V Committee’s Findings Related to NTPR Dose Reconstruction Program

The committee’s evaluation of the NTPR dose reconstruction program considered not only the validity of central and upper-bound estimates of dose for the assumed exposure scenarios obtained in dose reconstructions, but also the approaches used to determine the veteran’s exposure scenario. The committee’s findings regarding scenario determinations, estimates of external and internal doses and related uncertainty, and estimates of total organ doses from all pathways are discussed below, with examples taken from the 99 individual dose reconstruction cases sampled and from reconstructions for other veterans who provided written consent for use of their records. In parallel with the discussions in Chapter IV, Section V.A discusses scenario determination, Section V.B the estimation of external dose, Section V.C the methods of estimating internal dose, Section V.D the dose reconstructions for occupation forces in Japan, and Section V.E the estimates of uncertainty and upper-bound doses from all radiations and exposure pathways combined. Section V.F summarizes the committee’s findings regarding dose and uncertainty estimates obtained by the NTPR program.

V.A DETERMINATION OF EXPOSURE SCENARIOS

V.A.1 Introduction

As discussed in Section I.C, the most important part of the dose reconstruction process is the determination of a participant’s exposure scenario. Because exact histories do not exist for individual veterans, the analyst often has to reconstruct a scenario or a set of possible scenarios on the basis of plausible assump



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 124
V Committee’s Findings Related to NTPR Dose Reconstruction Program The committee’s evaluation of the NTPR dose reconstruction program considered not only the validity of central and upper-bound estimates of dose for the assumed exposure scenarios obtained in dose reconstructions, but also the approaches used to determine the veteran’s exposure scenario. The committee’s findings regarding scenario determinations, estimates of external and internal doses and related uncertainty, and estimates of total organ doses from all pathways are discussed below, with examples taken from the 99 individual dose reconstruction cases sampled and from reconstructions for other veterans who provided written consent for use of their records. In parallel with the discussions in Chapter IV, Section V.A discusses scenario determination, Section V.B the estimation of external dose, Section V.C the methods of estimating internal dose, Section V.D the dose reconstructions for occupation forces in Japan, and Section V.E the estimates of uncertainty and upper-bound doses from all radiations and exposure pathways combined. Section V.F summarizes the committee’s findings regarding dose and uncertainty estimates obtained by the NTPR program. V.A DETERMINATION OF EXPOSURE SCENARIOS V.A.1 Introduction As discussed in Section I.C, the most important part of the dose reconstruction process is the determination of a participant’s exposure scenario. Because exact histories do not exist for individual veterans, the analyst often has to reconstruct a scenario or a set of possible scenarios on the basis of plausible assump

OCR for page 124
tions. Problems arise because “plausibility” can be subjective. It is often difficult, 50 years after most of the atmospheric tests, to verify even a veteran’s participation status with certainty. For example, the original list of veterans provided for the earlier Five Series study (see Section I.B.6) was to have indicated all participants in five test series, but it erroneously omitted more than 20,000 participants and included some 8,000 who were later determined to be nonparticipants. The committee was generally impressed with the extensive historical research carried out by JAYCOR to document the whereabouts and roles of veterans who took part in the testing program. JAYCOR had to locate and piece together deteriorating, obscure, and often almost-unreadable records (morning reports, ship logs, unit histories, and so on) from diverse archival sources. With such sources, the dates of arrival and departure, where a veteran was quartered, and so on, could usually be documented. In contrast, the veteran’s specific duties and the time he spent in various locations (such as on contaminated ships) were typically difficult to document with certainty. Procedures to be followed by the NTPR program for dose reconstructions, as laid out in 32 CFR 218.3, specify that “possible variations in the activities, as well as possible individual deviations from group activities, with respect to both time and location, are considered in the uncertainty analysis of the radiation dose calculations.” There is also an expectation that a veteran will be given the benefit of the doubt in determinations used to adjudicate a claim for a nonpresumptive disease under 38 CFR 3.311. As stated in 38 CFR 3.102, “when, after careful consideration of all procurable and assembled data, a reasonable doubt arises regarding service origin, the degree of disability, or any other point, such doubt will be resolved in favor of the claimant” (see also Section I.C.3.2). In many of the records examined by the committee, however, the participant did not appear to have been given the benefit of the doubt regarding the assumed exposure scenario or film-badge dose, including the time and place of exposure. In reviewing the 99 cases, which were randomly sampled within strata, the committee found at least 20 in which a veteran’s external exposure scenario appeared to be incorrect, incomplete, or suspect (for example, see cases #15, 22, 27, 32, 33, 37, 40, 47, 53, 73, 77, 81, 83, 84, 87, 88, 89, 93, 97, 98, and 99). The inaccuracies were often due to insufficient follow-up by an analyst with the participant or other members of his unit. Examples are discussed below. One tendency the committee saw in the 99 cases was for the analyst to assume that an activity that allegedly violated radiation safety (rad-safe) or operational guidelines in place at the time did not happen. For example, an analyst often assumed that decontamination crews did not stay longer than the allowed times on contaminated ships, that radiation safety monitors and other personnel did not go beyond the 10 R h−1 demarcation line, or that badges that were issued and then returned had, in fact, been worn (not left in a drawer). If the date of issue of a film badge was missing, it was often assumed to have been the recorded date of turn-in of the veteran’s previous badge.

OCR for page 124
Such pragmatic assumptions reflect the analyst’s need to complete the calculations and seem also to reflect a tendency to idealize human behavior, particularly military behavior. Such assumptions tend to deny that chaos, confusion, and a perceived need among leaders to ignore rules to complete the task at hand may drive what happens in the field, particularly when a nuclear weapon has just been detonated. The commander of a decontamination crew may have been focused on getting a ship decontaminated and may have considered the rad-safe guidelines to be unnecessarily restrictive and thus not to be taken literally. The rad-safe limit line was not “drawn in the sand,” and forward units were sometimes unsure about their exact location relative to that line and to ground zero. Communication of radiation intensity from rad-safe monitoring personnel to commanding officers in the field was sometimes unreliable. Generic estimates of shielding and time spent indoors versus outdoors used to estimate external dose are questionable for some participants. For example, some participants on ships claimed that because of the heat they slept on deck, where they would not have been shielded at all (see case #28). The assumed 50% shielding factor for participants on Pacific islands may be too high for those who were billeted in tents or thin metal structures that may have had many open windows at night (see Figures V.A.1 and V.A.2). Thus, as discussed later in this chapter, generic dose estimates on ships and islands may not be reasonable estimates of the doses to some unit members. FIGURE V.A.1 Typical metal buildings used at Enewetak during Operation CASTLE.

