is not listed as a compensable disease for any of the RECA populations. Nonetheless, the committee was asked to elaborate on the justification for its exclusion. Prostate cancer is the most common type of cancer diagnosed and the second most common cause of cancer death after lung cancer (32% in men in the United States in 2004). The American Cancer Society estimates that prostate cancer will account for 13% (about 230,000) of all new cancer cases and 10% (about 30,000) of all cancer deaths in US males (Jemal, 2004). It occurs
only rarely in men less than 40 years old, but the incidence doubles for each decade of life thereafter (Ross and Schottenfeld, 1996).
Under RECA, none of the designated populations is compensated for prostate cancer. On the basis of earlier epidemiologic studies, the susceptibility of prostate cancer to induction by ionizing radiation is very low or absent, especially in connection with the chronic or low exposure potentially experienced by the RECA populations. Studies of incidence and mortality among the atomic-bomb survivors have shown little indication of increased risks of prostate cancer during the 45 years of followup (Preston et al., 1994; Preston et al., 2003). There is no epidemiologic evidence that uranium miners are at increased risk for prostate cancer (Roscoe, 1997).
A study of a cohort of workers employed at the UK Atomic Energy Authority found an increased risk of prostate cancer among a small group of workers who had relatively high total external radiation doses and who also had been monitored for internal radionuclide contamination (Beral et al., 1985). A followup case-control study (136 cases in men with prostate cancer diagnosed between 1946 and 1986 and 404 matched controls) explored the relationship between cases and occupational exposures, particularly to radionuclides (Rooney et al., 1993). Although prostate cancer was statistically significantly increased in men known to be internally contaminated with or at risk of exposure to any of several radionuclides, no association was found with exposure to uranium. The most recent update of mortality (1946 through 1997) in this cohort (Atkinson et al., 2004) found no significantly increased risk of prostate cancer (200 deaths, SMR = 92, 95% CI = 79.9-102.2). That finding is consistent with those of the several followup studies of other cohorts of nuclear-industry workers in the United States, the UK, and Canada (Cardis et al. 1995; Zablotska et al., 2004; Howe, 2004; Howe et al., 2004).
Conclusion There is no convincing epidemiologic evidence that prostate cancer is a radiogenic disease or that the RECA populations are at increased risk for it.
Members of the public asked why downwinders are not eligible under RECA for compensation for uterine cancer but are compensated for ovarian cancer.
The main reason for the apparent discrepancy in compensability is that ovarian and uterine cancers differ in their susceptibility to induction by radiation. On the basis of earlier epidemiologic studies, ovarian tissue is moderately susceptible to radiation induction of cancer, whereas the susceptibility of uterine tissues is very low to absent (Mettler and Upton, 1995). Neither BEIR V, ICRP (1991), nor UNSCEAR (2000) has reported risk estimates for uterine cancer. The most recent report of mortality among the atomic-bomb survivors also failed to show an increased risk of uterine cancer (Preston et al., 2003); the incidence of benign
uterine tumors (Yamada et al., 2004) showed a dose-related increase in risk that decreased with time since exposure.
Conclusion No convincing epidemiologic evidence suggests that uterine cancer is induced by ionizing radiation, so there is no basis for its compensation under RECA.
Nonmalignant brain tumors are not compensable under RECA in any RECA-eligible populations.
A well-documented increased incidence of meningioma has been found in Hiroshima atomic-bomb survivors. The incidence increased with dose and time. The incidence of meningioma among the Hiroshima survivors in 5-year intervals after 1975 was 5.3, 7.4, 10.1, and 14.9 cases 10−5 PY. The incidence classified by a distance from the hypocenter of 1.5-2.0 km, 1.0-1.5 km, and less than 1.0 km was 6.3, 7.6, and 20.0 cases 10−5 PY, respectively. The incidence classified by dose to the brain of 0-0.099 Sv, 0.1-0.99 Sv, and more than 1.0 Sv was 7.7, 9.2, and 18.2 cases 10−5 PY, respectively (Shintani et al., 1999).
A later study by Preston et al. (2002) of the same cohort of atomic-bomb survivors with respect to radiation dose found a statistically significant dose-related excess of nervous-system tumors (NSTs), (ERR/Sv = 1.2, 95% CI = 0.6-2.1), schwannoma separately (ERR/Sv = 4.5, 95% CI = 1.9-9.2), and all NSTs other than schwannomas (ERR/Sv = 0.6, 95% CI = 0.1-1.3). The risks of several of the other individual NSTs, including meningioma, were increased, but the increases were not statistically significant. On the basis of those findings, the authors concluded that there was an increased risk of NSTs even with radiation doses of under 1 Sv (100 rem).
A 2004 study (Yonehara et al., 2004) of the medical histories of some 80,160 Japanese atomic-bomb survivors found about a 6% increase in the risk of some type of tumor in the brain or spinal cord over a lifetime. However, for schwannomas, the risk rose to about 40%, although even with this increase these tumors are rare. Schwannomas are tumors of the nerve sheath and usually occur along nerves of the spine and along the auditory nerve in the brain; they are dangerous because of their location. Of the more than 80,000 survivors, the authors found 55 cases of schwannoma, which is about 20 more than would be expected to occur in a typical population without known radiation exposure. They also found 35 pituitary adenomas and 88 meningiomas. As in the 1999 Shintani study, there was a strong dose dependence for meningiomas (P = 0.004) but no significant correlation between incidence rate and age atexposure.
Conclusion RECA did not include compensation for benign tumors. Two benign brain tumors, intracranial meningiomas and intracranial schwannomas,
have been found to be increased following high dose, high-dose rate exposures in the atomic-bomb survivors. Due to the low brain doses received by downwinders and onsite participants, there is little likelihood that an increased occurrence or incidence will be noted that could be attributed to nuclear-weapons testing.
