VI Findings Related to Other Issues

VI.A QUALITY ASSURANCE AND DOCUMENTATION

The NTPR program provided written and oral information to the committee describing its quality assurance (QA) practices, including standard operating procedures (SOPs) and documentation requirements. That information, which is summarized in Section IV.G, consisted of a document on SOPs (DTRA, 1997), letters in response to the committee’s inquiries (see Appendix D), and oral discussions at open meetings of the committee. The following discussion summarizes the committee’s findings related to QA and documentation of procedures and individual dose reconstructions.

According to DTRA (Schaeffer, 2001a), the SOPs (DTRA, 1997) and 32 CFR Part 218 serve as the only written guidelines and procedures for the conduct of dose reconstructions. However, the SOPs really constitute a statement of approach and general principles followed by the NTPR program rather than a manual documenting the procedures used to reconstruct doses. Furthermore, the committee found that the SOPs are incomplete, are out of date, and contain no references to supplement the limited text. Many of the methods used to estimate doses are not discussed at all, nor are the methods that are used to estimate upper bounds. Details of the reconstruction methods are neither discussed nor referenced in the SOPs.1

1  

Some procedures are reported in unit dose reports, specialized reports (for example, Banks et al., 1959; Barss, 2000; Egbert et al., 1985; Barrett et al., 1986; Goetz et al., 1991), or unpublished SAIC internal memoranda (for example, Ortlieb, 1991; 1995; Flor, 1992; Schaeffer, 1995).



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 231
VI Findings Related to Other Issues VI.A QUALITY ASSURANCE AND DOCUMENTATION The NTPR program provided written and oral information to the committee describing its quality assurance (QA) practices, including standard operating procedures (SOPs) and documentation requirements. That information, which is summarized in Section IV.G, consisted of a document on SOPs (DTRA, 1997), letters in response to the committee’s inquiries (see Appendix D), and oral discussions at open meetings of the committee. The following discussion summarizes the committee’s findings related to QA and documentation of procedures and individual dose reconstructions. According to DTRA (Schaeffer, 2001a), the SOPs (DTRA, 1997) and 32 CFR Part 218 serve as the only written guidelines and procedures for the conduct of dose reconstructions. However, the SOPs really constitute a statement of approach and general principles followed by the NTPR program rather than a manual documenting the procedures used to reconstruct doses. Furthermore, the committee found that the SOPs are incomplete, are out of date, and contain no references to supplement the limited text. Many of the methods used to estimate doses are not discussed at all, nor are the methods that are used to estimate upper bounds. Details of the reconstruction methods are neither discussed nor referenced in the SOPs.1 1   Some procedures are reported in unit dose reports, specialized reports (for example, Banks et al., 1959; Barss, 2000; Egbert et al., 1985; Barrett et al., 1986; Goetz et al., 1991), or unpublished SAIC internal memoranda (for example, Ortlieb, 1991; 1995; Flor, 1992; Schaeffer, 1995).

OCR for page 231
The SOPs provided to the committee (DTRA, 1997) contain a provision for periodic review and updating. The version provided to the committee had not been modified in several years, even though significant changes had occurred in the program. An example of an important change that should have triggered an update of the 1997 SOPs is the routine assessment of beta dose to skin that began in 1998. Documentation of procedures for determining beta dose to skin is important because the number of claims filed for skin cancer under the nonpresumptive regulation has increased dramatically over the last 4 years. The committee was unable to locate formal documentation detailing when particular procedures were implemented or revised,2 although those changes might have been documented in internal memoranda not reviewed by the committee. Examples include implementation of large-particle inhalation dose coefficients, revision of the upper-bound gamma dose for target-ship boarding parties at Operation CROSSROADS, and the upper-bound analysis for sums of film-badge readings. QA is not discussed in any detail in the SOPs, nor are many important procedures, such as details on how film-badge uncertainties are calculated. Because of the evolution of the program and the lack of formal documentation of changes in policy available either publicly or in the individual files, it is difficult for a reviewer to evaluate a given dose reconstruction to ensure that up-to-date approved and consistent procedures were used. Some of the case files contained no narrative discussions of the dose assessments. Dose reconstruction memoranda prepared by the NTPR program supposedly contain references to all methods used. However, references in dose memoranda are often internal NTPR program memoranda.3 A veteran reviewing such a dose report would have no ready access to the referenced documents. Dose reconstructions by different analysts for similar scenarios often referenced different internal memoranda or reports for the same method. 2   For example, in a letter to DTRA, the committee requested documentation regarding the use of a lower upper-bound estimate of external dose for participants who boarded target ships during Operation CROSSROADS than given in the unit dose report (Weitz et al., 1982). The reply (Schaeffer, 2002b) stated that the upper bound had been revised, but no specific documentation as to when and why was supplied. Similarly, the committee requested specific information as to how the upper bound for sums of film-badge readings was calculated. In reply, the committee was told that the NTPR program followed recommendations of the NRC (1989) and supplied an unannotated copy of the computer program used to calculate upper bounds. Comments in the program listing clearly indicated that the method recommended by the NRC had been slightly modified. However, no additional documentation justifying the modifications was supplied. 3   The committee found a number of instances in which the references cited in a dose memorandum were not the appropriate or correct source of the information used to calculate the dose. Often, the cited reference contained only a reference to another document that discussed the procedure.