OCR for page 124
FIGURE V.A.2 Tents on Parry Island at Operation CASTLE. Some sources of information about veterans were not used as well as they might have been. For example, it is not apparent that information in “File A” (see Section I.B.3) for individual veterans was always considered. Additionally, the veteran himself and his buddies were rarely contacted, nor were civilian radiation-safety personnel who often accompanied participant groups during planned activities. That approach might reflect a difference in worldview between a researcher and a claims adjudicator or government contractor, but it is our view that additional and sometimes useful information could have been obtained from the veterans themselves. The questionnaire that was administered in the early days of the NTPR program was very sketchy. It included such questions as “Were you issued a badge?” and “Did you wear it?” When questions came up in the scenario reconstruction about what specific activities a veteran was involved in, the veteran apparently was almost never asked for clarification. The committee’s impression is that the contractor assumes that the veteran himself should not be regarded as a reliable source of information. When, on occasion, a veteran came forward with an account of what happened on the sometimes-chaotic day of a weapon test, his story may have been discounted by the analyst and may not even have influenced the calculation of uncertainty, that is, the assigned upper bound of the dose. Examples illustrative of those points are detailed below. V.A.2 Discussion of Selected Cases Illustrating Scenario Determination Problems In this section, we discuss some of the 99 sampled cases and additional files submitted by veterans. These cases are listed in Appendix B.

OCR for page 124
Case #22: The participant claimed that he was present at Operation IVY. However, his service records had been damaged, and his claim that he participated in IVY could not be verified. He was not given the benefit of the doubt in evaluating his claim for a nonpresumptive disease, and no dose was calculated for possible participation in IVY. Nor was the estimated upper bound of his assigned total dose (from his participation in other test series) adjusted to reflect his possible participation in IVY. He was not contacted to investigate his claim further. Case #53: This case provides a good example of inconsistent application of assumptions used in estimating the external dose and upper bound from boarding target ships at Operation CROSSROADS. The dose memorandum states that the veteran was given the benefit of the doubt by assuming that he participated in two-thirds of the target-ship boardings by his unit. However, the calculations in the case file are based on only one-third of the boardings. In other cases involving target-ship boarding (for example, cases #45 and 49), the veterans were usually given the benefit of the doubt by assuming that they participated in all boardings (see Figure V.A.3). FIGURE V.A.3 Sailors sweeping deck of ship.

OCR for page 124
Case #77: This veteran was a member of the 50th Chemical Platoon at Operation TEAPOT, and much of his film-badge information has been lost. From film-badge data summaries that have been found, it is known that several members of the 50th Chemical Platoon, which made up the Desert Rock Radiological Safety (Rad-Safe) Section for TEAPOT, received external doses that greatly exceeded the operational limit of 6 rem, but it is not known who those individuals were. The veteran in question was informed that a reconstructed dose of 3.12 rem was his “most probable dose,” but he was given the benefit of the doubt by assigning him the operational limit of 6 rem instead. The fact that no upper bound was provided implies that the dose of 6 rem would be considered as a 95th percentile of this veteran’s dose in any adjudication process (the veteran did not file a claim for compensation). The veteran’s personal narrative was provided to the analyst. He stated that he was assigned as rad-safe monitor for two colonels from the Pentagon, who were “dressed in silver suits covering every part of their body, including shoes. They taped all seams with a tape comparable in appearance with duct tape. I watched all this while wearing only a T-shirt and fatigue pants. I was curious as to what they knew, what I didn’t know, and what they weren’t telling me.” He goes on to describe what happened next (apparently, this incident occurred at Shot MET): About ten (10) minutes after detonation of the 22 kiloton device, I entered the blast area to find instruments the two Colonels had placed in the area. (We had previously met to acquaint me with the location and critique the recovery.) When I arrived at the site it was very dark, dusty and windy. I can’t recall the exact readings, but they were high. I returned to meet the Colonels, who had driven their van onto a road leading into the site. I reported that the recovery area was very hot, and they would have to work very quickly. I led them back to the instrument location. They recovered their instruments and packed them into boxes. In the recovery area fires were still burning. There was a lot of smoke and dust, and the mushroom cloud was still visible. The ground around us was black. The winds were strong. We had passed several mannequins burning, and I learned later this was a test of fireproof clothing. The mannequins burned and the clothes did not. We were several minutes in the area. We then left and I never saw the two Colonels again. I stopped to brush myself off, as I was covered with dust. I made a note of the time spent in the area. I remember thinking that the two Colonels had exceeded 5 Roentgens – more like 6 – and that my double trip into the area would place me even higher. (For example, see Figure V.A.4.) The veteran goes on to give details about several other tests, one of which again suggests the potential for an inhalation dose: One of the major studies undertaken by the 50th Chemical Platoon was to try to correlate a radiation pattern between the ground and the air. In order to be sure these readings from the air were accurate, it was necessary to have men on the

OCR for page 124
FIGURE V.A.4 US Army observers examining dummies set up about 3,000 yards from ground zero during dry run for Operation TEAPOT Shot MET. ground to check them. As part of this group, I was assigned to be a ground monitor. The exercise took place at a site where a nuclear detonation had occurred. I am unsure of the exact reading, but our location was radioactive enough to gather data from aircraft flyovers. After the first series, it was decided that the aircraft probe was not accurate. We stopped for a couple of days while a lead shield was built to protect the probe in the aircraft from every angle except straight down. We then spent a few more days testing this new device. Adjustments were made, and we were in and out of the area several more times. We took readings all the way to ground zero. This exercise was the dirtiest of my stay. Every day we were covered with dust from our travels through the test site. We had no protection and were inhaling dust constantly. I remember thinking our lungs must have looked like our clothes. I do not remember if we had film badges. He then describes an operation (apparently at Shot APPLE-II) in which he became disoriented near ground zero: During the test known as the Survival City Shot, I was assigned to locate a large group of military vehicles. I made several trips through the area prior to the test to orient myself to the location of these vehicles. They consisted primarily of 2 1/2 ton – 3/4 ton trucks and jeeps.