Several members of families with thyroid disease were brought to the attention of the committee as indicative of cancer clusters related to radiation from fallout. Familial clusters of disease are well known and new knowledge of genetics adds new tools for use in epidemiology inquiries into their pathologic basis. Individual immunologic differences and differences in intensity of exposure to agents in the environment modulate disease susceptibility, and recent advances in genetics make it increasingly possible to identify persons who are more likely than others to develop a disease, whether it is infectious or malignant. Inherited genes confer differences in cancer risk proclivity, and mutations that develop during life also influence individual risk. Ovarian, breast, and thyroid cancer are examples of cancers of which there is familial clustering, with multiple members of a family likely to be afflicted. In some cases, family members share the gene mutation; in others, the genetic basis is less well established. The apparent clustering of thyroid disease in some families in high-dose areas is consistent with the familial linkage that is a common feature of thyroid disease. The frequency of such clusters would be expected to increase in populations in which the incidence of thyroid nodules and thyroid cancer is increased, as in heavily exposed regions.
Knowledge regarding the clustering of cancer in specific people is hampered by many factors: multiple causative agents, long latent periods, inherited familial disease patterns, and the lack of a signature of a specific cause, especially when multiple factors are involved.
It is well known that cancer is common and that common things can aggregate. A simple example illustrates the frequency with which one might expect to sense an apparent cluster of cancers in a family, a neighborhood, or a community (Tversky and Kahneman, 1974; Rothman, 1987). Cancer causes about 25% of deaths in the United States (National Vital Statistics Report, 2002; 50:16:13). In a family of two, if there were no common cause, there would be a 75% × 75% = 56% chance of no cancers, a 18.75% × 18.75% = 38% chance of one cancer, a 25% × 25% = 6% chance of two cancers, and therefore a 44% chance of one or two cancers. In a family of three, there would be a 75% × 75% × 75% = 42% chance of no cancers, a 42% chance of one cancer, a 14% chance of two cancers, a 2% chance of three cancers, and therefore a 16% chance of two or more cancers. In a family of five, there would be a 37% chance of two or more cancers and a 10% chance of three or more cancers. In a neighborhood of 20 people, it can be
shown that there would be a 91% chance of three or more cancers, a 59% chance of five or more cancers, and a 10% chance of eight or more cancers. In a small village of 100 people, it is virtually certain that there will be 20 or more cancers and an 85% chance that there will be 30 or more cancers. Because such statistical clustering may be perceived as having a common cause, it will be worrisome and more likely recalled. The increased availability of examples and the inference of a common cause may lead many people to assume that they were caused by exposure to ionizing radiation whereas in reality they may merely demonstrate that common things are common.
The issue regarding potential cancer clusters is complicated for several reasons:
Many suspect cancer clusters are reported, and almost all are readily discounted on minor inquiry. Every year, state and local health departments are asked to respond to more than 1,000 inquiries on such matters, and they are able to respond to only a few inquiries.
Environmental clusters are especially difficult because it is hard to establish the potential cause, and to select appropriate comparison groups for statistical analysis.
Rare events often appear to come in groups. If the outcome is winning in gambling, that is agreeable and accepted as the way things should be. If the outcome is undesired, such as an adverse health effect, one seeks to identify its cause to avoid further unwanted consequences, and one seeks compensation if it is available. As the population ages, cancer incidence and mortality increase, and increased numbers of cancer clusters among old people are to be expected.
Environmental contamination from any source attracts concern regarding possible adverse health effects. Cancer distribution in populations living near nuclear power plants has been investigated to evaluate possible increases in radiation-related disease, but epidemiologic studies have failed to find a positive correlation; this might be expected because the low radiation doses released in normal plant operations (Jablon et al., 1991).
Conclusion Validated clusters of disease have involved particular exposures resulting in disease of a specific or closely related type (such as infectious disease, and industrial exposures). Many diseases of similar types occur sporadically and most clusters can not be attributed to specific causes. When they occur with increased frequency and in greater intensity, then a causal connection may be established. No evidence of unusual or unexpected clustering cancer or other diseases in exposed populations is known to the committee, although our attention was called to a possible association with multiple sclerosis. Compensation for persons with RECA-eligible disease should be compensated on the basis of their radiogenicity and the PC/AS value adopted by RECA.
RECA compensated downwinders and onsite participants for specific cancers (see Table 2.1). No questions were addressed to the committee regarding the cancers not addressed specifically in this section. Those include cancers of the bile ducts, esophagus, gall bladder, non-Hodgkins’ lymphoma, pancreas, pharynx, small intestine, stomach, and urinary bladder. Risk coefficients increased above those expected in a nonexposed population were the basis for their current RECA compensation, and no suggestions for changes were brought to our attention in the information-gathering meetings or in our review of new published findings. Data from the atomic-bomb survivors provides the major quantitative basis for RECA decisions concerning compensation for those conditions.
Chronic Renal Disease Among Uranium Miners
It was brought to the committee’s attention that under RECA, uranium miners are not eligible for compensation for chronic renal disease (CRD), a nonradiogenic disease, although uranium millers and ore transporters are eligible for such compensation (see Table 2.1). The committee asked to justify that exclusion.
Scientists have established that exposure to high doses of soluble uranium causes CRD and other damage to the kidneys in animals and humans (Hursh and Spoor, 1973). However, insoluble uranium, as generally found in uranium ore, has not been associated with CDR (Clayton and Clayton, 1981). As noted in Chapter 4, uranium millers, in contrast with miners, have the potential for exposure to soluble compounds of uranium during conversion of uranium ore to yellowcake, and renal dysfunction among actively employed millers has been identified, although the clinical significance of the finding is unclear (Thun et al., 1985). Millers may have a potential for the chemically toxic effects of uranium compounds to the kidneys. However, the recently updated studies of mortality among uranium miners (Darby et al., 1995; Roscoe, 1997) and millers (Pinkerton et al., 2004) have failed to show an increase in the risk of CRD in miners or to support earlier findings of a statistically significant risk of CRD in millers. Those studies are limited by small numbers and other methodologic problems in their ability to identify such risks.
Conclusion The committee concludes that there is no good epidemiologic evidence of increased risk of CRD among uranium miners and thus no epidemiologic basis for including it as a compensable disease in miners.