OCR for page 231
VI.A.1 Quality Assurance As noted in Section IV.G.2, the statement of work in the DTRA solicitation for NTPR program support required that the contractor provide QA monitoring for the program in database management, dose assessment, and veteran assistance. In response to the solicitation, JAYCOR and SAIC submitted a technical proposal that specified QA measures for the three program tasks. The committee did not see the JAYCOR-SAIC proposal or any detailed QA procedures, but it did receive a letter from the NTPR program that described the QA program (Schaeffer, 2002e); this description is included in Appendix D and summarized in Section IV.G.2. The committee noted that its sample of 99 case files contained little evidence of uniform application of basic QA measures. Dose calculations often were not signed, dated, or initialed by the analyst. Many of the typed assessments included typed initials of analysts and dates, but several did not. In files containing several recalculations of dose, the lack of dates made it difficult to determine which was the most recent. Although dose assessments are supposed to be reviewed before release to VA or to the veteran (DTRA, 1997), the files generally contained no documentation to show that the reviews occurred or who performed them. Many files did, however, contain logs in which activities associated with the files (but not final reviews) were noted and dated. Dose assessments were transmitted to VA or the veteran by a letter signed by the DTRA program manager, thus indicat-ing final managerial approval. The committee also found several examples of poor quality control that resulted in errors in the calculation or reporting of external dose. In case #2, the reported dose failed to account for a film-badge exposure during an earlier test series. The participant also was assumed to be present during Operation GREENHOUSE for a shorter time than indicated by his service record. In case #84, the dose memorandum references an incorrect unit dose report. In case #87, the dose memorandum assigns a dose of 0.4 rem but references an SAIC memorandum that indicates that the dose was 0.8 rem. In case #88, the dose memorandum and a letter in the file from the NTPR program to the veteran give the dose as 1.0 rem, but the SAIC database lists the total external dose as 1.8 rem, with no upper bound; according to the referenced exposure scenario, if the veteran is given the benefit of the doubt and is assumed to have been in an armored personnel carrier rather than a tank during a maneuver, his dose would have been 1.7 rem with an upper bound of 2.7 rem. Published reports of the NTPR program did not indicate that they had been subjected to peer review. Reports are often published without documentation of their review process, and the committee did not see any written description of a process by which the NTPR program reviews its documents. As is the case with some of the committee’s other concerns, peer reviews may have occurred informally or even formally, but they could not be verified by the committee, because

OCR for page 231
neither the reviews nor any indications that they had occurred were recorded in documents available to the committee. Some reports contained erroneous statements, which suggested to the committee that effective peer review had not occurred.4 VI.A.2 Documentation of Dose Calculations The SOPs (DTRA, 1997) list the documentation to be supplied to a veteran in connection with a dose assessment. On the basis of the committee’s reviews of the sample of 99 case files, the documentation was generally supplied. However, detailed reviews of the calculations used for dose reconstructions5 were central to the committee’s work. Those reviews were hampered by the uneven and generally poor documentation found in the individual files. Many calculations were recorded by hand and were illegible. Variables often were not defined, and committee members had difficulty in understanding the calculations. One file (case #77) even had important information recorded on an unsigned adhesive note that appeared on the side of a page in the record. Most of the records for the 99 cases in the committee’s sample were considered to be inadequately documented, but we identified 28 case files in which the calculations were particularly poorly documented (cases #1, 3, 4, 6, 8, 9, 18, 19, 21, 26, 37, 38, 40, 43, 47, 49, 53, 57, 59, 60, 65, 68, 73, 77, 78, 82, 85, and 98). Comments on a few selected examples follow. Case #3: Dose calculations were handwritten and scribbled although one part of the work was dated and signed. No upper-bound calculation was evident, although this assessment was performed in 1997. Case #6: The handwritten dose calculations were unclear, hard to read, and difficult to interpret. The dose summary sheet had a typed date and initials, but the dose calculations themselves were not dated or initialed. Case #18: The handwritten dose calculations had insufficient detail. Although the typed dose assessment was dated and included the analyst’s name, the dose calculations themselves were undated and the analyst unidentified. 4   In Barss (2000), for example, the linear energy transfer (LET) of photons was compared with the LET of beta particles, but LET is a quantity that applies only to charged particles, such as betas, and not to photons. Another statement in the report claimed that “although only a small fraction of the associated mixed fission product external gamma dose is actually deposited in the basal cell layer (of the skin), it is assumed that the entire whole body external gamma dose was deposited in the basal cell layer for upper bounding.” That statement illustrates a misunderstanding or misuse of the quantity “dose.” The committee is of the opinion that those inexact statements would have been identified in a peer review. 5   Examples include calculations for combining film-badge data, applying unit dose reconstructions, assessing internal dose, and assigning uncertainty or upper bounds.

OCR for page 231
Case #53: The file contained a hard-to-read handwritten excerpt, regarding doses on the USS Skate, that was attributed to “the back-up file.” The hard-to-read handwritten calculations of daily target-ship doses and upper-bound dose calculations were not dated, and the analyst was not identified. Case #57: The calculations in the file were mostly handwritten, with some use of mathematical software. No source was given for a handwritten table concerning generic doses and their “error factors” for ship-related activities. The handwritten calculations were difficult to read, and what could be read was difficult to interpret. Case #77: The file had an important inference about the impossibility of the veteran’s being involved in a dose-producing activity (Chemical Biological Radiological Team training) attached as an adhesive note (see discussion of this case in Section V.A.2). Case #78: A dose of 150 mrem was assigned to account for the dose received during island “sweeps” in connection with Operation GREENHOUSE. There were neither details on how these “sweep” doses had been calculated nor a citation. Internal dose calculations were handwritten and difficult to follow. The calculations were not dated or initialed by the analyst. VI.A.3 Summary In summary, the committee did not see a comprehensive manual of SOPs that is complete and current and covers all methods. The committee saw no manuals that document when changes in procedures occurred, that contain copies of computer programs, that document a formal quality assurance and quality control (QA-QC) program, or that contain copies of all internal NTPR program procedures memoranda referenced in dose reports. Although the NTPR program (Schaeffer, 2002e) cited the Environmental Protection Agency’s QA guidelines concerning the importance of SOPs,6 and the committee agrees with this approach, we did not see sufficient evidence of its implementation in the case files and other documents reviewed. The lack of a comprehensive document explaining the dose reconstruction methods was also an important concern to the National Research Council committee that reviewed the program in 1985 (NRC, 1985b). Assessment documents in the case files often are undated and show no authorship. Some assessments are difficult to follow, even for scientists, because internal documentation is inadequate. In fairness, others are good, but the dispar 6   These guidelines (EPA, 2001) state that: “The development and use of SOPs are an integral part of a successful quality system as it provides individuals with the information to perform the job properly, and facilitates consistency in the quality and integrity of a product or end result.”