OCR for page 124
I especially remember the layout of Survival City with its city street and completely furnished houses. There were even families of mannequins set in the houses. There was a two story brick building which had been built especially for the test. It was kind of a landmark because it was the tallest structure on the desert except for the bomb towers. Farther from ground zero was a completely equipped mobile home park. A large number of civil defense people were at this test. I entered the test site shortly after the blast, with a team, seeking the ten Roentgen line [10 R h−1]. I could not find the vehicles. They had been parked less than a mile from ground zero. The ground was black, the two story building was gone, and I became disoriented for a few minutes as I drove around looking for some trace of the vehicles. While I was looking, a call came over the radio that all troops were being pulled from the area due to a wind shift. When I found my way back I had been inside the ten roentgen line. I did not stop my jeep to take a reading. I was alone at this time, and was relieved to find my way back. I believe my exposure was quite high for this event. It was very windy, with dust and smoke. I had no protective clothing or equipment. The analyst only peripherally considered this narrative in the dose reconstruction. Regarding the first account, about accompanying the two colonels after Shot MET, the analyst writes that the veteran “did not provide sufficient information to identify the specific project that he supported on shot day.” Because the veteran commented on seeing burning mannequins, the analyst decided to assign him to Project 40.20, the Clothing Test Project, and accordingly assigned him a dose of 0.20 rem appropriate to that group, apparently discounting the veteran’s statement that “I learned later this was a test of fire-proof clothing.” Evidently, no inhalation dose was considered. Regarding the project to assess the correlation between readings on the ground and air-based readings, the analyst comments that although the veteran described this as a “major study,” “such a project is not listed, per se, among the Desert Rock projects at operation TEAPOT.” The closest documented match that the analyst could find was Project 40.19, CBR Defense Team Training, and the veteran’s dose from that activity was accordingly based on a reconstruction that had been done for that group, with the comment that his “dose resulting from this activity was certainly less than 1.7 rem.” Again, no inhalation exposure was considered, nor was any allowance made for the possibility that the veteran’s account may reflect an activity that was not represented in other surviving records from the time. A note in the file states that because this veteran was a PFC (private first class) at the time, he could not have been involved in CBR team training and, therefore, the dose of 1.7 rem noted above should be subtracted from his dose. However, the 1.7-rem piece of his dose was not replaced with a more accurate estimate. Regarding the third narrative, related to Survival City in connection with Shot APPLE-II, the analyst found other records that supported the veteran’s

OCR for page 124
claim that he was involved. However, some details of the veteran’s account were evidently discounted. The analyst’s report states that: the scenario is questionable since rad-safe monitors did not travel alone in jeeps and there was no reason to send anyone into the shot area to ‘search for’ the test vehicles since their locations were well known. Moreover, it was not the function of the 50th Chemical Platoon to locate vehicles, but merely to accompany project personnel who were to evaluate damaged vehicles. The analyst goes on to assign the veteran a dose for this shot on the basis of a reconstruction that had been done for 573rd Ordnance Company personnel and accompanying rad-safe monitors. In the end, the analyst made an argument that the veteran’s overall dose could not have exceeded the operational limit of 6.0 rem. The argument was based on information that seven members of the 50th Chemical Platoon evidently did exceed the limit and were restricted from further radiation-related work, but this veteran evidently was not restricted. The analyst states that “the dose calculation … does not consider [the veteran’s] allegation that he became disoriented while searching for some test vehicles and spent a few minutes in a high-radiation area. The dose resulting from such an excursion cannot be estimated without more specific information.” This narrative illustrates two points. First, if given the opportunity, veterans sometimes can provide detailed and compelling accounts about their experiences. The men who participated in these atomic tests knew that they were making history at the threshold of the nuclear age. Although memory is not totally reliable, such experiences are not easily forgotten. Second, although it is inherently difficult for an analyst to take scenario uncertainty into account quantitatively, a better effort could be made to acknowledge that such uncertainty exists and to account for it. Although the committee did not try to recompute the veteran’s dose, there was consensus that his true external dose could have greatly exceeded the assigned 6 rem, and that there was also the potential for substantial inhalation dose and beta dose to the skin, exposure routes that were not considered. Contributed case: Another example, not among the 99 sampled cases but a record that was randomly pulled from the Science Applications International Corporation (SAIC) files and then used with the permission of the veteran, concerns an Air Force helicopter technician. In this case, assumptions made throughout the dose reconstruction did not appear to give the veteran the benefit of the doubt. Other personnel involved, whose names and ranks were provided to the analyst by the veteran, could have provided supplemental information, but the record does not indicate that any follow-up contacts were attempted. The case is particularly interesting because it involved highly unusual, or possibly unique, conditions of exposure, which can place considerable demands on the analyst in developing an exposure scenario that fits the particular circumstances.

