Mortazavi and Doss present the crude rates of cancer incidence by dose for males and females in their Figure 1, and by inspection assert that there is a threshold dose. Crude rates and visual inspection are poorly suited for supporting statistical inference. We stated that the best estimates for threshold values were greater than 0 but were not distinguishable from zero via formal statistical tests. Our analyses were not limited to linear-non-threshold models. Indeed, one major new result in this report was the non-linearity of the male dose response.
Epidemiological data from the atomic bomb survivors have been meticulously recorded for decades and are a priceless resource for understanding the long-term human health effects of exposure to ionizing radiation. However, despite the size of the cohort and the number of survivors who received low doses, the power to detect radiation effects or characterize the shape of the dose-response at low...