Chronic Lung Disease
Cor pulmonale, pneumoconiosis, pulmonary fibrosis, and silicosis are currently compensated under RECA for uranium miners, millers, and ore transport-
ers. No issues relating to compensation for those diseases were raised at the four information-gathering meetings.
Several people asked the committee for its recommendations directed toward a reappraisal of the justification for reimbursement for specific autoimmune diseases. The pathophysiologic mechanisms underlying autoimmune disorders (such as rheumatoid arthritis, lupus erythematosus, multiple sclerosis, type 1 diabetes, autoimmune thyroiditis, multiple sclerosis, and related diseases) are not well understood, but they are presumed to share some common features, each condition having variable presentations. With the possible exception of autoimmune thyroiditis, none has a clear relation to radiation exposure.
Autoimmune processes with increased circulating antithyroid antibodies can lead to clinically manifest hypothyroidism. A definitive diagnosis of autoimmune thyroiditis according to current WHO standards is complex. It depends on various combinations of specified concentrations of antithyroid peroxidase, antithyroid globulin, and thyroid-stimulating hormone in combination with thyroid abnormalities detected with ultrasonography, palpation, cytology (when available), and surgical pathology (when available). The different studies of irradiated populations rarely provide data adequate to support a definitive diagnosis. And a positive correlation with dose is rarely demonstrated.
Of the many published studies on irradiated populations, the only one in which many of those considerations were addressed involved 27 cases of autoimmune thyroiditis that were observed in 2,587 members of the Adult Health Study (atomic-bomb survivors). A positive correlation was found with doses up to about 0.7 Gy; the correlation inexplicably decreased at higher doses (much below cell-killing levels). None of the studies of other irradiated populations (Chornobyl, Marshall Islands, Southern Utah, and Hanford) has demonstrated a positive correlation between autoimmune thyroiditis and thyroid dose. Current cohort studies by the National Cancer Institute (NCI) of children after the Chornobyl accident may afford a more definitive answer when the prevalence data and dose correlations are analyzed.
Conclusion There are no convincing epidemiologic data from which to calculate radiation risk estimates for autoimmune diseases and thereby to justify their inclusion as a new compensable category of disease under RECA.
Benign Tumors: Thyroid Nodules
None of the RECA populations is compensated for thyroid nodules. Thyroid nodules are common in the general population, and they increase in number with
age. Benign thyroid nodules are more frequent in persons who live in regions with low dietary iodine. The diagnosis of thyroid nodules is based on palpation of a neck mass, ultrasonographic or nuclear-medicine screening, and an open or closed biopsy (typically a fine-needle aspiration) to determine the nature of the lesion. At operation and at autopsy, pathologists frequently find microscopic clusters of cells (micropapillary-microfollicular lesions), which are often referred to as occult cancer, or “pathologist’s cancer,” as opposed to cancers that are thought to affect health adversely. While conducting routine autopsies at the Mayo Clinic, Woolner LB (Surgical Forum, 1954) (Mortensen et al., 1955) found, as they took more sections through the thyroids of persons who died from nonthyroid diseases, a proportionate increase in the number of microscopic nests of thyroid-cancer cells. The literature does not ordinarily classify such lesions as clinically significant. Occult cancers are by definition found only when the glands are removed surgically or at autopsy.
Benign thyroid nodules have been found to be increased after high thyroid doses from radiation beams (typically x rays) and in patients after exposure to iodine-131 (131I). In heavily exposed Marshall Islanders, thyroid nodules were increased by as much as 8 fold more than thyroid cancer especially after childhood irradiation (Robbins and Adams, 1989). The Republic of the Marshall Islands settlement has compensated exposed persons who had thyroid cancer and benign thyroid nodules (the differential compensation depended on whether surgery was performed and, if so, its extent). Among Hanford downwinders, thyroid nodules were increased in frequency, but the magnitude of the increase was not significant at the low 131I doses received (Davis et al., 2004b).
Data on the induction of hypothyroidism after exposure to high radiation doses come from the radiation-therapy literature. No reports of an increase in hypothyroid rates in children who received x ray therapy for other than thyroid conditions (tinea capitis, thymic enlargement, and lymph-node Rx) have been published. It is difficult to attribute diminished thyroid-function (hypothyroidism) to low doses of 131I administered to patients in whom there was reason to suspect the presence of a thyroid disease at the time the dose was administered. After higher doses used in therapy for hyperthyroidism, the results are more clear cut. A large fraction of adults given high doses of 131I in therapy for thyrotoxicosis become hypothyroid. After the first year, thyrotoxicosis patients treated with 131I become hypothyroid at a constant rate of 2.3-4.4% per year (the increment depends on 131I dosage) (Becker et al., 1971). A review of thyroid-function status after 131I therapy for Graves’ disease in 116 patients under 20 years old revealed a similar but steeper response. After doses of around 50 Gy, 30% of the children were hypothyroid in a year. The rate of increase slowed thereafter; 40-50% of the subjects were hypothyroid in 10 years (Read et al., 2004).
In patients with autoimmune diseases, such as Graves’ disease, the body reacts against the gland; by what is now known to be an autoimmune process. It is not clear how much of the decrease in thyroid function after 131I is due to the radiation and how much is due to the immunologic damage to the gland’s functional capacity. It is likely that the same level of damage from 131I to a normal thyroid gland would require a higher dose than would a Graves’ disease patient’s gland. A threshold dose of 2-4 Gy has been postulated for hypothyroidism induced after external beam radiation with x or gamma rays (Williams, 1991). Threshold doses of 10 Gy for hypothyroidism after 131I and 2 Gy after external x- or gamma-ray radiation have also been suggested (NCRP, 1991). Both internal and external photon sources were noted to have induced hypothyroidism with thresholds of about 50 and 20 Gy, respectively. No increased incidence of hypothyroidism has been found among the NTS downwinder, Hanford or Chornobyl populations, and it is unlikely that thyroid doses from NTS fallout could exceed threshold levels in persons without preexisting thyroid abnormalities.