OCR for page 231
ity makes it clear that there are not uniformly applied standards of quality. The committee was told that all assessments are reviewed, but there is not even simple documentation of reviews in the files (for example, date of review and signature of the reviewing authority), much less memoranda or other written documentation of the substance of the reviews. A comprehensive program explaining QA-QC objectives and procedures should be developed and documented. The program should ensure that all doses are checked and correctly calculated, that consistent and up-to-date-methods are used, that all dose estimates are fully and adequately documented, and that backup records and calculations in files are complete, legible, annotated, and dated. All dose assessments should be documented clearly, and the veteran’s file that contains details of the calculations, backup material, and so on, should be complete, legible, and comprehensible. Any references not readily available, such as internal memoranda, should be included in the file. All entries in the file should be typed or clearly written and dated, and the author should be clearly identified. Methods of validating film-badge doses, such as routine comparisons with generic dose reconstructions or with data on other veterans who performed the same type of activity, should be part of such a program. Some QA-QC should be performed by outside experts, possibly under the direction of an advisory board. VI.B COMMUNICATION WITH ATOMIC VETERANS As discussed in Section III.B, communication with an atomic veteran concerning compensation decisions is the primary responsibility of the VA Regional Office (VARO), which receives dose reports and other correspondence from DTRA. But the response time has often been lengthy, particularly before the recent effort to respond in a more timely manner. In case #32, for example, the NTPR program wrote to the VARO on October 20, 1995, apologizing at the outset for a delay in follow-up of an inquiry first filed on February 6, 1995. It appears to have taken that long for the veteran’s participation in an atmospheric test to have been confirmed and a radiation dose assessment completed. In our sample of cases, it was not uncommon for at least 6 months to pass between the VARO inquiry and the NTPR report on the results of a dose reconstruction. When a veteran feels aggrieved, he is owed accurate and responsive communication by the government. In any effort of this magnitude and complexity, there will be errors in communication, delays, failures to respond appropriately, and questionable judgments about suitable content in a response. But there were enough lapses within the committee’s sample cases to cause concern. In some cases, intervention by members of Congress who inquired on behalf of constituent veterans or family members seemed to speed the process or to revive an investigation years after it appeared to have concluded. In a few other instances, an attorney was retained to pursue the matter on behalf of a veteran. On the whole, veterans and their families have needed patience and determination to resolve their concerns.

OCR for page 231
Over the years, as represented in our sample cases, written communications with veterans or their families were courteous, and doses were reported accurately. When much time had elapsed between letters, which was often the case, suitable apologies were expressed. When veterans described serious health problems, regret and sympathy usually were offered. Nonetheless, veterans often had reason to find the process frustrating. In one instance (case #68), the VARO received a letter from DNA reporting a dose to the skin, but the claim was for cancer of the kidney. The Veterans Benefits Administration identified the error about 6 weeks later. The same file exemplifies a problem in many of the cases in which the veteran was deceased. As part of information gathering, a letter from the VARO to the widow asks 20 questions of a highly detailed nature that the widow is unlikely to know how to answer, such as: “Is there personal history of smoking? If yes, provide the starting date, what was smoked (cigarettes, pipes, cigars, etc.), how much (number of packs, cigars, or pipes, etc., daily), and the date stopped (if applicable).” “Give the organization or unit attached to (ship, tank group company or squadron) and the rank at the time of the test? If different for more than one test, provide this information for each test.” “Provide a detailed description of duties and other activities during the entire period of each test.” “How far away was ground zero at the time of each test shot?”; and “Was there any direct contact with contaminated materials? If yes, describe the specific circumstances.” Although the widow is informed that she may write “I do not know” for any question, it is hard to understand the point of this exercise. The respondent is asked to provide such specific information that the reliability of her responses leaves wide scope for doubt. The committee recognizes that there are legal requirements for service-connected benefits, but the failure to modify the questions out of consideration for the widow’s circumstances seems excessively formalistic. Obtaining accurate detailed information from deceased veterans’ families is sometimes hampered by veterans’ unwillingness to discuss their nuclear-test-related activities when they were alive. In at least one instance among those sampled by the committee (case #32), the veteran’s family reported his reluctance to violate his sense of patriotic duty by discussing the atomic-testing program because he had been told, for reasons of national security, not to discuss his role with anyone. That sometimes made a veteran reluctant to apply for additional benefits when he became ill. As a result, only an incomplete account of his service activities was available to his surviving family. When one veteran was deposed by VARO counsel concerning a claim, he refused on security grounds to disclose all details of a rocket test shot (case #36).

OCR for page 231
Attitudes about the process and the candor of VA and DTRA varied. Although expressions of patriotism characterized many of the veterans’ approaches to the dose reconstruction process, there were also those in our sample whose mistrust was demonstrated in remarks they made about the circumstances of their exposure. It is striking, for example, that several veterans alleged that they were used as a “human guinea pig” or a “human monitor” or that they were in an “experiment.” It is not clear that anyone in authority could dissuade veterans of such views, but delays and bureaucracy surely do not lessen such suspicions. More recent direct communication between veterans and DTRA evinces an improved effort to tailor responses to the veteran’s circumstances. Yet in at least one exchange (case #60), the veteran asked several specific questions concerning bioassay (see Section VI.D), sampling criteria in the comparable U.S. population that did not participate in atomic testing, and the name and address of the agency conducting the sampling, but the response did not address the second and third questions. Over the years, various attempts have been made to provide veterans with general information on what was being learned about their radiation exposures. For example, a letter from the Navy to a veteran (case #49) includes the following statement, which is found in a number of such letters in the early to middle 1980s: “An important finding to date is that radiation exposures to the participants were generally quite low.” The letter also refers to “the consensus of the medical community” that “the risk of any adverse health effect from exposures such as experienced by nearly all test participants is very, very slight.” It is not clear what “consensus” the letter intended to identify. About a year later, a letter from the NTPR program to another veteran (case #58) reports the conclusions of a 1985 National Research Council report concerning mortality of test participants (NRC, 1985a) and includes a press release of the report. The committee found such generic statements regarding the low incidence of cancer among veterans and low probability of radiation as a causal agent for cancer possibly confusing to laypersons. Although attempts to provide general information and reassurance on the significance of radiation doses and risks are laudable, the committee finds the way in which scientific views have been used in these communications to be somewhat troubling. Appearing in the context of an individual request for record review, dose reconstruction, and potential compensation for a disease suffered by a veteran, the communications have a defensive quality that is at best unpersuasive for the veteran’s case and at worse an overt attempt to “expert” the veteran into retreat. Veterans and their families may, and apparently often do, regard the source of this information, which is the authority responsible for analyzing radiation exposure, as operating under a conflict of interest. Letters sent to veterans often were ambiguous in their presentation of the risk and the extent of scientific agreement. A letter sent in 1996 (case #15) stated that