OCR for page 124
The veteran had been trained in maintenance of F-84G aircraft that were used for cloud sampling after nuclear detonations in the South Pacific (see Figure V.A.5). He arrived at Kwajalein on September 30, 1952, and was present for both detonations in Operation IVY. After Shot MIKE (November 1, 1952), two F-84G sampler planes had to leave the radioactive cloud because one got into trouble and “went into a spin” and the other followed it. The first one could not return to land and the pilot went down in the sea with his plane. The other plane just made it to Enewetak but had a rough landing, blowing out two of its tires. The veteran was flown to Enewetak to change the wheels and tires, refuel the plane, and use a power source to restart its engine so that it could return to Kwajalein. The downed F-84G must have still been holding its very hot air samplers on its wings and nose. On his return to Kwajalein, the veteran recalled that he required more than 4 h of showering before the Geiger-counter reading on him came down to acceptable levels. The veteran’s initial dose reconstruction, as reported to him in 1983, assigned him a dose of 0.000 rem. He complained right away. In 2000, he filed a claim for service-connected disability. The analysts revisited the calculations at that time, and a revised dose assessment was reported. The second dose reconstruction began with the fact that 4 days after Shot MIKE, the external exposure rate at 4 in. from the pylon of the F-84G that he had serviced was recorded as 0.10 R h−1. That was extrapolated back in time (on the basis of a decay rate of t−1.2) to the time when the veteran would have been on Enewetak changing the tires, but this extrapolation evidently did not take into FIGURE V.A.5 F-84G cloud-sampling aircraft.

OCR for page 124
account the presence of cloud samples while the veteran was working on the plane and the likelihood that the plane lost some of its radioactivity in a washdown after its return to Kwajalein. It was assumed that it had taken him 1 h of work close to the hot plane to get both tires changed, refuel it, and restart it. It was assumed that the veteran spent that time near the landing gear at a distance of 1 m from the contaminated fuselage (“his arms being extended”). The landing gear and blown tires were assumed to be uncontaminated because they would have been “tucked inside” the plane. The committee did not attempt to do a dose reconstruction for the veteran, but the committee took issue with every assumption that was applied and considers the assigned upper bound of 0.8 rem to be much too low to adequately reflect the uncertainties in scenario definition and estimation of dose. The extrapolation of the measured exposure rate backward in time is complicated by the fact that the plane would have had its highly radioactive air samplers removed immediately on its return and the possibility that it cooled off during the 2-h flight back to Kwajalein and was hosed down before day 4 to begin its decontamination. Elsewhere in the dose reconstruction report, the analyst calculates doses that the veteran might have received in later work where he decontaminated F-84s, mentioning that the planes were routinely decontaminated within a day of their return from flying through the mushroom cloud. The analyst states that: During the mornings following both shots (2 November and 17 November) the F-84G aircraft were moved to a decontamination ramp at Kwajalein, where they would be thoroughly scrubbed and washed down. The average radiation intensity upon landing of the F-84G’s was 2.5 [R h−1]. As the readings were taken of various aircraft parts, the average was likely indicative of radiation levels at 4 inches from the surfaces of aircraft components that personnel were likely to spend the majority of their time maintaining. Engine/intake area decontamination effectiveness was about 50 percent; smooth surfaces were about 95-98 percent. The highest surface contamination zones on the aircraft were leading edges, air intakes, and engines. Even if the wings were decontaminated with an effectiveness of only 90%, it follows that the measured reading of 0.1 R h−1 on day 4 should have been multiplied by 10 before extrapolating it back to shot-day levels. On that basis, it seems reasonable to suppose that the estimated dose during the tire-changing event was too low by at least a factor of 10. Again, this conclusion does not take into account the presence of air samplers, which would increase the extent of underestimation of the veteran’s dose. Other assumptions made in the scenario reconstruction do not seem to give the veteran the benefit of the doubt. The assumption that the landing gear and tires were not contaminated seems doubtful. Potentially, the well in which the landing gear is housed during flight may serve as a trap for radioactive particles. The metal cover over the wheel well swings down when the landing gear is extended, and the cover presumably was contaminated. Finally, the assumption

OCR for page 124
sures of most participants were due mainly to inhalation and that intakes by ingestion usually were insignificant. The committee recognizes that estimation of inhalation doses to atomic veterans is difficult. Given the lack of data on airborne concentrations of radionuclides at locations and times of exposure and data on amounts of radionuclides excreted in urine or feces, inhalation doses can be estimated only by using indirect methods that involve substantial uncertainty. It also is likely that in some exposure scenarios, such as those involving exposure to suspended neutron-activation products in soil at the NTS or exposure to descending fallout at the NTS or on residence islands in the Pacific, inhalation doses were inconsequential compared with external doses that could be monitored with film badges or field instruments. In scenarios in which inhalation doses should be much lower than external doses, uncertainties in methods used by the NTPR program to estimate inhalation dose are unlikely to be important. The committee’s detailed evaluation of methods used in the NTPR program to estimate inhalation doses to atomic veterans is given in Section V.C.3 and summarized in Tables V.C.5 and V.C.7. In some respects, the methods should tend to overestimate inhalation doses. In other respects, however, the methods involve substantial uncertainty or they should tend to underestimate inhalation doses to such an extent that it is often difficult to determine whether estimated doses to atomic veterans are credible upper bounds, as intended by the NTPR program. Furthermore, the committee has identified exposure scenarios in which neglect of resuspension of previously deposited fallout by the blast wave produced in most detonations at the NTS almost certainly has resulted in underestimation of upper bounds of inhalation doses by a factor of at least 100. Such scenarios are important because thousands of participants at the NTS could have been exposed to substantial airborne concentrations of fallout that was resuspended by a blast wave. The committee also identified other cases in which an inhalation dose of zero was assigned to an organ in which a veteran’s cancer occurred but there is little doubt that there was some inhalation exposure. On the whole, the committee has concluded that methods used in the NTPR program to estimate inhalation doses to atomic veterans have important shortcomings that center around three issues. [1] Most estimates of inhalation dose to participants at the NTS and in the Pacific depend on estimates of concentrations of radionuclides deposited on the ground or other surfaces or distributed over a depth in surface soil at locations and times of exposure. Those estimates are based, in part, on measurements of external photon exposure with film badges worn by participants or field instruments, combined with calculations of external exposure rates per unit concentration of radionuclides on the surface. However, especially in scenarios involving exposure to descending or resuspended fallout, the reliability of the methods of estimating concentrations of radionuclides

