David J. Pawel

Stable Chromosome Aberrations in Atomic Bomb Survivors: Results from 25 Years of Investigation

Abstract Kodama, Y., Pawel, D., Nakamura, N., Preston, D., Honda, T., Itoh, M., Nakano, M., Ohtaki, K., Funamoto, S. and Awa, A. A. Stable Chromosome Aberrations in Atomic Bomb Survivors: Results from 25 Years of Investigation. Radiat. Res. 156, 337–346 (2001). Frequencies of stable chromosome aberrations from more than 3,000 atomic bomb survivors were used to examine the nature of the radiation dose response. The end point was the proportion of cells with at least one translocation or inversion detected in Giemsa-stained cultures of approximately 100 lymphocytes per person. The statistical methods allow for both imprecision of individual dose estimates and extra-binomial variation. A highly significant and nonlinear dose response was seen. The shape of the dose response was concave upward for doses below 1.5 Sv but exhibited some leveling off at higher doses. This curvature was similar for the two cities, with a crossover dose (i.e. the ratio of the linear coefficient to the quadratic coefficient) of 1.7 Sv (95% CI 0.9, 4). The low-dose slopes for the two cities differed significantly: 6.6% per Sv (95% CI 5.5, 8.4) in Hiroshima and 3.7% (95% CI 2.6, 4.9) in Nagasaki. This difference was reduced considerably, but not eliminated, when the comparison was limited to people who were exposed in houses or tenements. Nagasaki survivors exposed in factories, as well as people in either city who were outside with little or no shielding, had a lower dose response than those exposed in houses. This suggests that doses for Nagasaki factory worker survivors may be overestimated by the DS86, apparently by about 60%. Even though factory workers constitute about 20% of Nagasaki survivors with dose estimates in the range of 0.5 to 2 Sv, calculations indicate that the dosimetry problems for these people have little impact on cancer risk estimates for Nagasaki.

Improved Estimates of Cancer Site-Specific Risks for A-Bomb Survivors

Abstract Pawel, D. J., Preston, D. L., Pierce, D. A. and Cologne, J. B. Improved Estimates of Cancer Site-Specific Risks for A-Bomb Survivors. Radiat. Res. 169, 87–98 (2008). Simple methods are investigated for improving summary site-specific radiogenic risk estimates. Estimates in this report are derived from cancer incidence data from the Life Span Study (LSS) cohort of A-bomb survivors that are followed up by the Radiation Effects Research Foundation (RERF). Estimates from the LSS of excess relative risk (ERR) for solid cancer sites have typically been derived separately for each site. Even though the data for this are extensive, the statistical imprecision in site-specific (organ-specific) risk estimates is substantial, and it is clear that a large portion of the site-specific variation in estimates is due to this imprecision. Empirical Bayes (EB) estimates offer a reasonable approach for moderating this variation. The simple version of EB estimates that we applied to the LSS data are weighted averages of a pooled overall estimate of ERR and separately derived site-specific estimates, with weights determined by the data. Results indicate that the EB estimates are most useful for sites such as esophageal or bladder cancer, for which the separately derived ERR estimates are less precise than for other sites.

Uncertainty in estimating probability of causation in a cross-sectional study: joint effects of radiation and hepatitis-C virus on chronic liver disease

Exposure to other risk factors is an important consideration in assessing the role played by radiation in producing disease. A cross-sectional study of atomic-bomb survivors suggested an interaction between whole-body radiation exposure and chronic hepatitis-C viral (HCV) infection in the etiology of chronic liver disease (chronic hepatitis and cirrhosis), but did not allow determination of the joint-effect mechanism. Different estimates of probability of causation (POC) conditional on HCV status resulted from additive and multiplicative models. We therefore estimated the risk for radiation conditional on HCV status using a more general, mixture model that does not require choosing between additivity or multiplicativity, or deciding whether there is interaction, in the face of the large uncertainty. The results support the conclusion that POC increases with radiation dose in persons without HCV infection, but are inconclusive regarding individuals with HCV infection, the lower confidence bound on estimated POC for radiation with HCV infection being zero over the entire dose range. Although the mixture model may not reflect the true joint-effect mechanism, it avoids restrictive model assumptions that cannot be validated using the available data yet have a profound influence on estimated POC. These considerations apply more generally, given that the additive and multiplicative models are often used in POC related work. We therefore consider that an empirical approach may be preferable to assuming a specific mechanistic model for estimating POC in epidemiological studies where the joint-effect mechanism is in doubt.