OCR for page 231
“the doses received by over ninety-nine percent of the test participants were less than the current Federal guidelines for radiation workers which permit external exposures of 5 rem gamma per year.” The next paragraph begins by saying that “While medical science has no proof that exposure to low levels of ionizing radiation is hazardous to health, it is generally assumed by scientists that even low levels of exposure carry some slight risk.” Quite apart from questions about the accuracy of the latter assertion, the two statements taken together may be confusing to a layperson. This language appeared in letters from the NTPR program to atomic veterans for years. An earlier version found in the file for case #63 sought more aggressively to downplay the significance of radiation exposure. It included the statement that: “In sum, the studies described above have not revealed a basis for concern by participants of the atmospheric nuclear tests, or by the veterans of the Hiroshima and Nagasaki occupation forces over an increased risk of adverse health effects due to radiation.” The magnitude of the beta dose to skin and the associated risks were downplayed for years after it became clear that there was a stochastic risk of skin cancer at doses experienced by some participants (see Section V.B.6). In some cases, veterans seem to have received a more tailored response when they communicated directly with the NTPR program and bypassed the VARO. Sometimes, however, the tailored responses unfolded over a long period. Case #9 exemplifies the bureaucratic odyssey that sometimes characterizes multiple dose assessments and the tenacity required by the veteran. After a 1978 application for service-connected disability benefits, a 1980 letter from a Navy NTPR official to the veteran offered a medical examination and reported a “careful search of dosimetry data” at the test shot, which was said to have revealed an exposure within then-prevalent occupational standards. After what seems to have been a series of further contacts, a 1992 letter included then-common language that “radiation exposures to the participants were generally quite low” and that “the consensus of the medical and scientific community is that the risk of any adverse health effect from exposures such as experienced by nearly all test participants is very, very slight.” DTRA completed the latest dose report on this veteran on June 20, 2000. The long period of several dose assessments often resulted in letters concerning revised dose reconstructions that could as easily have undermined a veteran’s confidence in the process as reinforced it. One instance is a letter in 1995 (case #56) that reports a revision in an estimate provided 10 years earlier. On the whole, communication with veterans, although courteous and accurate with respect to reporting doses, has sometimes not been timely, has varied in detail, has conveyed mixed messages, and, considering the time and effort involved, has failed to inspire confidence that the process was fair, orderly, and expeditious.

OCR for page 231
VI.C THE LOW-LEVEL INTERNAL DOSE SCREEN VI.C.1 Introduction As mentioned in Section I.D, one of the key issues of concern to the atomic veterans has been the application of a low-level internal dose screen to eliminate the need for a more detailed estimation of internal dose in many dose reconstructions. In an attempt to provide some insight into this issue, the committee undertook a thorough evaluation of the low-level internal dose screen and provides the following discussion about its use in dose reconstructions. The methods applied in the NTPR program to estimate internal doses to atomic veterans are discussed in Section IV.C, and the committee’s evaluation of the methods is presented in Section V.C. As emphasized previously, intakes of radionuclides by inhalation are expected to be more important in most cases than intakes by ingestion or absorption through the skin or open wounds. Early in the NTPR program, it was recognized that the task of performing detailed calculations of inhalation dose to all participants in the atomic-testing program could be overwhelming, given that data needed to estimate intakes of radionuclides, including measured concentrations of radionuclides in air at times of exposure or amounts of radionuclides excreted in urine or feces, generally were lacking. It also was believed that inhalation doses usually would be insignificant compared with external doses. Inhalation doses would be insignificant, for example, when most of the airborne radionuclides to which a participant was exposed were shorter-lived, photon-emitting fission or activation products. Those types of radionuclides produce relatively high external doses per unit activity concentration in the environment because of the long distances of travel of higher-energy photons in air, but relatively low internal doses because of their short residence times in the body and the low absorption in body tissues of higher-energy photons emitted by radionuclides in the body. On the basis of those considerations, a method was developed that could be used to quickly evaluate potential inhalation doses to groups of participants who were exposed under similar conditions and to eliminate from further consideration groups whose members most likely did not receive a significant inhalation dose. Such a process is referred to as screening. A method of evaluating the potential importance of inhalation exposures, referred to as the low-level internal dose screen, was first developed and applied to participant groups at the NTS (Barrett et al., 1986). The same screening method was later applied to participant groups in the Pacific (Goetz et al., 1991). This section discusses the general requirements of a screening method and the low-level internal dose screen developed in the NTPR program. Particular attention is paid to concerns that the atomic veterans have expressed about use of the internal dose screen in dose reconstructions, especially when a veteran files a claim for compensation for a radiation-related disease.

OCR for page 231
VI.C.2 General Requirements of Screening Methods Any method of screening that is used to draw conclusions about the significance of potential radiation doses, or lack thereof, incorporates two basic elements: A model to estimate dose in an assumed exposure scenario. A dose criterion to define a level of exposure below which there is no concern. If the dose estimated with the model is below the dose criterion, potential doses in the assumed scenario are considered to be so low that the associated health risks are insignificant. Given the basic elements listed above, a method of screening must satisfy two conditions to meet its intended purpose: The model used to estimate dose must tend to overestimate doses for any exposure conditions that could be encountered. The dose criterion must correspond to a dose that clearly is insignificant with regard to potential health risks to an exposed person. It also is helpful if the model used to estimate dose is simple and transparent, so that others can understand that it is likely to overestimate dose. VI.C.3 Assumptions Used in Internal Dose Screen The methods used to estimate inhalation doses to participant groups for purposes of screening are the same as those described in Section IV.C.2. That is, the same four basic exposure scenarios and the same methods of estimating inhalation dose in each scenario were used in screening. For example, when a participant group was assumed to be exposed to resuspended fallout that had been deposited on the ground, the inhalation dose to an organ or tissue of concern is estimated on the basis of (1) estimates of the relative activities of radionuclides produced in shots of concern combined with an assumption of no fractionation of radionuclides in fallout except for the removal of noble gases, which are used to estimate the relative activities of radionuclides in fallout deposited on the ground; (2) measurements of external photon exposure or exposure rate due to the deposited fallout with film badges or field instruments combined with calculations of the exposure rate per unit activity concentration for the assumed mixture of radionuclides, which are used to estimate the absolute activity concentrations of radionuclides on the ground; and (3) an assumed resuspension factor, which is applied to the estimated concentrations of radionuclides on the ground to obtain an estimate of the concentrations in air. The dose criterion used in screening of potential inhalation exposures of participant groups is a 50-year committed equivalent dose of 0.15 rem to bone,