OCR for page 124
that are important contributors to inhalation dose has not been demonstrated and therefore is unknown. Methods of estimating inhalation doses based, in part, on measured external photon exposures were criticized by a previous committee of the National Research Council (NRC, 1985b). The essence of the criticism was that the methods lacked scientific credibility and that their reliability is therefore unknown. Similarly, on the basis of an evaluation of simultaneous measurements of airborne concentrations and ground deposition at the same locations near the NTS during periods of atomic testing, Cederwall et al. (1990) concluded that the relationship between airborne and surface concentrations of fallout is too complex to be treated adequately by simple approaches, such as use of a deposition velocity. The present committee shares those concerns about the reliability of methods used in the NTPR program to estimate concentrations of radionuclides that are potentially important contributors to inhalation dose. The previous National Research Council committee suggested that urinanalysis should be used to assess the validity of methods used in the NTPR program to estimate internal dose (NRC, 1985b). The present committee also believes that some indication of reliability is essential if estimates of inhalation dose are to be considered credible. However, because of experience with a bioassay program that was recently undertaken to assess internal exposures to plutonium and difficulties with the use of present-day measurements to estimate intakes that occurred many years ago, as discussed in Section VI.D, the committee believes that urinanalysis is not likely to provide useful information on the reliability of methods used to estimate inhalation doses to atomic veterans. A potentially more fruitful approach would be to compare estimated radionuclide concentrations in deposited fallout or in neutron-activated soil used in the NTPR program with measurements that were made at the NTS or in the Pacific after the period of atomic testing ended. As noted in Section V.C.3.2, comment [6], radionuclide concentrations in surface soil over portions of the NTS that were affected by fallout were measured extensively during the 1980s. Important constituents of fallout on which data were obtained are 241Am, 238Pu, 239,240Pu, 60Co, 90Sr, and 137Cs. Although data on shorter-lived radionuclides in fallout are lacking, measurements of longer-lived constituents and knowledge of the relative activities of different fission and activation products that were produced in each shot presumably could be used to assess the reliability of estimated concentrations of all radionuclides in deposited fallout that are used in dose reconstructions. An illustration of the importance of those data is provided by an analysis presented in Appendix E. In addition, later measurements of 152,154,155Eu in surface soil could be used to assess the reliability of estimated concentrations of activation products at the NTS. Similarly, concentrations of radionuclides in fallout deposited on residence islands in the Pacific have been estimated in many studies, some of which began

OCR for page 124
during the period of atomic testing (see, for example, Wilson et al., 1975; Robison et al., 1997b; Simon and Graham, 1997; Donaldson et al., 1997). Those data could be used to assess the reliability of estimated concentrations of radionuclides in fallout deposited on residence islands that are used in dose reconstructions and the potential importance of inhalation doses, and they may also be useful in assessing the reliability of estimated concentrations of radionuclides in fallout on ships. The potential importance of the data is illustrated by the following example. On residence islands at Enewetak Atoll, the total deposition of plutonium reported by Wilson et al. (1975) is about 0.3-25 nCi m−2. If we assume that those data define a 90% confidence interval of plutonium concentrations and use the same assumptions about uncertainties in parameter values as in the example analysis of a scenario involving resuspension caused by walking or other light activities discussed in Section V.C.3.3—except that an assumption about fractionation is not needed when concentrations of plutonium on the ground are measured—we find that a central estimate of inhalation dose to the lung is about 10−4 mrem h−1, and an upper bound (95th percentile) of a probability (uncertainty) distribution is about 0.02 mrem h−1. Those results indicate that inhalation doses due to resuspension of longer-lived radionuclides in fallout deposited on residence islands in the Pacific are unlikely to be important in most cases. That conclusion is supported by later assessments of doses to native Marshall Islanders from inhalation of plutonium (Robison et al., 1997b; Sun et al., 1997b). Knowledge of amounts of shorter-lived radionuclides in fallout relative to plutonium could be used to infer possible inhalation doses due to resuspension of all radionuclides deposited on residence islands. The committee is particularly concerned about two assumptions used in the NTPR program to estimate concentrations of radionuclides in fallout deposited on the ground or other surfaces. The first is an assumption of no fractionation of radionuclides in fallout except for removal of noble gases. That assumption almost certainly results in substantial underestimates of concentrations of refractory radionuclides (such as plutonium) in fallout at the NTS and in the Pacific. An assumption of no fractionation is especially important at the NTS because accumulation of fallout plutonium during the period of atomic testing presented an important inhalation hazard to thousands of participants who engaged in activities in forward areas. The second is an assumption, used to calculate external exposure rates per unit concentration of radionuclides in deposited fallout, that the source region is a surface of infinite extent. That assumption is reasonable at the NTS and on residence islands in the Pacific, but it probably results in underestimates of concentrations of radionuclides in fallout deposited on ships. Estimates of concentrations of radionuclides on the ground or other surfaces used in dose reconstructions are of crucial importance because calculated inhalation doses in most scenarios depend on those estimates. The committee is not aware of any efforts by the NTPR program to assess the reliability of those estimates at the NTS or in the Pacific. If a key element of a method on which

OCR for page 124
estimates of dose depend has unknown reliability, all estimates of dose based on the method are called into question unless it can be demonstrated by other means that the method as a whole most likely results in substantial overestimates of dose. The committee does not believe that it has been demonstrated that the method as a whole tends to overestimate inhalation doses. [2] An important deficiency in dose reconstructions for many participants at the NTS is the lack of consideration of resuspension of previously deposited fallout by the blast wave produced in aboveground detonations. When combined with the frequent neglect of aged fallout that accumulated at the NTS during the period of atomic testing and the general neglect of fractionation in fallout, neglect of resuspension caused by a blast wave could result in underestimates of upper bounds of inhalation doses by a factor of at least 100 in some scenarios in which participants engaged in activities in forward areas within a few hours after a shot, and perhaps by a factor of as much as 1,000 in the worst cases. The issue of neglect of resuspension caused by the blast wave produced in a detonation in all dose reconstructions at the NTS and the possible degree of underestimation of upper bounds of inhalation dose in some scenarios due to neglect of blast-wave effects are discussed in Section V.C.3.2, comment [7]. The committee believes that neglect of effects of a blast wave on inhalation exposures of participants in forward areas after detonations at the NTS, combined with the frequent neglect of aged fallout that accumulated during the period of atomic testing at the NTS and neglect of fractionation in fallout, is an important deficiency for which there is no apparent explanation. The potential importance of resuspension caused by a blast wave on inhalation doses is demonstrated by an analysis in Appendix E. Neglect of blast-wave effects is important not only because of the likelihood of large underestimates of inhalation dose but also because thousands of participants at the NTS (maneuver troops and close-in observers) probably were exposed to fallout that was resuspended by a blast wave, and credible upper bounds of doses to organs of concern could have exceeded 1 rem in many cases. [3] Dose coefficients for inhalation of radionuclides (equivalent doses to specific organs and tissues per unit activity intake) have substantial uncertainty that has not been taken into account in the NTPR program. In the worst cases, such as the dose coefficient for the lung from inhalation of plutonium, a credible upper bound of a dose coefficient based on current ICRP recommendations and a full accounting of uncertainty is more than a factor of 10 higher than values used in dose reconstructions for atomic veterans. Dose coefficients for inhalation of radionuclides are uncertain because of uncertainty in the associated dosimetric and biokinetic models and in the biological effectiveness of alpha particles. Evaluations of those uncertainties have been

