Re: Research Subject Doses

[JMUCKERHEIDE@delphi.com]

>         I would say that the only sure thing is that the risk is very
> small (e.g. not larger than the linear estimate), and it is just as likely
> to be beneficial as harmful.
>
> Bernard L. Cohen
> Physics Dept.
> University of Pittsburgh
> Pittsburgh, PA 15260
> Tel: (412)624-9245
> Fax: (412)624-9163
> e-mail: blc+@pitt.edu

It would seem to me, based on the weight of the actual epidemiological and
biological evidence, that it would be "more likely" to be beneficial as
harmful, though even more likely that "it's not enough to effect health either 
way". (You'ld get more DNA damage from drinking a cup of coffee or running a
hundred yards. I haven't seen evidence of the stimulation of the immune
response that can prevent/reduce cancer at 0.5 rem.) 

However, in all of this discussion, it would seem that a primary source
reference missing is the HPS' own Position Statement: that the radiation
health effects data provides no basis to calulate risk at less than 5
rem/year, or 10 rem lifetime. 

(Re BEIR V, reading their own details acknowledges clear thresholds for many
endpoints in which they then draw a straight line anyway. The argument is that 
this is ok for "rad protection policy/planning", but does not reflect actual
potential health effects should be taken to heart.) 

Regards, and best to all for the new year,

Jim Muckerheide
jmuckerheide@delphi.com
muckerheide@mediaone.net

> On Mon, 29 Dec 1997, Chuck Pickering wrote:
> 
> > Happy Holidays,
> > 
> > All you Linear and Non-linear combatants please sharpen your pencils. 
> > I am reviewing a human subject research study that proposes to expose 
> > the subjects to a dose (TEDE) of 0.5 Rem. Extrapolating LINEARLY from 
> > BEIR 5 data suggests that the risk from participating in this study is 
> > about 4 deaths per 10,000 (10 Rem = 800 per 100,000). 
> > 
> > I am deliberately not giving you all the information on radionuclide, 
> > dose schedule, administered dose, etc. I am interested in your 
> > thoughts on extrapolating in this fashion and if you have suggestions 
> > on what the risk might be if one extrapolated using a non-linear 
> > method.
> > 
> > Chuck Pickering, RSO
> > University of Southern California
> > (213) 342-2200
> > (213) 342-2201 Fax
> > cpickering@engr-serv.usc.edu
> > 
> 
> 
>