OCR for page 231
been used to estimate inhalation doses when a veteran filed a claim for compensation, contrary to the assertions by DTRA noted above.10 On the basis of the committee’s efforts to understand the low-level internal dose screen and its use in dose reconstructions, the committee appreciates the confusion that can result when the NTPR program states that the internal dose screen is not used in dose reconstructions for veterans who file claims for compensation but there appears to be evidence to the contrary in the documentation of dose reconstructions or in letters sent to veterans that summarize the results. However, any such confusion can be resolved if one recognizes that the methods of estimating inhalation dose that were used to develop the screening codes for participant groups at various tests are the same as the methods used to estimate inhalation doses in individual dose reconstructions (see Section VI.C.3). In judging the adequacy of an assessment of inhalation dose in the dose reconstruction for an individual veteran, the most relevant questions concern whether the assumed exposure scenario for that person is reasonable, according to knowledge of his activities and the radiation environment in which those activities took place, and whether the models, parameter values, and other assumptions used to estimate inhalation dose in the assumed scenario provide credible upper-bound estimates. If credible upper-bound estimates of inhalation dose have been obtained for a veteran, it is of secondary concern whether the documentation of the dose reconstruction or the letter sent to the veteran summarizing the results includes statements or reported doses that indicate that the low-level internal dose screen may have been used to estimate inhalation dose or may have influenced the calculation. If the screening code assigned to the veteran’s unit in a low-level internal dose screen report was based on reasonable assumptions, the inhalation dose based on the screening code and the assigned external dose for the unit should provide a credible upper bound. However, if the assumptions used in developing the screening code for the veteran’s unit are incorrect and do not describe the conditions of exposure of the veteran, the inhalation dose based on the screening code and the assigned external dose will not provide a credible upper bound.11 10   The NTPR program also has acknowledged that there were instances when a veteran filed a claim for compensation but an individual dose reconstruction was not performed (Schaeffer, 2002d). That situation occurred when the exposure scenario for the veteran was judged to be adequately described by assumptions used in an existing dose reconstruction for the veteran’s unit in which the low-level internal dose screen had been applied. In such cases, JAYCOR apparently reported doses estimated in the unit dose reconstruction but an individual dose reconstruction was not performed by SAIC. Thus, in a sense, the low-level internal dose screen was used to estimate inhalation doses to a few veterans who filed claims for compensation. 11   A clear example discussed in Section V.C.3.2, comment [6] and [7], and Appendix E involves inhalation exposures of participants in forward areas after Operation PLUMBBOB Shot HOOD. The screening code assigned to several units at Shot HOOD was based on an assumption that the only radionuclides inhaled were neutron-activation products in soil (Barrett et al., 1986), even though substantial amounts of plutonium and longer-lived fission products undoubtedly were present.

OCR for page 231
It is important to bear in mind that the credibility of a dose reconstruction does not depend on how the analysis is described. Rather, the credibility of a dose reconstruction depends only on the assumptions used in calculating dose, whether the dose was calculated correctly on the basis of the assumptions, and whether the calculated dose was reported correctly. The description of a dose reconstruction may indicate that there are problems in communicating to a veteran how the analysis was performed, and poor communication certainly can affect a veteran’s belief about the credibility of an analysis. However, the issue of communication to the veteran is separate from the issue of the credibility of the dose reconstruction itself. VI.D BIOASSAY PROGRAM TO ASSESS INTERNAL EXPOSURES TO PLUTONIUM VI.D.1 Description of Plutonium Bioassay Program In the late 1990s, the NTPR program undertook an effort to determine whether plutonium bioassay testing of atomic veterans, specifically measurement of the amounts of plutonium in urine samples, could be used to estimate intakes that occurred during participation in the weapons-testing program (Schaeffer, 2002f). If intakes of plutonium could be estimated on the basis of bioassay testing, the estimated intakes could be used to investigate the reliability of the methods of estimating inhalation doses to participants discussed in Section IV.C.2, which are based on data other than measured concentrations of radionuclides in air or amounts of radionuclides excreted in urine or feces shortly after exposures occurred. Bioassay testing for the purpose of assessing the reliability of methods of internal dose estimation was recommended in a previous review by a committee of the National Research Council (NRC, 1985b) discussed in Section V.C.2. Plutonium was selected as the radionuclide to be studied because it is an important component of fallout, it presents a significant inhalation hazard, and it is tenaciously retained in the body after an intake (a substantial fraction of plutonium absorbed into blood is retained for the rest of life, even if intakes occurred at an early age), and because there appeared to be suitable methods of analysis that could detect very low concentrations of plutonium in urine. It is the committee’s understanding that data on the amounts of plutonium in urine samples provided by atomic veterans that have been collected so far have not been used in assessing internal dose in any dose reconstructions. The committee found no evidence to the contrary in its review of dose reconstructions for individual veterans. VI.D.2 Discussion of Plutonium Bioassay Program On the basis of information on results obtained in the plutonium bioassay testing program (Schaeffer, 2002f), the committee has concluded that it will be

