Re: Wing: Descriptive Epidemiology by Any Other Name...

[Bob Flood <bflood@SLAC.Stanford.EDU>]

At 08:56 PM 2/27/97 -0600, you wrote:
>Bob, This "logical", but Dan's premise is utterly false. This is NOT a
>matter of an "unacceptable methodology". This methodology is used VERY
>extensively in science.

Apparently I have created an inaccurate view of my position on this issue.
Let me clarify things:

I had the pleasure of hearing Dr. Lucky expound on radiation hormesis about
10 years ago and was thoroughly pusuaded. After listening to agruments on
boths sides for another few years, I came to my own conclusion that the
hormesis phenomenon may exist and has strong logical arguments to support
it, but the reality of the math dictates that no definitive case-control
study is practical to prove the LNT hypothesis or hormesis. Simply put, the
background is too high to find a small harmful or beneficial effect.

I find Dr. Cohen's study just as pursuasive. He has gone to extraordinary
lengths to challenge the method (ecological study) and the data and invited
an unlimited number of others to do the same, and I have seen no pursuasive
counter agruments.

My original point on the Wing study was this: Dr. Cohen has shown that an
ecological study can be done with great scientific integrity and produce a
valuable result. However, I was seeing on Radsafe some descriptions of
Wing's study as fundamentally flawed because it used the ecological
approach. I disagree with such an assessment. It is not the analytical
approach that is flawed. As usual in life, the devil is in the details. A
very good overview of the Wing study was distributed on Radsafe and showed
enough technical flaws in the study to convince me that the report's
conclusions are not justified.


Bob Flood
Stanford Linear Accelerator Center
(415) 926-3793     bflood@slac.stanford.edu
Unless otherwise noted, all opinions are mine alone.