A Review of Two Hanford Environmental Dose Reconstruction Project (HEDR) Dosimetry Reports: Columbia River Pathway and Atmospheric Pathway (1995)

These reports are substantially improved over the one for the feasibility study, which the committee reviewed earlier (National Research Council, 1994), and the authors are to be commended. The text is generally clear and better organized. The figures are well designed, but in order to convey the results of the dose reconstruction more effectively and convincingly, the graphs should be reviewed for consistency of symbols and made self-explanatory by showing units and distance scales. The reports could be improved, and to this end the committee offers the following recommendations.

• A fuller definition of source terms, model validation, and demographiccharacteristics ascribed to the three representative individual typeson whom doses are computed should be included. The details are presentedin other reports, which are cited, but a brief discussion of thesecrucial topics here would make the reports more self-contained andcomprehensible and make the results more credible to the averagereader.

• A detailed discussion of model validation for the various pathwaysshould be included to account for possible discrepancies betweencalculated and measured values. The validity of the models used iscentral to their acceptability. Without such discussion, the modelcomputations appear to be divorced from reality and of dubious credibility.It is essential, therefore, that the reports explain thoroughly howthe calculations were validated to an acceptable degree of accuracyand which segments of the models represent justifiable projections;this explanation should include the ranges of the doses receivedby representative, well-described individuals, rather than by thosemaximally exposed.

• Worker body-burden data should be considered as a means for validatingmodels.

• The maximally exposed individual and how this individual is relatedto different pathways of exposure should be described more clearly.

• The scoping analysis used to evaluate the relative importance ofthe various pathways at different periods and used to select themost important impacts for more detailed assessment should be discussed.A tabular presentation of the lesser pathways and their screeninglevels would also be helpful.

• Information should be collected on the estimates of doses for peoplewho are in the Hanford Thyroid Disease Study or who request suchdose estimates. This information could be useful in evaluating thedose-reconstruction effort and its ability to address the concernsof the public.

• The committee recommends that a summary table describing models foreach pathway be added to the report. Such a table might include 10-20of the most important parameters included in the models, and thedefault values used when values for the parameters are not available.In addition, a clear presentation is needed of the range of parametervalues used to produce 100 realizations of estimated doses for anygiven individual.

• More details on the specific sources of uncertainty considered inthe uncertainty analyses should be presented including the size andshape of the assumed distribution of each source. It would also behelpful, in this regard, to indicate the range of possible choicesfor the ingestion dose conversion and transmission factors and wherein these ranges the values used in the analyses lie.

• The use of only five radionuclides in the assessment of the doseobtained through the water pathway should be justified earlier inthe text than it is now. Because the contribution of the water pathwaysto the total dose is small, except possibly in the case of the indigenousAmerican Indian populations, this restriction of the assessment tothe five major contributing radionuclides appears warranted, butthe rationale for the restriction should be presented earlier.

• Dietary histories based on memory have many short-comings, and historiesobtained 40 or more years after the critical events could be subjectto systematic biases that would compromise epidemiologic findings.The degree of reliability of the anamnestic reporting of food intakeshould be evaluated. This evaluation will be pivotal in the estimationof sample sizes in all later epidemiologic studies and in decidinghow much credence to place in them.

• A special effort should be made to present the estimates of absorbeddose in the thyroid in a manner more convincing of their accuracy(reliability). The acceptability of the proposed thyroid-diseasestudy hinges on the reliability of those doses, particularly in theyoungest and most vulnerable portion of the population.

• Some rationale should be presented for the reduction in the numberof locations used in the RATCHET model (2,091) and the DESCARTESmodel (1,102).

• Consideration should be given to the changing patterns in breastfeedinghabits and the use of dried milk over the years of this assessment.Milk is the major source of iodine-131 exposure of nursing infants,and such changes could lead to overestimation or underestimationof the thyroid dose.