OCR for page 231
difficult to obtain reliable estimates of internal doses to participants during periods of atomic testing. The measured activities of plutonium in urine are sufficiently low in most cases that background levels of plutonium in urine due to global fallout can obscure the signal derived from participation in atmospheric tests. There also are problems with the method of bioassay itself at these low levels, and analyses of split samples revealed poor agreement in pairs of measurements. Furthermore, even if activities of plutonium in urine could be measured reliably, there is substantial uncertainty in applying assumed biokinetic models for the behavior of plutonium in the body and rates of excretion (which may vary among individuals) to obtain estimates of intakes that occurred many years ago. That uncertainty would need to be evaluated and taken into account in assessing the extent to which recent bioassay data could be used to assess the reliability of methods that have been used by the NTPR program to estimate intakes of plutonium by atomic veterans. VI.E RETROACTIVE RECALCULATIONS OF DOSES AND RE-EVALUATIONS OF PRIOR COMPENSATION DECISIONS An important challenge throughout the more than 20-year existence of the NTPR program has been the need to adapt to changes in applicable laws and regulations and to incorporate improvements in the scientific foundations of dose reconstruction and methods of estimating radiation risks and probability of causation of radiation-related diseases. For example, one issue is how to incorporate improvements in science into the methods of dose reconstruction and the decision process used to evaluate claims for compensation while striving to treat all claims equitably. The committee was concerned about what is done within the NTPR program to re-evaluate dose reconstructions when a claim for compensation was denied but it is thought that later changes in laws, regulations, or methods of reconstructing doses or estimating probability of causation might have affected the outcome if they had been in place at the time of the claim. For example, suppose that a veteran filed a claim for compensation for kidney cancer before this form of cancer was declared to be presumptive in 38 CFR 3.309 (see Section I.B.4), but the claim was denied because of a low estimate of dose in the veteran’s dose reconstruction (see case #99). The question in this case is: Would the veteran’s claim have been reopened after kidney cancer was declared to be a presumptive disease and would compensation have been granted from the time the presumptive law was passed, assuming that the veteran’s participation in the atomic-testing program is adequately established? Or suppose that a veteran filed a claim for skin cancer before 1998, when such claims were routinely denied without assessment of possible skin doses on the basis of the presumption that doses of around 1,000 rem were required to cause skin cancer and participant doses of this

OCR for page 231
magnitude were not credible (see Section V.B.6.3). Would the veteran’s claim have been reopened and an assessment of dose to the skin performed for use in reevaluating the previous compensation decision after the method for estimating beta dose to the skin was developed and claims for compensation for skin cancer were granted in some cases when the estimated equivalent dose to skin was as low as a few tens of rem (see case #9)? In response to a verbal inquiry on the issue of retroactive recalculations of dose and re-evaluations of prior compensation decisions in cases in which claims had been denied, the committee was informed that VA generally does not take the initiative to reopen cases when a change in law, regulations, or methods of reconstructing doses or estimating probability of causation of a radiation-related disease could have affected the compensation decision. Nor does VA or the NTPR program inform individual claimants about changes that could have affected their denied claims. Rather, the NTPR program disseminates such information, for example, in public announcements or statements submitted to veterans organizations, such as the National Association of Atomic Veterans, for publication in a newsletter. Once the information is disseminated, the responsibility lies with the veterans or their representatives to request that a prior claim be reopened and re-evaluated. The committee has identified several specific issues that could be important with regard to retroactive recalculations of doses and re-evaluations of prior compensation decisions. The issues are summarized as follows (all examples apply to claims for compensation for nonpresumptive diseases under 38 CFR 3.311): Re-evaluation of claims that were denied before the disease of concern was declared to be presumptive in 38 CFR 3.309. Re-evaluation of claims for skin cancer that were denied before development of the method of estimating beta doses to skin and before the large reduction in the skin dose that could be judged to be at least as likely as not to cause skin cancer. Re-evaluation of claims that were denied when the wrong disease was identified in the dose reconstruction (that is, the dose to the wrong organ or tissue was estimated).12 Retroactive calculation of upper bounds of external dose in claims that were denied before such upper bounds were calculated routinely (that is, in cases in which only a central estimate of external dose was obtained in the dose reconstruction). 12   The committee encountered one case in its review of 99 randomly selected dose reconstructions for individual veterans in which the wrong disease was assumed (case #91); this error presumably could be corrected only if information sent to the veteran that summarizes the dose reconstruction and compensation decision specifically identifies the disease for which the claim was denied.

OCR for page 231
Retroactive calculation of internal doses in claims that were denied before development of the FIIDOS code (Egbert et al., 1985) in cases in which internal dose was not estimated in the dose reconstruction (see case #89 and 99). Retroactive recalculation of inhalation doses in claims that were denied before an assumption of large particles (AMAD, 20 μm) was used routinely in exposure scenarios in which mostly large particles presumably were inhaled and an assumption of large particles, instead of respirable particles (AMAD, 1 μm), would increase the estimated dose to the organ or tissue of concern. Re-evaluation of cases in which a veteran did not file a claim for compensation but stated that he had a radiogenic disease specified in 38 CFR 3.311.13 Retroactive re-evaluation of decisions to deny claims for compensation that were made before so-called screening doses (CIRRPC, 1988) discussed in Section III.E were used routinely in evaluating claims, especially when the screening dose to the organ or tissue of concern is lower than the dose criterion that was used previously in evaluating claims. The issue of retroactive recalculations of dose and re-evaluations of decisions to deny compensation also will arise if methods of dose reconstruction are revised in response to the committee’s evaluations presented in this report, because appropriate changes should result in increases in credible upper-bound estimates of dose in many cases. It is not the committee’s intent to criticize policies of the NTPR program and VA concerned with informing veterans whose claims for compensation were denied about changes in laws, regulations, or methods of reconstructing doses or estimating probability of causation that could affect their claims. The committee recognizes the effort that would be required to take the initiative to re-evaluate every denied claim whenever there is a change in some aspect of the program that could affect claims. Nonetheless, veterans might view the NTPR program more favorably if, for example, individual veterans were informed when changes in methods of estimating doses are made that might result in increases in their previously assigned doses or when policies affecting evaluations of claims are changed and were reminded that they can request a revised dose reconstruction. For example, after recent changes in the methods of calculating beta dose to the skin and evaluating claims for compensation for skin cancer, individual veterans with a previously denied claim for skin cancer, and the community of atomic veterans as a whole, could have been informed that doses to the skin are now being calculated in a different way and, furthermore, that more claims for skin cancer are being granted on the basis of a re-evaluation (lowering) of doses that could cause skin cancer. A 13   The committee encountered one such case in its review of dose reconstructions for individual veterans (case #55).

