Reply to Fellman

摘要

Dear Editors: We appreciate feedback and criticism on our paper (Hoffman et al. 2011) from Alan Fellman. His arguments are based on the premise that at doses and dose rates well below limits of epidemiological detection of a statistically significant dose-response, quantification of risks with uncertainty is not defensible. In such cases, he believes that quantified risk estimates can do more harm than good.We disagree for several reasons. First, the inability of epidemiology to detect a statistically significant excess risk at low doses and low dose rates is not, by itself, evidence of an absence of risk. The presence of risk at low doses is inferred by (1) observations of an essentially linear dose response for incidence of radiogenic cancers above limits of epidemiological detection (at doses ranging from about 10 mSv to above 200 mSv, depending on the study and specific conditions that affect the power of the study to detect risk), (2) the difficulty in identify-ing biological mechanisms that conclusively explain the absence of risk at low doses (in the presence of ubiquitous exposures to background radiation), and (3) the inability to conclude that background radiation exposure is itself without risk.