Conclusion Hypothyroidism is increased after high radiation doses delivered externally or internally. Data on the incidence of hypothyroidism caused by 131I in normal populations are lacking. A strong dose-related increase has been noted in patients who have autoimmune thyroid disease. No evidence of an increased incidence of hypothyroidism has been noted in any of the populations exposed to increased doses from 131I. There is no convincing evidence that the incidence of hypothyroidism is likely to be increased by the doses received from NTS fallout by RECA-defined populations.
Type 2 Diabetes
Although Type 2 diabetes (non-insulin-dependent) is not known to be increased after radiation exposure, the disease is reported to be increased in Native American populations. A disproportionately high fraction of uranium miners were Native Americans. The complications of diabetes include CRD, which cannot be differentiated clinically from effects of high doses of uranium in the kidney. The doses of soluble uranium compounds that miners are likely to achieve are unlikely to reach or exceed thresholds at which CRD has been observed. However, the committee considered the possibility that there is a synergistic interaction between diabetes (and its propensity for CRD) and enhanced sensitivity to uranium compounds. The question is being addressed by research at the University of New Mexico, so further consideration of this issue awaits release of the study results.
Conclusion There is no convincing evidence that the incidence of type 2 diabetes is likely to be increased by the radiation doses received by RECA-defined populations.
Cardiovascular Disease and Stroke
A small but statistically significant increase in cardiovascular-disease and stroke mortality (in 1950-1985) with increasing dose was identified among atomic-bomb survivors (Shimizu et al., 1992). The dose-response data indicate that the risk is likely to be negligible below 0.5 Sv. Analysis of data on mortality (1950-1997) from noncancer diseases except diseases of the blood and blood-forming organs has strengthened but not explained the association with respect to nonmalignant circulatory, respiratory, and digestive system diseases (Preston et al., 2003). The increase in risk is about 14%/Sv, which is about 10% less than the increase in the risk of radiogenic cancer and appears not to be influenced by age at exposure or attained age. A review of data from 26 studies of populations exposed to radiation doses of 0-5 Sv, showed that six of the studies had reasonable power to detect a cardiovascular effect if one existed. One study found supporting evidence, but five did not. McGale and Darby (2005) concluded that epidemiologic data have not provided clear evidence of increased risk of circulatory diseases after doses of 0-4 Sv, as had been suggested by the atomic-bomb survivor studies.
Conclusion There is no convincing evidence that cardiovascular or cerebrovascular disease is likely to be increased by the radiation doses of the magnitude received by RECA-defined populations.
Radiation-induced cataracts are known to occur after high doses of ionizing radiation to the lens of the eyes. Most of the reported cases have been in adults who received relatively high doses in occupational exposures. Studies in the atomic-bomb survivors documented cataracts in children within the first 10 years after their exposure; some of the increase was attributed to neutrons. Radiation-protection standards have long recognized this fact and assigned high relative biological effectiveness factors (RBEs) to neutrons and charged particles (ICRP, 1991). Reports on the Chornobyl accident have also suggested an increased incidence in cataracts in adults (Junk et al., 1998).
Well-documented studies carried out in Scandinavia in infants less than 1 year old who received radiation therapy to the face (for treatment of hemangiomas) found a significant increase in cataracts many years later. Typically, there was a big difference in the dose to the two eyes and a strong correlation between cataract frequency and dose to the eyes closest to the beam. The increased frequency of cataracts was best represented by a linear dose-response relationship (Wilde and Sjostrand, 1997). The lowest dose (0.1 Gy) at which very mild effects were noted was higher than that expected from NTS fallout (CDC-NCI, 2001).
Conclusion There is no convincing evidence that the incidence of cataracts is likely to be increased by the radiation doses received by RECA-defined populations.
In Utero Exposure to Radiation
Increased sensitivity of the fetus to the effects of many toxic agents, including ionizing radiation, is well documented. The subject has been reviewed recently and extensively in ICRP (2001a; 2001b) and UNSCEAR (1993, 2000) and in Mettler and Upton (1995, Chapter 8). Much of the detailed information is derived from animal studies buttressed by human data. In the first week of pregnancy, the fetus is growing most rapidly, and fetal sensitivity to radiation and other toxins is highest at this stage. The probability of damage depends on the amount and kind of fetal exposure (radiation, chemical, viral, hypoxia, and so on) and the time during pregnancy when the injury is received. The earliest effect noted in animal studies during the preimplantation period is loss of the shedding of the damaged embryo. Effects induced later in pregnancy depend on the fetal tissues; most sensitive are tissues that are experiencing the greatest growth rate then. Effects can include somatic or germline mutations, congenital malformations, and decreased organ cell mass (the functional impact expressed as decreased functional capacity observed later in life). Mental retardation has been noted in offspring of Japanese atomic-bomb survivors who were exposed during weeks 8-15 in utero, when neuronal migration rates are highest. On the basis of early estimates of dose (gamma and neutron), the risk is compatible with a linear nonthreshold, but a threshold dose of around 20 rem could not be excluded (UNSCEAR, 1993; ICRP, 2001a and 2001b).
Data on cancer after whole-body irradiation of the fetus come from studies of the Japanese atomic-bomb survivors, and from studies of results of using diagnostic medical x-ray during pregnancy (pelvimetry). Two cases of leukemia were observed among 807 atomic-bomb survivors exposed in utero—marginally higher than the number observed in the comparison group (P = 0.054). There was no evidence of a dose response correlation in the in utero period, as there were no cases of leukemia in the higher dose groups (Delongchamp et al., 1997).
X-ray pelvimetry was a common diagnostic procedure in the era before ultrasonography, and magnetic resonance imaging became available. It was used to detect and avoid delivery problems due to disproportions between the maternal pelvic anatomy and fetal anatomy, particularly head size and fetal position. Epidemiologic studies have documented an increased incidence of leukemia in the exposed offspring. Current estimates indicate that the doses received by the fetus were 10-20 mGy and that there was about a 1.4 fold increase in the risk of leukemia thereafter (MacMahon and Hutchinson, 1964); a British study reported an OR of 1.23 (95% CI = 1.04-1.48) (Mole, 1990). The lack of consistency between the findings in the atomic-bomb survivors and in x-ray pelvimetry patients raised concern that the selection of patients for pelvimetry may have been
biased by inclusion of mothers who had problems that foretold complications. In such a case, the maternal condition would have been in some individuals responsible for the procedure—that is, an effect caused the procedure, rather than the procedure causing the effect—thereby compromising the validity of cause-effect conclusions. Additional pertinent discussion is presented in Boice and Miller (1999), Mettler and Upton (1995), and Brent (1999).