OCR for page 231
properly designed database would facilitate efforts to identify veterans with previously denied claims for particular diseases and to locate the veterans and inform them of changes in the dose reconstruction program and in the laws and regulations that could affect their claims. VI.F IMPLICATIONS OF COMMITTEE’S FINDINGS VI.F.1 The Central Issue The committee’s review of the program of dose reconstruction for atomic veterans was prompted by concerns about the adequacy of the methods used to estimate dose. Those concerns are important because estimated doses to atomic veterans are used in evaluating claims for compensation for nonpresumptive diseases. Although previous scientific reviews expressed concerns about some aspects of the methods of dose reconstruction (see Section V.C.2) and the atomic veterans themselves have questioned many assumptions used in dose reconstructions (see Section I.D), the central issue of concern to the veterans has been the apparently very small number of claims for compensation for nonpresumptive diseases that have been granted. When a legal and regulatory structure is established to provide compensation to atomic veterans who later experience radiation-related diseases on the basis, in part, of results of dose reconstructions but the odds of receiving compensation appear to be very low, it is understandable that the veterans would question their estimated doses and the value of dose reconstruction. In accordance with the requirement to give the veterans the benefit of the doubt in evaluating claims for compensation for nonpresumptive diseases (see Section I.C.3.2), dose reconstructions should provide credible upper bounds of possible doses to participants, taking into account uncertainties in estimating dose that are an inherent part of any dose reconstruction. Indeed, it is the policy of the NTPR program to provide “high-sided” estimates of dose. However, as discussed at length in Chapter V, the committee has concluded that dose reconstructions for atomic veterans do not consistently provide credible upper bounds of possible doses. Rather, in many cases, credible upper bounds would be substantially higher than alleged “high-sided” dose estimates in existing dose reconstructions. The reasons for substantial underestimation of upper bounds of possible doses in many cases are related primarily to two recurring issues: assumptions about exposure scenarios that do not give the veterans the benefit of the doubt or do not conform to plausible conditions of exposure, taking into account information on the veterans’ activities and reasonable assumptions about the radiation environment in which the activities took place; and inadequate accounting of uncertainty in data, models, and parameter values used to estimate external and internal doses.

OCR for page 231
Given the committee’s findings on deficiencies in methods and assumptions used in dose reconstructions and the belief, based on these findings, that upper bounds of possible doses to atomic veterans are often underestimated substantially, the central question is: What are the implications of the committee’s findings with respect to claims for compensation for nonpresumptive diseases? That is, if credible upper bounds of dose had been obtained in all dose reconstructions, what would have been the effect on the number of claims granted for nonpresumptive diseases? VI.F.2 Number of Claims Granted for Nonpresumptive Diseases Before addressing the question posed above, we should consider the issue of the number of claims for compensation for nonpresumptive diseases that have been granted. As of October 2001, about 14,000 claims had been filed for presumptive (38 CFR 3.309) and nonpresumptive (38 CFR 3.311) diseases, including about 4,400 claims from veterans who served in occupation forces in Japan (VA, 2001). On the basis of information provided by VA (1996), the atomic veterans believed that only about 50 claims had been awarded for nonpresumptive diseases. If the estimate of about 50 claims granted for nonpresumptive diseases is correct, the odds of a successful claim indeed appear to be very low (about 1% or less).14 However, in response to a written inquiry, the committee was informed that about 1,600 claims for compensation for nonpresumptive diseases had been granted, including about 350 claims by members of occupation forces in Japan (Flohr, 2001). If that estimate is correct, the odds of a successful claim would be much higher (perhaps 15-20%). In an effort to resolve this important issue, which is central to the veterans’ concerns about the value of dose reconstruction in their compensation program, the committee examined the outcomes of about 300 claims for compensation, some of which involved presumptive diseases, and a list of the outcomes of recent medical opinions by VA on claims for nonpresumptive diseases. The committee’s investigation indicated that the proportion of successful claims for nonpresumptive diseases has been around 1% or less, excluding awards for skin cancer since 1998.15 Therefore, statements that the number of successful claims for nonpresumptive diseases other than skin cancers since the regulations in 38 CFR 3.311 were promulgated is on the order of 50, as previously reported and as believed by the 14   The rate of success of claims for nonpresumptive diseases cannot be estimated more precisely because VA does not maintain a database that gives the breakdown of successful and unsuccessful claims for presumptive and nonpresumptive diseases separately. 15   Few, if any, claims for skin cancer were granted before 1998, but the number of claims granted since 1998 is such that the current rate of granting claims for all nonpresumptive diseases may be nearly 10%.

OCR for page 231
veterans, indeed appear to be reasonable. A report by GAO (2000) also indicated that the number of claims granted for nonpresumptive diseases is low. VI.F.3 Implications of Findings for Past Compensation Decisions As discussed in detail in Chapter V, the committee has concluded that there are many deficiencies in the methods of dose reconstruction being used in the NTPR program. However, the implications of the deficiencies with regard to evaluating claims for compensation for nonpresumptive diseases are difficult to assess. In many cases, it is unlikely that possible underestimation of upper bounds of doses to atomic veterans would have affected a decision about compensation, even though the degree of underestimation of upper bounds could be substantial. As an example, consider the potential importance of deficiencies in the methods of estimating inhalation doses that were identified by the committee (see Section V.C). The committee’s belief that underestimation of upper bounds of inhalation doses probably would not have affected a decision on compensation in many cases is based on two considerations. First, the lowest doses required to grant a claim for compensation can be high, depending on the cancer of concern and the participant’s age at the time of exposure and time of diagnosis (see Section III.E). For example, if a veteran was a smoker and filed a claim for lung cancer, doses to the lung of greater than 25 rem normally were required on the basis of screening doses that have been used since the late 1980s (CIRRPC, 1988); doses of about 10 rem or higher were required for several other cancer types, including colon cancer (lung and colon cancers are common in veterans’ claims). All screening doses were developed on the basis of an assumption that the risk of a specific cancer is a linear function of dose, without threshold. Second, difficulties in estimating inhalation doses notwithstanding, inhalation doses received by many participants almost certainly were low, judging by known conditions of exposure, and often were much lower than external doses that were monitored with film badges or field instruments. Similarly, the committee believes that external doses to most participants were sufficiently low that a re-evaluation of external doses based on the committee’s findings about deficiencies in methods of estimating credible upper bounds probably would not result in estimates that exceed the screening doses used in evaluating claims. Thus, for most veterans who filed claims for nonpresumptive diseases other than skin cancer, the committee believes that there is little chance that a reassessment of external and inhalation doses would affect a past compensation decision. That conclusion applies, for example, to participants for whom a credible upper bound of external dose is less than 1 rem and who were exposed only to descending fallout, resuspended neutron-activation products in soil, or deposited fallout that was resuspended by walking or other activities that did not involve vigorous disturbance of surface soil. In such cases, upper bounds of inhalation doses

