Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 Exposure Definition and Measurement The National Research Council's 1985 study (Robinette et al., 1985) of participants in the five nuclear weapons test series on which the current report focuses used dose data provided by the Defense Nuclear Agency's (DNA) Nu- clear Test Personnel Review Program (NTPR). Using data painstakingly col- lected from diverse sources, NTPR staff and contractors attempted to assign to each individual participant a valid estimate of the radiation dose received (Gladeck and Johnson, 1996~. Initial plans for the new Five Series Study in- cluded the use of these individual dose assignments. The committee charged with oversight of the present study created a working group, with external expertise, to review DNA dosimetry estimation procedures and results. Based on the working group's findings, the full committee issued a letter report (IOM, 1995; reprinted as Appendix A in this report), stating that the dosimetry estimates were not appropriate for dose-response analyses in the con- text of epidemiologic studies. In this chapter, we describe (1) the background and limits of the NTPR dosimetry work as it relates to this study's protocol, (2) alter- native exposure surrogates considered, (3) decisions made for the analyses in this report, and (4) possibilities for farther investigations. * John Till, Ph.D.' served as chair of the dosimetry working group' which included one other oversight committee member (Clarice Weinberg, Ph.D.) and three external dosimetry experts (F. Owen Hoffman, Ph.D.; Keith J. Schiager' Ph.D.; and John Taschner). 36
EXPOSURE DEFINITION AND MEASUREMENT DNA-PROVIDED DOSE ESTIMATES Individual Doses 37 The Nuclear Test Personnel Review (NTPR) database contains a dose as signment for each participant, derived in most cases through reconstruction based on duty assignments. In less than half of cases, the assigned dose is based on one or more film badges worn by the participant or on a film badge worn by another participant in the same unit (cohort badging). Ideally, exposure measurements would be (1) individual-specif~c; (2) re- corded by time, duration, and dose; (3) sensitive to different components of ex- posure (e.g., alpha, beta, or gamma radiation); (4) previously validated for use in similar situations; (5) quantitative and at least theoretically reproducible; (6) complete, in that they cover all exposures for all involved people; and (7) ac- cepted by all interested parties. As stated above, based on our examination of the NTPR dosimetry data, we do not believe that these data are appropriate for the individual-specific assignments necessary for the type of epidemiologic com- parisons on which this report is based. A working group of the Five Series Study oversight committee assessed the basis and quality of the data upon which dose assignments were made and con- cluded that they were not suitable for dose-response analyses in epidemiology (IOM, 1995, p. 2~: The Working Group concluded that there has been a lack of consistency over time in NTPR dose estimation methods and, in particular, in the methods of as- signing "high-sided" doses, that is, doses in which uncertainties are resolved in favor of assigning higher doses rather than lower doses. In some cases, because of the existing compensation program, procedures for assigning doses have been different for those who did and did not file a claim for a radiogenic can- cer. Neither the dose assignment methods nor the database itself are thoroughly documented. In addition, uncertainties have not been estimated in a consistent manner and do not incorporate all potential sources of variability inherent in the dosimetry. The conclusions also state, "Although there is anecdotal evidence that indi- vidual doses may have been greatly underestimated in individual cases, the overall tendency may have been to overestimate both external and internal doses" (p. 13~. Individualized dose reconstructions are "generally only carried out if there is a specific institutional or legal need for a refined estimate" (p. ll~for example, when a veteran or a survivor files a claim for health or death benefits. Because reconstructed doses are more likely to be overestimated than others (IOM, 1995), these NTPR doses differ systematically based on health status, which is closely tied to our study endpoints. We think that using them would introduce sufficient bias to render the epidemiologic analysis of these data useless. Veterans, mean
38 THE FIVE SERIES STUDY while, have expressed concerns that the assigned doses are significantly lower than justified, based on their firsthand experiences at the test site. Alternative Uses of Dose Data We looked for other, indirect, quantitative dose measures obtainable from the NTPR database. We hypothesized that either the number of badges issued to an individual or the total dose derived from badge data might be a more reliable measure of individual exposure than the reconstructed total dose discussed above. Based on DNA background material, we hypothesized that the participants most likely to be exposed to ionizing radiation would have been issued more badges than those believed less likely to be exposed. Facing the same dosimetry question in our earlier study of Operation CROSSROADS participation (John- son et al., 1996), we had looked for relationships between the number of badges issued to an individual and both the total dose assigned to that individual and the dose assigned to the individual using badge data alone. Finding no relationship, we rejected using the number of badges as an exposure surrogate for the CROSSROADS study and, now, for this study as well. The assignment of individual-level dose surrogates based on badged dose was also considered and rejected. The issuance of personal dosimeters varied by service and series. An estimated 45 percent of all atmospheric nuclear weapons test participants (DTRA, 1999), and 52 percent of all participants in the five test series in particular (IOM, 1995), have individual radiation dose information on record (see Appendix A). The REDWING series had the largest number of par- ticipants who were issued a personal dosimeter (82 percent), and UPSHOT- KNOTHOLE had the lowest (13 percent). This reflects a change in procedures over time, not a difference in the anticipated exposures. The value of the do- simetry information for use in epidemiologic studies is questionable, however, even in instances in which relatively high proportions of personnel were badged (IOM, 1995~. Individuals who were badged did not wear their badges continu- ously during their exposure, according to participant and DTRA accounts. Thus, an individual's cumulative dose from film badges may well give an incomplete picture of total dose. We examined the CROSSROADS dosimetry data to de- termine if an individual's assigned dose and badged dose were proportional, which would allow an assumption that badges were indicative of the total dose accrued by the individual. We found that individuals with very similar badged totals had widely disparate assigned doses due to differing dose reconstructions. Other approaches were considered and rejected. For example, we might use the highest recorded doses and the zero doses, hoping that dosimetry was more reliable at the extremes. However, the committee was concerned that the NTPR database did not adequately distinguish between truly zero doses and unknown or unmeasured doses. Finally, we explored how one might use dosimetry as an indicator rather than a measure of radiation. Again, each approach to the development of an in
