February 19, 1992

Radiation Dosage Estimation and Health Risk-Reply

Author Affiliations

University of North Carolina at Chapel Hill School of Public Health
Oak Ridge (Tenn) Associated Universities

JAMA. 1992;267(7):929-930. doi:10.1001/jama.1992.03480070045013

In Reply.  —Drs Maienschein and Peelle's conclusion that "best estimates" of the radiation-cancer dose response from our study of workers at the Oak Ridge National Laboratory1 should be reduced by nearly an order of magnitude due to radiation dose underestimation is based on faulty logic. It is true that underestimation of doses of more highly exposed workers would spuriously elevate the estimated dose response. However, because of their higher exposures, those are the workers whose dosimeters were least likely to have been exposed below the detection threshold. Exposure underestimation for workers in the lowest-dose categories would actually mix the risk experience of minimally exposed workers with that of more highly exposed workers, resulting in underestimation of the radiation-mortality dose-response relationship.Maienschein and Peelle misrepresent the annual dose trends shown in the Figure in our article.1 There is no evidence of a discontinuity of the trend in 1956 attributable