The radiation dose to the fetus from radionuclides in the environment depends on the element, on the amount ingested or inhaled by the mother, on whether the material passes via the placenta into the fetus and how long it remains in the different organs. The primary radionuclides of interest from fallout are iodine-131 (131I), cesium-137 (137Cs), and strontium-90 (90Sr).
131I taken into the mother’s body concentrates in her thyroid, and what is in the body that does not concentrate there can cross the placenta and accumulate in the fetus’s thyroid. The dose it receives depends on the mother’s intake and the fetal stage of development. The fetal thyroid first appears as a discernable organ at 10-12 weeks after fertilization, so 131I intake before then is not thought to be detrimental to its subsequent structure or function. The risk of thyroid damage is presumably highest around the transition from the first to second trimester, and 131I continues to be a threat to the fetal thyroid throughout the rest of pregnancy. Additional fetal total-body dose is received from gamma rays from 131I in the mother, but the fetal thyroid dose is 0.1-1% of what the fetal thyroid would have received if the 131I had been there. Similarly, the dose to fetal tissues from 131I in the mother is less than 1% of the dose received by the mother’s thyroid.
In extreme circumstances, very high doses to the fetal thyroid can lead to thyroid ablation and severe problems (cretinism) in the newborn if not rapidly corrected. Smaller doses can lead to depressed thyroid function. The risk of thyroid complications is likely to be highest after in utero exposure, although the magnitude of the risk is not well established. Of the 12 people who were exposed in utero to fallout at the Marshall Islands, two were found to have adenomatous nodules (one exposed to 190 cGy at 10 weeks of gestation, and one to 870 cGy at 23 weeks), and one had a probable occult papillary cancer (after exposure to 110 cGy at 33 weeks) (Howard et al., 1995). Three cases with pathology in a population of 12 (25%) is a remarkably high incidence rate.
Cesium behaves like potassium in the body. It accumulates in circulating blood cells and in muscle. Intake into the body is primarily through the eating of meat from animals that grazed on contaminated lands, but it can also be through the eating of other foods when surface contamination is high. The estimated thyroid
and red marrow doses to an adult from NTS fallout was estimated at 0.009 mGy for those tissues from 137Cs and 0.002 mGy from 134Cs, both of which radionuclides emit beta and gamma rays (NCI-CDC, 2003). Stannard (1988) reviewed research related to fallout radionuclides in the environment. For more detailed data, see http://www.atsdr.cdc.gov/toxprofiles/tp157-c3.pdf, accessed February 28, 2005.
Strontium behaves like calcium in the body and localizes primarily in bone. The organs with the highest dose are bone, bone marrow, and the lower large intestinal wall. More data based on direct measurements in bone are available on 90Sr than on any of the other fallout nuclides (ICRP, 2001a; 2001b). The range of ratios of activity of 90Sr in fetal bones to activity in maternal bones is 0.5-1.0 (Roedler, 1987). The dose to adult bone marrow from NTS fallout to the adult bone marrow is 0.02 mGy.
The Environmental Protection Agency (EPA) has published a set of internal-dose conversion factors for standard persons of various ages (newborn; 1, 5, 10, and 15 years old; and adult) in its Federal Guidance Report No. 13 (EPA, 1999). For example, EPA has estimated that the dose equivalents after ingestion of 1 Bq of 90Sr are 2.77 × 10−8 and 2.77 × 10−7 Sv, respectively, for the adult and infant (assuming an integration time of 50 years for an adult following the initial exposure). For 89Sr, these values are 2.57 × 10−9 Sv and 3.59 × 10−8 Sv, respectively. Age-specific dose coefficients for inhalation and ingestion of any of the radioactive isotopes of strontium by the general public can be found in ICRP 71 (ICRP, 1995) and 72 (ICRP, 1996), respectively. Dose coefficients for inhalation and ingestion of strontium radionuclides can be found in EPA Federal Guidance Report No. 11 (EPA, 1988). Dose coefficients for external exposure to radioisotopes of strontium in air, surface water, or soil contaminated to various depths can be found in EPA Federal Guidance Report No. 12 (EPA, 1993). There is no evidence that 90Sr doses from NTS fallout would rise to the point where increased incidence or mortality from bone cancer or leukemia would be detectable.
Beta or gamma emitters taken into the mother’s body and from external radiation from terrestrial activity during pregnancy contribute to doses to the mother and fetus (whole-body dose, including dose to thyroid, bone marrow, and other organs). Fallout from the NTS weapons tests occurred during a long span of time relevant to an individual fetus’s vulnerable period. It is unlikely that detrimental effects would be discernible from the low doses received during in utero exposure. However, due to heightened radiation sensitivity assigned to the in utero period, it should be added to the NTS testing period during which the mother was eligible for RECA-defined benefits.
Conclusion Sensitivity to radiation exposure is higher in utero than at any other time of life. In the downwinders, the largest dose contribution to the fetus came
from 131I and from external exposure. The contributions from 137Cs and 90Sr taken into the body are significantly less than the other pathways. The committee concludes that it is important that the in utero period be included in determining residence eligibility and radiation dose of NTS releases for all subjects at risk during the fallout-testing interval.
PSYCHOLOGIC CONSEQUENCES OF RADIOLOGIC THREATS
No radiobiologic evidence indicates that interactions between radiation and molecules in the body cause the psychologic conditions or sequelae chronicled in humans after an event that involves a threat or actual release of radiation or radioactive materials. Thus, there is no credible evidence of a positive dose-effect correlation.
