Low-dose Extrapolation of Radiation-Related Cancer Risk

Draft document: Low-dose Extrapolation of Radiation-Related Cancer Risk
Submitted by John G. Merkle, Volunteer, ATSDR Work Group, Oak Ridge, TN
Commenting as an individual

The calculations in Table 2.4 of ICRP Committee 1 Draft Task Group Report 12/421/04, dated 12/10/04, are of interest with respect to the evaluation of public health assessments for the Oak Ridge, TN, area being prepared by ATSDR. Using the book by Kahn and Sempos, "Statistical Methods in Epidemiology", Oxford University Press, 1989 (see p. 31) as a reference, I have attempted to reproduce the calculated population sizes in Table 2.4. The calculation for 1 Gy works, but to get the values of N for the subsequent three smaller doses, I have to switch the value of the standard normal deviate associated with alpha and the null hypothesis from 1.65 to 1.00. I would appreciate a reply indicating your thoughts on this apparent discrepancy. Is it real, and if so, am I making a mistake, or is there one in Table 2.4? Thanks for your consideration, John Merkle