OCR for page 124
available for use in dose reconstructions at least since 1994 (see Section V.C.3.2, comments [1] and [2]). The conclusion that the upper bound of a dose coefficient for inhalation could be underestimated by a factor of more than 10 in the worst cases takes into account the presumed bias of most dose coefficients used in dose reconstructions to overestimate dose when a particle size (AMAD) of 1 μm is assumed (see Tables V.C.1 and V.C.2). The substantial uncertainty in dose coefficients is important because it affects all calculations of inhalation dose to participants. Uncertainty in dose coefficients should be acknowledged and taken into account in the NTPR program if credible upper bounds of inhalation doses to atomic veterans are to be obtained. V.C.6 Conclusions on Credibility of Estimated Upper Bounds of Inhalation Dose All estimates of inhalation dose to atomic veterans obtained in the NTPR program are reported as single values without uncertainty, and those estimates are intended to provide upper bounds of possible inhalation doses. Thus, the key question in evaluating methods of estimating inhalation doses used in dose reconstructions is whether the methods provide credible upper bounds. If they do, estimates of inhalation dose to atomic veterans are appropriate for use in evaluating claims for compensation for radiation-related diseases. However, if estimates of inhalation dose are substantially less than credible upper bounds, the veterans are not given the benefit of the doubt and, depending on the magnitude of possible doses from all exposure pathways, their claims for compensation may not be evaluated fairly; that is, a veteran’s claim could be denied even though a credible upper-bound estimate of dose, taking all exposure pathways and uncertainties into account, would qualify the veteran for compensation. As discussed in Section V.C.3.3, the committee does not believe that the question of whether estimates of inhalation dose obtained in the NTPR program are credible upper bounds can be given a single answer that applies to all exposure scenarios for participants at the NTS and in the Pacific. However, partly on the basis of conclusions obtained in previous reviews by committees of the National Research Council (see Section V.C.2), the NTPR program has often claimed that its methods of calculating inhalation dose provide overestimates of dose (the doses are “high-sided”), the implication being that the claim applies generally (see, for example, Schaeffer, 2001b). Therefore, the question is whether the methods of estimating inhalation doses provide credible upper bounds in all or nearly all cases. The present committee’s review of methods of estimating inhalation dose used in the NTPR program has been considerably more extensive than previous reviews by other committees of the National Research Council. The present committee considered many issues involved in estimating inhalation doses that were not evidently considered in previous reviews. Furthermore, the present committee had

OCR for page 124
access to documentation of methods that was not available when the first review was conducted in 1985; and for the first time, extensive and detailed evaluations of dose reconstructions for individual veterans who filed a claim for compensation or who requested information on their doses were conducted. On the basis of its review, the present committee has reached a different conclusion about methods of estimating inhalation dose used in the NTPR program from the one based on previous reviews. Its conclusion is summarized as follows: Methods used in the NTPR program to estimate inhalation doses to atomic veterans do not consistently provide credible upper bounds. Furthermore, the extent of underestimation of upper bounds is a factor of at least 100 in important scenarios involving maneuver troops and close-in observers at the NTS who were exposed to old fallout that was resuspended by the blast wave produced in a detonation. There are some important scenarios in which estimates of inhalation dose obtained in dose reconstructions probably are credible upper bounds, as intended by the NTPR program. An example of such a scenario discussed in Section V.C.3.3 is exposure to descending fallout throughout the period of descent on residence islands in the Pacific when cancer in an internal organ other than an organ in the GI tract is the disease of concern, although an unequivocal conclusion is difficult even in this scenario because of the unknown reliability of methods used by the NTPR program to estimate concentrations of radionuclides in descending fallout. It also seems likely that estimates of inhalation dose in scenarios at the NTS involving suspension of neutron-activation products in surface soil are credible upper bounds, given that assumed resuspension factors are likely to be considerable overestimates for radioactive materials that are fixed in soil. Estimates of inhalation doses to occupation forces in Japan discussed in Section IV.D also should be credible upper bounds if they are based on an assumption that exposure occurred only at locations of highest fallout. However, the types of exposure scenarios for which estimates of inhalation dose obtained in dose reconstructions probably are credible upper bounds are somewhat limited. In many frequently occurring scenarios, such as scenarios of exposure to previously deposited fallout in forward areas at the NTS, the committee believes that uncertainties in assumptions used to estimate inhalation dose are sufficiently important that doses estimated by the NTPR program may not be credible upper bounds even if some parameter values used in the calculations, especially resuspension factors, are credible upper bounds. Even in scenarios involving exposure to descending fallout, exposure during the entire period of fallout probably was a rare occurrence at the NTS, in which case concentrations of radionuclides in air could be underestimated, depending on when exposure occurred; and the committee again notes that concentrations of radionuclides in fallout that descended on ships in the Pacific may be underestimated. Furthermore, in some dose reconstructions, it is evident to the committee that upper bounds of inhalation doses to atomic veterans have been underestimated by large