In recent decades, however, physicians and other scientists have become increasingly aware of the psychological consequences of a variety of real or perceived threats, including those involving ionizing radiation; they have documented that such responses can seriously affect the health of individuals and communities (IOM, 2003a; Schlenger and Jernigan, 2003). Various articles and reviews have also reported findings of several studies of the psychologic consequences observed among survivors of radiologic threats that occurred before 1990 (Bromet et al., 1990; IAC, 1991; Ricks et al., 1991); these studies confirmed earlier findings of increased emotional, behavioral, and psychologic stress that in some cases have persisted for many years in survivors of radiation accidents.
More recent information on the cascade of effects of emotional threats that involve radiation covers the nature of and risk factors in psychologic responses to such threats and their health consequences (Ginzberg, 1993). Psychologic effects have been described as the major public-health outcomes seen to date of the Three Mile Island (Kemeny, 1979) and Chornobyl reactor accidents (NEA/ OECD, 2002). Separating effects that may have been related to the accidents themselves from those triggered by their social, economic, and institutional effects is difficult; adding to the complexity, in the case of the Chornobyl accident, was the dissolution of the former Soviet Union.
The psychologic outcomes observed among individuals and populations after sudden and unexpected radiation threats or events range from mild transient signs and symptoms of anxiety—in its more severe form, generalized anxiety disorder (GAD)—to major depression or minor depression (dysthymia) and posttraumatic stress disorder (PTSD). Those conditions, of course, are not peculiar to radiation threats or events.
In the aftermath of sudden and unexpected events, those diagnoses have been most prevalent in people with acute radiation injuries, physically unharmed children, mothers of young children, and the disadvantaged. They tend to persist for long periods after the event but gradually decline in intensity; they can be exacerbated by events such as anniversaries and heightened media interest that rekindle
memories of the event. The effects can also be seen in physically unharmed persons at locations remote (low dose) from threats or events. However, their duration declines with increasing distance from the site of the threat (Collins and de Carvalho, 1993; Ursano et al., 1994; Becker et al., 2001; Adams et al., 2002; Bromet et al., 2002; Yamada and Izumi, 2002) and presumably with decreasing intensity of news reports concerning the incident. Lee (1996) introduced the term chronic environmental stress disorder to describe the psychologic effects observed in populations subject to additional stressors associated with an sudden catastrophe, such as the Chornobyl accident.
The populations at risk of radiation exposure from the US nuclear-weapons program with which we are concerned in this report did not, by most accounts, undergo sudden, unexpected catastrophic events, nor did the weapons program cause or threaten serious physical or emotional harm in the short term. Rather, as described to the committee by members of the public, the atmospheric tests often were exciting, apparently safe, sometimes dramatic events and spectacles that were announced in advance. The public and test-site workers observed the events, sometimes with encouragement to do so, with little or no guidance as to personal safety or protection. They and many physicians and scientists were largely unaware, especially in the early years of the nuclear-testing program, of the potential for long-term harm from fallout.
Later observations of unexpected, unaccounted-for health effects, accumulating knowledge and growing public awareness about the delayed health risks of radiation exposure, and gradual loss of trust in authority and professional figures could have prompted considerable stress in those populations and increased their risk of GAD, depression, and other psychiatric disorders. In contrast, uranium miners were working in environments already known to be harmful but without adequate information or protections to limit radon exposure. Their loss or lack of trust in the system as they or their colleagues developed lung cancer and other respiratory diseases that had previously been linked to their work exposures also could have contributed to stress-related anxieties and depression. The committee has been unable to identify any data that show that the psychologic effects of such chronic environmental factors have been evaluated either during the testing and mining periods or, more recently, among the RECA or other populations with similar radiation-exposure experiences.
Anxiety disorders of the types observed most frequently after radiation events also are among the more common psychiatric conditions among the US general population. Many are treatable but often go unrecognized. Prevalence rates from community-based surveys are 1.8-3.3% for depression within the preceding month and 4.9 -17.1% for depression at any time in life (Pignone et al., 2002); an earlier report put the lifetime prevalence of depression at about 20% (Mulrow et al., 1999). The lifetime prevalence of GAD is estimated to be 5% (Fricchone, 2004). It is unlikely that former or current psychiatric disorders in individuals in
the RECA populations are directly attributable to the biologic interactions between radiation and DNA or other cellular targets.
Conclusion There is no convincing epidemiologic evidence that the incidence of mental disorders is increased by the radiation doses associated with the radiation exposure of the downwinders or other RECA-eligible populations. And, there are no data whereby the psychologic effects of the exposure experiences can be evaluated in the RECA populations. Thus, there is no epidemiologic basis of compensation under RECA for psychologic disorders that developed during or since the end of the US weapons programs, but screening for depression may warrant consideration because RECA populations as a group might be at higher risk than the general population.
Based on currently available scientific evidence, the committee recommends that no additional diseases be added to the list of diseases that should be considered for compensation under RECA.
ADDITIONAL POPULATIONS OCCUPATIONALLY AT RISK FOR RADIATION EXPOSURE
Additional Occupational Groups Working in Underground Mines
The committee also considered additional occupational groups at risk for radiation exposure in uranium mining and milling operations, specifically, core drillers and geologists. The committee concludes that core drillers and geologists who worked in the underground mines should be considered in the same category as uranium miners. They worked side by side with the miners collecting samples to assay the ore bodies, and they would have been subject to the same exposures as the miners.
Core Drillers and Geologists Working on the Surface
Many core drillers and geologists were involved in exploratory work on the surface. Using drilling and other techniques, they sampled the subsurface soil to locate and define the extent of ore bodies. The committee could not locate sufficient sampling data on their work environment to evaluate the magnitude of possible exposures. Their work generated dust loading, but much of the material drilled through is overburden rather than uranium ore, so in general the dust would not be expected to have radionuclide concentrations as high as the ore itself. Exposure to radon and its decay products would be expected to be relatively low because the work was on the surface.
Such exposure to airborne dust could lead to a nonradiation hazard and give rise to some forms of nonmalignant respiratory disease, such as silicosis. The severity of those exposures depends highly on the type of soil that is being drilled
and the resulting air concentrations. Crystalline silica in particular has a low threshold limit value time-weighted average as recommended by the American Conference of Governmental Industrial Hygienists (ACGIH, 2004), but the committee is not aware of any epidemiologic studies that indicated an increased incidence of nonmalignant respiratory disease among the workers in question.