Statistical issues in biological radiation dosimetry for risk assessment using stable chromosome aberrations

Biological dosimeters are useful for epidemiologic risk assessment in populations exposed to catastrophic nuclear events and as a means of validating physical dosimetry in radiation workers. Application requires knowledge of the magnitude of uncertainty in the biological dose estimates and an understanding of potential statistical pitfalls arising from their use. This paper describes the statistical aspects of biological dosimetry in general and presents a detailed analysis in the specific case of dosimetry for risk assessment using stable chromosome aberration frequency. Biological dose estimates may be obtained from a dose-response curve, but negative estimates can result and adjustment must be made for regression bias due to imprecise estimation when the estimates are used in regression analyses. Posterior-mean estimates, derived as the mean of the distribution of true doses compatible with a given value of the biological endpoint, have several desirable properties: they are nonnegative, less sensitive to extreme skewness in the true dose distribution, and implicitly adjusted to avoid regression bias. The methods necessitate approximating the true-dose distribution in the population in which biological dosimetry is being applied, which calls for careful consideration of this distribution through other information. An important question addressed here is to what extent the methods are robust to misspecification of this distribution, because in many applications of biological dosimetry it cannot be characterized well. The findings suggest that dosimetry based solely on stable chromosome aberration frequency may be useful for population-based risk assessment.

U.S. Environmental Protection Agency radiogenic risk projections: uncertainty analysis.

AbstractThe U.S. Environmental Protection Agency (EPA) has updated its estimates of cancer risks due to low doses of ionizing radiation for the U.S. population, as well as their scientific basis. For the most part, these estimates were calculated using models recommended in the recent National Academy of Sciences’ (BEIR VII) report on health effects from low levels of ionizing radiation. The new risk assessment includes uncertainty bounds associated with the projections for gender and cancer site-specific lifetime attributable risks. For most cancer sites, these uncertainty bounds were calculated using probability distributions for BEIR VII model parameter values, derived from a novel Bayesian analysis of cancer incidence data from the atomic bomb survivor lifespan study (LSS) cohort and subjective distributions for other relevant sources of uncertainty. This approach allowed for quantification of uncertainties associated with: 1) the effect of sampling variability on inferences drawn from the LSS cohort about the linear dose response and its dependence on temporal factors such as age-at-exposure, 2) differences in the radiogenic risks in the Japanese LSS cohort versus the U.S. population, 3) dosimetry errors, and 4) several other non-sampling sources. Some of the uncertainty associated with how risk depends on dose and dose rate was also quantified. For uniform whole-body exposures of low-dose gamma radiation to the entire population, EPA’s cancer incidence risk coefficients and corresponding 90% uncertainty intervals (Gy−1) are 9.55 × 10−2 (4.3 × 10−2 to 1.8 × 10−1) for males and 1.35 × 10−1 (6.5 × 10−2 to 2.5 × 10−1) for females, where the numbers in parentheses represent an estimated 90% uncertainty interval. For many individual cancer sites, risk coefficients differ from corresponding uncertainty bounds by factors of about three to five, although uncertainties are larger for cancers of the stomach, prostate, liver, and uterus. Uncertainty intervals for many, but not all, cancer sites are similar to those given in BEIR VII, which were derived using a non-Bayesian approach.

Uncertainties in Cancer Risk Coefficients for Environmental Exposure to Radionuclides. An Uncertainty Analysis for Risk Coefficients Reported in Federal Guidance Report No. 13

Federal Guidance Report No. 13 (FGR 13) provides risk coefficients for estimation of the risk of cancer due to low-level exposure to each of more than 800 radionuclides. Uncertainties in risk coefficients were quantified in FGR 13 for 33 cases (exposure to each of 11 radionuclides by each of three exposure pathways) on the basis of sensitivity analyses in which various combinations of plausible biokinetic, dosimetric, and radiation risk models were used to generate alternative risk coefficients. The present report updates the uncertainty analysis in FGR 13 for the cases of inhalation and ingestion of radionuclides and expands the analysis to all radionuclides addressed in that report. The analysis indicates that most risk coefficients for inhalation or ingestion of radionuclides are determined within a factor of 5 or less by current information. That is, application of alternate plausible biokinetic and dosimetric models and radiation risk models (based on the linear, no-threshold hypothesis with an adjustment for the dose and dose rate effectiveness factor) is unlikely to change these coefficients by more than a factor of 5. In this analysis the assessed uncertainty in the radiation risk model was found to be the main determinant of the uncertainty category for most risk coefficients, but conclusions concerning the relative contributions of risk and dose models to the total uncertainty in a risk coefficient may depend strongly on the method of assessing uncertainties in the risk model.

Adjustments to the Baseline Lung Cancer Mortality for Radon-Induced Lung Cancers in the BEIR VI Risk Models

where r0 is the baseline lung cancer mortality rate at age a in the population, b is the excess relative risk coefficient, which may be dependent on age and other parameters, z, such as smoking status, and w* is an effective exposure [e.g., in the BEIR VI Committee’s preferred models, w* is a sum of exposures weighted by a function of time since exposure and by a function of either exposure rate or exposure duration, and b is the product of a constant times age-specific and smoking-category correction coefficients (1)]. Suppressing the dependence on age and other variables for simplicity, the ‘‘excess risk’’ attributable to radon is then