OCR for page 124
factors. The most obvious cases involve exposure scenarios for participants in forward areas at the NTS, including maneuver troops and close-in observers, in which resuspension of substantial amounts of previously deposited fallout by the blast wave produced in a detonation has been ignored even though exposure to relatively high concentrations of resuspended radionuclides caused by the blast wave almost certainly occurred. For example, when the NTPR program has assumed that resuspension of previously deposited fallout was caused by walking or other light activity in cases in which blast-wave effects probably occurred but were ignored, the committee believes that upper bounds of inhalation doses are underestimated by a factor of at least 100, and perhaps by a factor of as much as 1,000 in the worst cases. Furthermore, in such cases, upper bounds of equivalent doses to some organs and tissues could have been substantially above 1 rem. Of paramount importance is the issue of whether deficiencies in methods of estimating inhalation dose identified by the committee could have affected decisions about compensation of atomic veterans. The committee believes that possible underestimation of upper bounds of inhalation doses by the NTPR program is unlikely to be important for most participants in the Pacific or occupation forces in Japan. Inhalation doses to most of those participants probably were too low for possible underestimation of upper bounds to have affected decisions about compensation. The committee also believes that neglect of possible ingestion doses in dose reconstructions is unlikely to be important for most participants at any site. However, the neglect of blast-wave effects, combined with the frequent neglect of aged fallout that accumulated during the period of atomic testing at the NTS and neglect of fractionation in fallout, is an important concern for thousands of participants who were exposed in forward areas at the NTS shortly after a detonation. On the basis of an example analysis of the effects of a blast wave on inhalation doses (see Appendix E) and screening doses that have been used in evaluating claims for compensation (see Section III.E), use of credible upper bounds of inhalation doses in scenarios involving resuspension by a blast wave could have changed decisions not to grant compensation in some cases, depending on the disease of concern (for example, lung cancer in a nonsmoker). The question of the importance of deficiencies in methods of estimating inhalation doses in the NTPR program with respect to evaluating claims for compensation for radiation-related diseases is discussed further in Sections VI.F and VII.C. V.D DOSE RECONSTRUCTION FOR OCCUPATION FORCES IN JAPAN As discussed in Section IV.D, the upper-bound external dose for the 195,000 troops who participated in the occupation of Japan or were prisoners of war at or near Hiroshima or Nagasaki was estimated, on the basis of very pessimistic assumptions, to be always less than 1 rem, even though the likely dose to most

OCR for page 124
participants was at least a factor of 10 lower (McRaney and McGahan, 1980). The dose from ingestion of contaminated food or water or inhalation of resuspended debris was also found to be insignificant. The highest possible dose is for a participant who was present throughout the entire operation and spent 8 h d−1 at the location of highest exposure rates. However, most troops were rotated, troops were billeted well away from contaminated areas, and the highest exposure rates occurred over an area of only about 0.1 km2. In examining a sample of 12 cases, the committee found that detailed calculations of worst-case upper-bound doses were carried out for most of the veterans, and the calculations included both internal and external doses. In those cases, the calculated upper bound was considerably less than the overall generic upper-bound value of 1 rem. The one exception was a person with a calculated upper-bound dose of 0.62 rem. At the other extreme, three veterans were given an upper-bound dose of zero because they did not have an opportunity to be close to contaminated sites (for example, they remained on board a ship in the Nagasaki harbor). In one case, the veteran was in a different part of Japan. The committee concurs with the assessment by the NTPR program that the dose to even the most exposed of the occupation troops in Japan from both internal and external exposure was probably well below 1 rem. V.E COMMITTEE EVALUATION OF METHOD OF ESTIMATING UNCERTAINTY IN DOSE AND UPPER BOUNDS As stated in Section II.A, dose reconstruction is an inexact science. Uncertainties in quantifying dose arise from uncertainties in the various components that must be brought together to calculate a dose: in reconstruction of the activity scenario, in characterization of the radiation environment through time and space, in parameters assumed for calculations (such as resuspension factors and decay factors for radiation fields), in characterization of the mixture of radionuclides produced by a particular detonation, and in quantifying exposures through various routes (such as inhalation, ingestion, and dermal exposure). Clearly, uncertainties in the dose assigned to an atomic veteran are highly relevant to the adjudication process, particularly for diseases not categorized as “presumptive,” that is, diseases whose probability of causation is evaluated, because those uncertainties can inform the decision regarding the merits of a claim for service-connected disability. According to 32 CFR 218.3, which describes the approach to dose reconstruction used in the NTPR program: “Due to the range of activities, times, geometries, shielding, and weapon characteristics, as well as the normal spread in the available data pertaining to the radiation environment, an uncertainty analysis is performed. This analysis quantifies the uncertainties due to time/space variations, group size, and available data. Due to the large amounts of data, an automated (computer-assisted) procedure is often used to facilitate the