Conclusion The committee concludes that there is no convincing evidence that radiation exposure of core drillers and geologists performing exploratory work in uranium areas resulted in adverse health effects. The committee proposes that the National Institute for Occupational Safety and Health or another appropriate government agency conduct a hazard assessment of the conditions in which exploratory core drillers and geologists worked and determine whether there was a significant risk of exposure to hazards linked to RECA-compensable diseases. If so, the committee proposes that these workers be considered for inclusion under RECA.
ADDITIONAL POPULATIONS ENVIRONMENTALLY AT RISK FOR RADIATION EXPOSURE
Nuclear Testing: Downwinders and Onsite Participants
The committee reviewed the locations where nuclear-weapons tests were performed. The current RECA downwinder population is concentrated in the area around the NTS, and the 1997 NCI 131I report (NCI, 1997) dealt with emissions from the NTS.
In RECA, Congress found that fallout from atmospheric nuclear tests exposed people to radiation that is presumed to have caused an excess of cancer and that this risk was borne by these people to serve the national security interests of the United States. The United States has conducted nuclear-weapons tests in areas other than the NTS, and populations exposed to fallout from these tests may also be considered as possible candidates for RECA compensation if Congress so chooses. The tests in question include the Trinity test near Alamogordo, New Mexico, and the Pacific tests. Onsite participants in the tests are already included under RECA, but RECA coverage may be extended to the downwinder populations in those areas.
Over the last several years, there has been a concern about the health effects associated with radioactive fallout that reached Guam during the testing of nuclear weapons in Micronesia. The Pacific Association for Radiation Survivors was formed. In 2002, a blue ribbon panel, authorized by the government of Guam, submitted the Committee Action Report on Radioactive Contamination in Guam between 1946 and 1958.
In March 2004, Robert Celestial provided written and oral testimony to the committee indicating that Guam did receive fallout from nuclear-weapons testing
in the Pacific. He included statements from retired Navy Lt. Bert Schreiber, who testified that “the Geiger counters were off scale” in November 1952. In addition to this, various support ships deployed at Bikini Atoll during Operation Crossroads were sent to Guam and elsewhere for decontamination.
In April 2004, the congressional delegation from the Pacific Island Territory of Guam submitted a petition to Congress to amend RECA to include Guam in the jurisdiction of downwinders and onsite participants.
The committee initiated an independent assessment of the radiologic consequences related to the weapons tests in the Pacific to people living on Guam. The details of the assessment are presented in Appendix C.
Conclusions As a result of its analysis, the committee concludes that Guam did receive measurable fallout from atmospheric testing of nuclear weapons in the Pacific. Residents of Guam during that period should be eligible for compensation under RECA in a way similar to that of persons considered to be downwinders.
The committee concludes that available evidence does not show that the general population of Guam was subjected to unwarranted radiation exposure resulting from the decontamination of naval vessels. Persons who have proof of their employment by a federal agency or its contractor in the process of decontaminating ships affected by fallout are already eligible for compensation as onsite participants under RECA.
Uranium Mining and Milling Materials Used for Construction or Other Purposes
The committee received testimony about the use of pre-December 31, 1971, uranium mine tailings and overburden in home construction. The experience with the use of uranium mill tailings in construction of homes and other buildings in uranium-mining areas indicates the potential hazard of this practice especially given that most people spend most of their time at home and many live in the same homes for decades. Others spend much time working in buildings that may contain tailings. Consequently, even a relatively small exposure rate could lead to an appreciable lifetime exposure.
This is a potentially important source of radiation exposure to the public. In the case of uranium mill tailings, the subject was addressed through the Uranium Mill Tailings Radiation Control Program, which evaluated and remediated “vicinity properties” and inactive mill-tailings piles and mills at 22 inactive mill sites across the United States. Earlier work had also been done by the federal government and the state of Colorado to remediate properties at Grand Junction, Colorado. That work was accompanied by some 4,000 measurements of indoor radon and radon decay products primarily in Grand Junction but also near other uranium mill locations, to determine whether particular buildings had radon con-
centrations high enough to require remediation. Uranium mine tailings would not be expected to have radionuclide concentrations as high as mill tailings, but the potential does exist for above-background concentrations of radium-226 (226Ra) that would cause high radon concentrations in buildings.
The committee recognizes the hazard posed by the use of mine tailings in buildings. Historically, this has been addressed by remedial action programs, including scoping surveys to evaluate the extent of possible problems, and actions, if needed, to restore buildings to acceptable radon concentrations. The question for the committee is whether compensation should be available for people whose health has been affected by the use of tailings in building material.
Compensation is awarded by the US government under RECA because of its responsibility for the direct harm produced by nuclear-weapons development—mining, milling, ore transporting, and activities associated with aboveground testing. Responsibility for harm resulting from the use of mine and mill tailings by others for activities outside the scope of nuclear-weapons production is complex. First, such harm is an indirect rather than direct consequence of nuclear-weapons development. Second, it is plausible that shared responsibility exists for the harm produced, especially in cases in which the potential hazards associated with the tailings were known by both the citizen-procurer and the mill or mine under contract to the United States. The responsibility would rest with the user of the materials, the US government, and possibly the contractor who controlled the materials. Whether one party bore more responsibility than another is, in part, an empirical question about knowledge of harm, control of materials, contractual arrangements, and so on. Third, if all parties were ignorant of the potential hazards associated with use of the materials, the duty to compensate weakens; compensation would be become supererogatory, “beyond the call of duty.”
The subject is ethically and empirically complex. The committee recommends that the appropriate agency reviews the data on radiation exposure levels obtained inside dwellings constructed from mill and mine tailings. The committee also recommends that their findings regarding potential health consequences of such exposures be evaluated to determine whether the PC/AS values based on these exposures rise to or exceed the levels used in RECA compensation.