EXPOSURE DEFINITION AND MEASUREMENT 39 dex of probable exposure that could be supported by the available dosimetry presented difficulties. In summary, seeing no evidence that the film badge data provided an exposure surrogate adequate for use in this study, we chose not to use them. POTENTIAL SURROGATE MEASURES OF DOSE After ruling out the use of DTRA-assigned doses, we considered various ways in which to categorize exposure. Some suggestions such as number of series, number of individual shots, and type and size of detonation were re- jected immediately because this information did not correlate closely with what we know about dose. For example, an individual who was assigned to units at four different test series, but was assigned to indoor locations far from the deto- nations, might have received no radiation dose, whereas another individual who attended only one test shot may have had responsibilities within meters of ground zero soon after the detonation. The number of series in which an individual participated and the individ- ual's branch of service were rejected as dose surrogates for similar reasons. We did conduct some analyses in an attempt to determine whether certain series could be considered as proxies for exposure. The results of some of these analy- ses are presented in Appendix E. They provide descriptive information, but we believe that they are difficult to interpret because of the many unmeasured and potentially heterogeneous circumstances that they represent. Given the wealth of anecdotal and written record descriptions of potential high-dose situations (e.g., troops involved in maneuvers such as Desert Rock Troop Brigades; those involved in cloud sampling; and radiation safety person- nel), we discussed how we might use historical, qualitative information to define high-dose groups. Because these task groups often were not defined by a spe- cific unit name, however, we could not identify which individuals to assign to the study category. Historical narrative records suggest that particular ships were subject to higher exposures than others; for example, individuals assigned to the USS Bairoko, USS Philip, Rongerik, Rongelap, or the boat pool. We could not look at RADSAFE (radiation safety) personnel for reasons of feasibility and interpretability. RADSAFE is not always a unit designation; radiation safety personnel may be spread out among other units and their potential exposures diluted by the lower doses of others in these units. Also, professional radiation experts may have used more safeguards (e.g., protection equipment) and re- ceived lower doses in exposure situations because of their expertise. One distinction among the participants that may be exposure related is the location of the test series. The weapons tested in the Pacific and Nevada test areas were primarily two different types of devices. For the Pacific test areas, fusion weapons were tested primarily, and for the Nevada Test Site, fission de- vices. Although these types of weapons are fundamentally different in the inter- actions that led to detonation, the residual radiation fields created in the two
40 THE FIVE SERIES STUDY locations are very similar. Nevertheless, because of environmental differences between the test areas and the tasks that personnel were expected to carry out, it is possible that the pathways of exposure were somewhat different. For example, military personnel who were exposed to radiation from contaminated ships in the Pacific tests were likely not to have received as much exposure via the inha- lation pathway as their counterparts at the Nevada Test Site where resuspended particulate matter could have been inhaled. Another example of potential differ- ences between exposures is the type of activities that servicemen were under- taking. In the Nevada tests, most personnel were exposed as a result of ground contamination in areas where they either witnessed the explosions or entered following the blasts. In the Pacific, many personnel were exposed after they went aboard ships that had been placed at varying distances from ground zero. Although it would be expected that the predominant exposure at both test areas would have been direct gamma radiation originating from surfaces, differences likely existed in the radiation dose fields and thus could have exposed some or- gans of the body in different ways. Although these scenarios of exposure between the two test areas are specu- lation, it is evident that the two environments within which servicemen were working differed greatly and these differences could have led to exposure being created and received in different ways. These potential differences in exposure suggest that there may be justification for considering a comparison between disease among veterans who received the majority of their dose from one loca- tion or the other. The Five Series Study design purposefully included partici- pants from both land and sea test sites. DECISIONS FOR THE ANALYSES IN THIS REPORT Based on the considerations described in the preceding sections, the com- mittee and staff decided not to use dosimetry data in the analysis. This decision was not taken lightly. The painstaking effort to develop the dose data was im- mense. The dose data, however, as previously described, do contain systematic differences that could affect the study's results in ways that are not well defined. Therefore, without looking at dose-outcome correlations, we made the decision not to use the individual-specific reconstructed or badged doses. The core study, therefore, is a comparison of the participant and referent cohorts. Status as a participant is taken to be the most reliable (though broad) indicator of exposure. The participant versus referent cohort dichotomy provides the largest group of people to study, size being important when considering rare outcomes such as leukemia. Using participant status to represent exposure in this case, potential exposure to ionizing radiation and possibly to other test- related environmental factors also presents many limitations regarding epide- miologic study. We cannot account for differences in potential radiation expo- sure among participants or their exposure to other ill-defined environmental or occupational factors, either related to or independent of nuclear test participation
EXPOSURE DEFINITION AND MEASUREMENT 41 (e.g., later employment as a radiological technician or a radiation worker in the nuclear power industry). FUTURE OPTIONS FOR USE OF DOSIMETRY Should the current study yield interesting findings, the oversight committee has discussed further avenues of research. Demonstrating an association be- tween dose and outcome would greatly support any finding of higher mortality among participants than referents. Once a specific outcome is selected (which was not feasible in this study because of the requirement that all participants and a range of outcomes be considered), an efficient design such as the nested case- control study could be used. Such studies require fewer subjects, making less prohibitive, for example, the per-person expense of custom dose reconstruction. For reasons explained in detail in its earlier report (IOM, 1995), the oversight committee believes that useful dose reconstructions can be achieved if certain guidelines such as unbiased selection of participants and technical consistency in methodology for dose estimation are followed.
