above. The essential function of quality assurance and quality control is to ensure that there is a systematic and auditable documentation of the procedures used in dose reconstructions and that the methods of analysis and the calculations themselves are free of important error. Proper documentation must be provided for all data, interpretations of data and other assumptions, and computer codes or other methods of calculation used to estimate dose. Procedures to be used in dose reconstructions and changes in the procedures, including when they occurred, must be properly documented. If complex computer codes are used, they should be verified to ensure that they do not introduce important error. External peer review is an important means of achieving quality assurance.
In general, proper quality assurance and quality control are essential to developing confidence in the dose reconstruction process and the resulting estimates of dose. The issue of quality assurance and quality control is discussed in more detail in Section IV.G.
Three other aspects of dose reconstructions for atomic veterans warrant special consideration. First, dose reconstructions have been used to evaluate claims for compensation by specific persons who incurred a disease that could have been caused by exposure to ionizing radiation during the atomic-testing program; in many cases, estimates of dose to a particular organ or tissue in which a cancer or other disease has occurred are compared with a specified dose as part of the process of deciding whether the disease was at least as likely as not to have been caused by radiation exposure (see Section III.E). Second, the claimant is to be given the benefit of the doubt in estimating dose. Third, the NTPR program has been going on for more than two decades, and there have been many advances in the science of dose reconstruction over that time. As discussed below, these considerations have important implications for the dose reconstruction process for atomic veterans.
Focusing on reconstruction of doses received by specific persons can place considerable demands on the dose reconstruction process, especially with regard to defining exposure scenarios, selection of parameter values for use in models to estimate dose, and treatment of uncertainty. For example, depending on the particular exposure situation, it could be inappropriate to ascribe average exposure conditions in a participant group to a specific person in that group. In its review of dose reconstructions for individual veterans, the committee encountered a substantial number of cases that clearly involved unusual, or even unique, exposure conditions.