Emissions from Uranium Mines or Mills
People living near uranium mines and mills may be exposed to airborne and waterborne effluents from them. Resulting radiation-associated risks are expected to be dominated by exposure to radon and its decay products. Radon emissions from underground uranium mines increased during the late 1950s when the mines began intensive ventilation. Radon is released from uranium mill tailings piles, as well as mine and mill ore storage piles. Windblown particulate
emissions were also generated from mine and mill operations, tailings piles, ore storage piles, and overburden storage piles.
Liquid discharges included water from uranium mines where the ore bodies were in aquifers and seepage of process liquids from mills. Radiation doses from those discharges generally are not expected to exceed those from the airborne emissions.
The committee is not aware of any measurements of radon concentrations in offsite areas from uranium mining (as opposed to uranium milling) from 1942 to 1971. Radon concentrations were measured in the 1960s by the Public Health Service around uranium mills at Durango and Grand Junction, Colorado, and Monticello and Salt Lake City, Utah (Shearer and Sill, 1969). Of 44 stations in areas around but not over the tailings piles, only two had above-background annual average radon concentrations of at least 37 Bq/m3 (1 pCi/L), specifically, 37 Bq/m3 (1 pCi/L) and 96 Bq/m3 (2.6 pCi/L). Using the 96-Bq/m3 concentration, an equilibrium fraction of 0.4, and an indoor fraction of 0.7, as used by the EPA (Marcinowski et al., 1994), we calculate a bounding radon exposure of 0.38 WLM/year. A person (nonsmoker) born in 1927, exposed at 0.38 WLM/year for 30 years (1942–1971), and having lung cancer diagnosed in 1990 at the age of 63 years, would have a PC/AS of 0.12 (BEIR VI duration model) or 0.25 (BEIR VI concentration model).
The committee recommends that the appropriate agency reviews historical data on radon concentrations in off-site areas near tailings piles of uranium mills used to produce uranium for the US nuclear-weapons program. The agency should determine whether exposures to those concentrations in off-site areas could result in PC/AS values that meet or exceed the RECA compensation criteria. If so, the agency should take the necessary steps to have these populations included in RECA.
DEFINED INTERVALS FOR WHICH COMPENSATION IS GRANTED
Uranium Mining, Milling, and Ore Transporting
During the public information-gathering meetings, the committee was asked to consider recommending extension of the uranium mining and milling interval in Section 5(a) of RECA. RECA compensation applies to uranium miners, millers, and ore transporters who worked between January 1, 1942 and December 31, 1971. The decision to stop compensation in December 1971 was taken because the United States purchased no uranium for weapons programs after that date. The decision was based on the issue of responsibility and liability rather than radiation exposure and the resulting health effects.
With respect to radon and its decay products, no significant difference exists between exposures in 1971, which are covered under RECA, and those in 1972, which are not. Any decision to compensate uranium workers exposed after 1971,
however, would be based on considerations of limits of responsibility. Such issues would be outside the scientific considerations discussed in this report.
Onsite Participants and Downwinders
The period covered by RECA for onsite participants and downwinders (January 21, 1951-October 31, 1958, and June 30-July 31, 1962) is the period of atmospheric testing of nuclear weapons at the NTS.
Radioactive material was also released from tests at the NTS other than the atmospheric tests. For instance, the underground Baneberry Test in 1970 inadvertently vented and was reported to have released 3.0 × 1015 Bq (80 kCi) of 131I (NCI, 1997). Releases from such tests were generally not as large as those from the atmospheric tests, and the tests were included in the 1997 NCI 131I study.
The committee recommends that the radiation doses and estimates of risks from the radioactive releases from all NTS nuclear weapons tests, including underground tests that resulted in atmospheric releases, be included in determining the PC/AS.
GROUPS AT RISK OF EXPOSURE OUTSIDE RECA’S TIME-SINCE-EXPOSURE INTERVALS
Several people testified to the committee that, in practice, the length of time spent in an affected area or since the first exposure is determined by date of birth and does not include the period in utero. This discrepancy means that some people are ineligible for RECA compensation because they do not meet the existing “time-since-exposure” criterion even though they were in utero and their pregnant mothers were in the area and at risk for exposure during the testing period.
In Chapter 6, the committee has recommended that a PC/AS-based process be used to determine the eligibility of compensation claims for people exposed to radiation in fallout. The PC/AS is determined from an estimate of the radiation dose that a person has received, and this dose must include any dose received in utero. In addition, determining PC/AS will take into account latency periods of each cancer type. Consequently, the committee’s recommendation already considers in utero exposures in determining eligibility.
OTHER ISSUES OF PUBLIC CONCERN REGARDING ELIGIBILITY FOR COMPENSATION
As pointed in Chapter 1, we heard an argument from the Navajo Nation that miners can use affidavits under some circumstances to establish employment history but that millers and ore transporters cannot. People are concerned about their failed attempts at using affidavits in establishing proofs of eligibility. The
Department of Justice confirms that uranium-mining employment history may be substantiated by affidavit under certain circumstances, but the use of affidavits to establish employment as a miller or ore transporter is not allowed (Federal Register, Vol 69, No. 56, pgs. 13630-1). The committee could find no relevant difference to warrant that restriction on the use of affidavits and believes that it creates an unjustified inequity. This is not, strictly speaking, a scientific matter. The Congress may wish to consider re-examining the restriction and allow millers and ore transporters to submit affidavits as proof of employment.
Likewise, we heard argument from the Navajo Nation that an affidavit should be allowed as proof of presence or residence for downwinder claimants. The Department of Justice confirmed that an affidavit is not allowed as proof of presence for Native American downwinder claimants, although they are allowed to use them to establish employment (responses from the Department of Justice to the committee’s questions, March 16, 2004). Members of the Native American community did not have the proofs of presence—such as utility bills, telephone bills, and mailing addresses—that were available to those living off reservations. The Department of Justice works as well as it is able within the current law to help such people. The committee finds a relevant difference between the Native American’s ability to establish residence and that expected of the non-Native American population. To achieve equity of treatment regarding the processing of claims, the committee believes that Native Americans should be allowed to submit affidavits as proof of residency. The Congress may wish to consider re-evaluating this restriction.