RE: Comments on LNT as a "prudent upper limit"

[jrcamero@facstaff.wisc.edu (John Cameron)]

An excerpt from the recent contribution  by Frank R. Borger commenting on
Science Withous Sense states:
>        Only ONE gripe. The Linear-non-threshold model was really adopted
>        as a prudent UPPER LIMIT of possible harm, in an era when we didn't
>        have the epidemiological studies, lab data etc we have now. The people
>        who adopted it were not "some geniuses" but just trying to do "the
>        best they could with what they had" at the time.

I disagree with Frank's interpretation. I certainly agree that they were
not "geniuses" but they were members of the world's most prestigous
radiation protection committee and they had access to the world's
experts..There was no need for them to adopt the unproved and  inherently 
unprovable LNT policy. The ICTP could recommend what ever safe level of
radiation dose they wished. 
        It is true that in 1977 when the ICRP adopted the LNT model,  "we
didn't have the epidemiological studies... we have now." However, there was
one very good "epidemiological study" which showed clearly that  the
carcinogenic effects of alpha particles in the skeleton had a huge
threshold of 10 Gy (200 Sv). . (Robley Evans Radium in Man, HPJ November
1974 pp 497-510). In science when there is overwhelming experimental
evidence contradicting an assumption, one must abandon the assumption. The
ICRP did not do so. 
       .The motives of the ICRP for adopting the LNT are not clear. They
claimed that the LNT model represented a consensus of the scientific
community. I don't recall that they took a vote on the matter. It seems
likely that if a vote were taken now that not many would support the LNT
model. Evans studies reported in his  Radium in Man article also showed
that the stochastic model of radiation carcinogenesis was not valid.
Incidence of bone cancer in the radium cases was nearly constant at about
30% over the dose range from 20 Gy to 500 Gy - a dose factor  of 25! The
incidence is supposed to be proportional to dose if the stochastic model is
correct.
         Evans' article is seldom referenced in any NCRP report. NCRP
Report 115, Sec. 9.2 on Radium does not quote any of Evans' articles but
shows a graph of risk per mean skeletal dose using Ra-224.  It claims to
show that "at very low doses risk did vary linearly with dose (Figure
9.1)." The figure shows 10 data points with their 2 sigma error bars. A
horizontal line at a risk per Gy of 0.013 intersects all the error bars up
to a dose of about 15 Gy (300 Sv). I asked Robley Evans about this graph.
He said that it was inappropriate to compare the biological effects of
short lived Ra 224 to that of long lived Ra-226. It seems to me
inappropriate for the NCRP to ignore the data of Evans which dramatically
contradicts the LNT model.  .
        The LNT was undoubtedly adopted by the ICRP with good intentions
but the net effect of its great exaggeration of risk at trivial doses has
cost the world a fortune.  John Cameron.
John R. Cameron 2678 SW 14th Drive, Gainesville, FL 32608-2050 
phone: 352/371-9865; fax 352/371-9866  e-mail: jrcamero@facstaff.wisc.edu