Re: Life Span Study of Hiroshima and Na

[John Goldsmith <gjohn@bgumail.bgu.ac.il>]

Jim:
	It is difficult to carry on a dialogue if we are not looking at 
the same information, but: (1) please get and study Shigematsu's book  if 
you want to examine the most recent RERF findings in the English language.
(2) I assume the raw data are in Japanese, and short of unlikely fluency 
and incredible patience would noy help you much.  Epidemiological data are
usually interpreted by grouping subjects with similar exposures.(3) I 
suppose you refer to the instability of dose relationships as "severe 
sawtooth results".  All Shigematsu's figures give confidence bounds.  
For example his figure 2, Chapter 18 (Life Span) shows a relative risk of 
non-malignant deaths for those less than 40 at the time of the bomb and 
followed during 1966-1985, of about 1.35 for 2-3 Gray(significant), 1.20 for 
3-4 Gray and 2.4 (significant) for 5-6 Gray. We (Goldsmith and Kordysh) have 
written elsewhere on why dose-response rates should not be linear (J. Exposure
Analysis and Env.Epidemiol. vol3. 259-276, 1993).  
	Shigematsu et al. cite a relative risk on p. 327 at 2 Gy for these 
data as 1.19 (95% C.I. 1.10-1.29), compared to cancer 1.78 ( 1.64-1.92)  These
are reasonable in my view. To have equipped each member of a target population
with a film badge or Geiger counter before dropping the bomb may have provided
individual exposure data, but would not otherwise be reasonable. 
	John Goldsmith( gjohn@bgumail.bgu.ac.il,) Prof. Epidemiology
 (On Wed, 27 Dec 1995 JMUCKERHEIDE@delphi.com wrote:

> Dr. Goldsmith,
> 
> [Group: Sorry about my "delayed" message re the RERF data. It was in a machine 
> that I had used for email when my primary computer was out, and I used it the
> other day for email and was surprized to send a message from the email
> reader.] 
> 
> However, the point that was explored then was that the raw RERF epi data is
> not available. There are only analyses done with the grouped data. This seems
> to have been confirmed (even by a major int'l figure who worked on BEIR V!) If 
> anyone knows whether there is any new info on this? (My curiosity was
> originally related to the effect of the selected bins on the severe "sawtooth" 
> results in the <40rem dose range.) Does anyone know whether the estimated
> individual doses are reported as single values, or just as ranges, which would 
> make the issue difficult to address and require a lot of work? 
> 
> > Detailed data were published this year by Shigematsu, I, et al in 
> > "Effects of A-Bomb Radiation on the Human Body" by Harwood Academic 
> > Publishers and Bunkodo Co. Ltd. It is well illustrated and translated 
> > from the Japanese. ISBN 3-7186-5418-0
> > Excess mortality for other than cancer and hematological disease during 
> > 1966-1985 is increasing above 2.5 Gy for those under 40 ATB, but not for 
> > those older than 40 ATB.   John Goldsmith( gjohn@bgumail.bgu.ac.il)
> 
> I appreciate the reference. I will try to find it. I have been using the "no
> effects at <200 rad" data (in Kondo, from Shimizu). You are indicating that
> this now reports excess mortality only above 250 rad? Only for people exposed
> at <40 years old? (Since >40 would now be >90 we can assume most has been
> expressed, although since the longevity is greater for the survivors, do we
> know if age-adjustment is adequately considered? especially for leukemia
> (which is common for people >90 who have had long excellent health - as we
> have recently seen in our friends and family.) 
> 
> Thanks.
> 
> Regards, Jim Muckerheide
> jmuckerheide@delphi.com
>