OCR for page 124
data-handling and the dose integration and to investigate the sensitivity to variations in the parameters used.” However, the committee did not see evidence in the case files that this kind of thorough uncertainty analysis was often done, although Monte Carlo methods can bring together sources of uncertainty in this way. The standard operating procedures (SOPs) document provided to the committee (DTRA, 1997) provides almost no information about how uncertainty is quantified by the NTPR program, and this complicated the committee’s review of methods used. The unit dose reports do provide uncertainty estimates, but they are usually estimates of the uncertainty in the average unit dose and, as discussed earlier, they may not provide a credible estimate of the uncertainty in the dose to the most exposed individuals in the unit. Furthermore, they often provide little detail regarding the specific method used, the exact correlations assumed or neglected, and the specific data used to calculate the upper bounds. Often, the reports acknowledge that the procedures used to combine various sources of uncertainty are based on approximate methods. The NTPR program’s intention with an upper-bound calculation is to provide at least a 95th percentile of the dose, that is, a dose that is intended to ensure that we can be at least 95% confident that the true dose is lower. Upper bounds estimated from film-badge data and from reconstructed gamma and neutron doses are combined in quadrature, assuming that they are uncorrelated, to arrive at an estimate of the upper bound in the total external dose. To the extent that the individual upper-bound estimates are credible and all doses and potential uncertainties are included, the upper-bound estimate for this sum is credible, provided that uncertainties in the increments of dose are independent—that is, not correlated—which they may not be because of repetitiveness of behavior and work responsibilities. If the components being summed are positively correlated, then the quadrature method will systematically underestimate the upper bound for the aggregated dose. Another problem arises in the context of combining uncertainties across different types of radiation. In recent years, after it became routine to report an upper bound for the external gamma plus neutron dose to VA, the sum of the estimated upper bounds of the gamma and neutron doses and the estimated “high-sided” internal organ dose has been used as the dose of record in evaluating probability of causation of a veteran’s claimed disease in the adjudication process. Summing upper bounds of external and internal doses would generally result in an overestimate of the upper bound of the total organ dose. However, as discussed earlier in this chapter, the committee found that in many cases the estimated upper bounds for external gamma and neutron dose were not credible and the “high-sided” estimates of internal and beta skin doses may not always reflect the 95th percentile dose (that is, a credible 95th percentile could be considerably higher). To the extent that the external gamma-plus-neutron dose upper bounds and inhalation dose estimates are reasonable estimates of at least 95th percentile or higher doses, the VA practice of summing the reported upper-bound external

OCR for page 124
dose and the “high-sided” inhalation dose will result in a high-sided estimate of the 95th percentile upper bound of the total organ dose.33 Although external and inhalation dose estimates are sometimes correlated to some extent, such as when both are based on the same exposure-rate measurement, most of the pertinent uncertainties involved are independent of each other. The estimated beta skin dose calculated by the NTPR program is directly related to the reported upper bound in gamma external dose. Thus, summing the reported beta dose estimate with the reported upper-bound gamma dose estimate will result in a credible estimate of the upper bound of the skin dose when the beta and gamma dose estimates both are credible upper bounds. The committee acknowledges that calculation of an upper-bound dose is itself an uncertain process. Furthermore, it is not clear how one ought to quantify effects of uncertainties in an activity scenario. For example, for external radiation exposure, NTPR program policy guidelines sometimes seem to target a best or even “high-sided” central estimate together with a 95th percentile upper-bound dose, and at other times seem to opt for only a “high-sided” estimate, in accordance with the benefit-of-the-doubt provision. For internal dose, the policy of the NTPR program is to provide a “high-sided” estimate that supposedly incorporates benefit of the doubt with respect to the exposure scenario. However, as discussed elsewhere in this chapter, the committee has concluded that assumed exposure scenarios often did not give the veteran the benefit of the doubt. V.F SUMMARY OF COMMITTEE FINDINGS REGARDING DOSE AND UNCERTAINTY ESTIMATES BY NTPR PROGRAM The central (“best”) estimates of external gamma and neutron doses to participants obtained by the NTPR program based on film-badge data and/or unit dose reconstructions are generally credible, provided that the assumed exposure scenario is reasonable. However, the committee has documented numerous examples in which the NTPR program has failed to establish the participant’s exposure scenario adequately; that is, plausible scenarios could be developed, on the basis of available information, that would have resulted in higher estimates of dose. The committee finds that estimates of uncertainty in external dose obtained by the NTPR program in unit dose reconstructions often are not credible and do not adequately reflect the upper bound (95th percentile) in the external dose to an individual participant, because deviations in individual exposure scenarios from the assumed group exposure scenario are not considered. Furthermore, the committee has identified a number of situations in which uncertainty in film-badge issuance dates, interpretation of data, and failure to give the veteran the benefit of 33   The equivalent dose to any specific organ from external gamma irradiation differs little from the reported whole-body dose because of the high penetrating power of the energetic photons emitted in detonations and by radionuclides in fallout and in activation products.

OCR for page 124
the doubt suggest the possibility of a much higher credible upper bound of the dose to an individual than reported by the NTPR program, even when the dose is based primarily on film-badge data. Upper-bound estimates of external dose should include consideration of the possibility of incorrect exposure scenarios, possibly missing or erroneous film-badge data, the impact of limited survey data, and other such factors. To give the veteran the required benefit of the doubt, some method should be devised to increase upper-bound estimates of external dose when there is reason to believe that any of those events may have occurred. The committee has concluded that, contrary to claims by the NTPR program, calculated internal doses from inhalation are not always “high-sided.” The committee has identified scenarios for which the method used by the NTPR program to estimate inhalation dose probably provides credible upper bounds (95th percentiles of possible doses or above). However, the committee has also identified important scenarios for which estimates of inhalation dose obtained by the NTPR program probably underestimate upper bounds by as much as a factor of 100 or more. Furthermore, organ equivalent doses could be substantial in some of those cases. The committee found that beta doses to the skin and lens of the eye, although claimed by the NTPR program to be “high-sided,” may not represent a credible estimate of the 95th percentile beta dose. Furthermore, beta doses from direct contamination of skin or clothing apparently have not been considered in dose reconstructions in any cases in which a veteran filed a claim for skin cancer. The committee believes that upper bounds of neutron doses reported by the NTPR program are not credible, because of neglect of the uncertainty in the biological effectiveness of neutrons. When neutron doses were important, estimated upper bounds of the combined gamma-plus-neutron doses obtained by the NTPR program may be low by as much as a factor of 5. The committee thus has concluded that the external gamma and neutron dose upper bounds and “high-sided” internal and beta skin and eye doses reported by the NTPR program often do not represent a credible estimate of the 95th percentile upper bound of the possible dose to an individual participant. As discussed in Section III.E, VA uses the sum of the reported external-dose upper bound and organ internal dose to evaluate the probability of causation of a claimed radiation-related disease. By using the upper-bound dose estimate to evaluate probability of causation, rather than the best (central) estimate, VA intends to give the veteran the benefit of the doubt. However, to the extent that the reported doses do not provide credible estimates of 95th percentile upper bounds of organ total equivalent doses, evaluations of probability of causation may be less favorable to the veteran than intended. Implications of the committee’s findings with regard to evaluating claims for compensation are discussed further in Section VI.F.