OCR for page 231
probably were less than 1 rem. Therefore, a credible upper bound of dose, taking into account uncertainty in estimates of external and internal doses, is unlikely to approach a screening dose that would qualify a veteran for compensation. However, the committee also believes that there almost certainly have been cases in which a veteran’s claim for compensation for a nonpresumptive disease was denied but would have been granted if a credible exposure scenario had been assumed and uncertainty in estimating dose had been taken into account properly. As an example, consider the committee’s concern about neglect of the effects of a blast wave on inhalation exposures in areas of accumulated fallout at the NTS (see Section V.C.3.2, comments [6] and [7], and Appendix E). When exposure to accumulated fallout occurred, potential inhalation doses relative to external doses were substantially higher than in cases of exposure to descending or freshly deposited fallout, mainly because of the increased importance of plutonium relative to photon-emitting radionuclides, and a blast wave produced high concentrations of airborne radionuclides, compared with other causes of resuspension, over a substantial area. That type of exposure scenario occurred at several shots in later test series at the NTS. Furthermore, there are cancers for which the lowest dose required to grant a claim for compensation is not high. For example, claims for compensation for lung cancer in nonsmokers have been granted when the estimated dose to the lung from internal and external exposure combined is as low as 4 rem (Otchin, 2002; see also Section III.E). The committee’s analysis of potential inhalation doses due to resuspension caused by a blast wave in areas at the NTS where fallout had accumulated throughout much of the period of atomic testing (see Appendix E) indicates that upper-bound estimates of lung doses of that magnitude are credible when uncertainties in data, models, and parameter values are taken into account. A similar conclusion may apply in other scenarios at the NTS in which inhalation exposures resulted from resuspension of relatively large amounts of deposited fallout by vigorous disturbances of surface soil. An example is exposure during assaults or marches behind armored vehicles (see case #88). In this scenario, the resuspension factor normally assumed in dose reconstructions (see Table IV.C.2) probably is a reasonable upper bound, but consideration of the accumulation of deposited fallout throughout the period of atomic testing, fractionation of radionuclides, and uncertainty in inhalation dose coefficients could result in increases of more than a factor of 10 in credible upper bounds of inhalation dose compared with estimates obtained in dose reconstructions. Depending on the organ or tissue in which a veteran’s cancer occurred, it is possible that credible upper bounds of the total dose from external and internal exposure could qualify the veteran for compensation in these cases. The committee has not attempted to estimate the number of past claims for compensation for nonpresumptive diseases that could be affected by its findings on deficiencies in the methods of dose reconstruction; such an effort is beyond the scope of the committee’s work. However, although the committee believes

OCR for page 231
that most past claims would not be affected by a reassessment of doses on the basis of the findings described in this report, it believes that some claims that were not granted would have been. VI.F.4 Implications of Findings for Future Compensation Decisions The implications of the committee’s findings on deficiencies in methods of dose reconstruction used in the NTPR program with regard to the number of claims for nonpresumptive diseases that might be granted in the future are similar to those described above. If methods of dose reconstruction were changed to be consistent with these findings, the committee expects that the outcome of most future claims would not be affected. That expectation is based on the presumption that the distribution of doses in future claims will be similar to the distribution in past claims and, therefore, that credible upper bounds of dose to most claimants would be too low for the VA to conclude that the veteran’s disease was as least as likely as not caused by his radiation exposure and thus qualify the veteran for compensation. Two additional factors that have not been important in the past may affect future claims for compensation. First, the list of presumptive diseases has been expanded to include 21 cancer types (see Section I.B.4). Of the cancer types for which radiation risks have been estimated from studies of the Japanese atomic-bomb survivors (Thompson et al., 1994), only cancers of the rectum, skin, uterus, prostate, and nervous system are now nonpresumptive; of these, skin and prostatic cancer appear frequently in veterans’ claims, but the others apparently are rare. Estimates of the risk per unit dose for these cancers tend to be lower than estimates of risks for many cancers that are presumptive diseases, and this results in higher screening doses that might qualify a veteran for compensation.16 Second, VA is updating its methods of evaluating the probability of causation of radiation-related cancers (see Section III.E). Consequently, the lowest dose that would qualify a veteran for compensation will increase for many cancers. Such increases are a direct result of improvements in the data on cancer risks in humans and attendant decreases in uncertainty in cancer risk estimates (that is, decreases in credible upper bounds of cancer risks per unit dose). Thus, if methods of dose reconstruction used by the NTPR program are changed to be consistent with the committee’s findings, the effect on future claims for compensation for nonpresumptive diseases is not likely to be substantial. However, the importance of the committee’s findings could increase when veterans file claims for compensation for presumptive diseases. In such cases, requirements for establishing a veteran’s status as a participant are more demand 16   Expansion of the list of presumptive diseases means, of course, that it is much easier for many veterans whose claims for compensation for nonpresumptive diseases were denied in the past to be compensated now under the presumptive law.

OCR for page 231
ing, and if adequate proof of participation is not available, claims are evaluated under the nonpresumptive regulation and a dose reconstruction is required. The committee does not know how often claims for compensation for presumptive diseases have had to be evaluated under the nonpresumptive regulation, but the number of such cases could increase in the future, given the expansion of the list of presumptive diseases, thus increasing the importance of changes in methods of dose reconstruction.