Reply to Cohen

[dj_strom@ccmail.pnl.gov]

On Mon, 15 Jan 1996 Bernard L Cohen <blc+@pitt.edu> wrote:
>Subject: Re: Feasibility of Low-Dose Epi: Minimum Detectable Amount 
>On Thu, 11 Jan 1996 dj_strom@ccmail.pnl.gov wrote:
>>
>>   For common diseases, such as lung cancer, power
>> calcuations will show that you're hosed without a huge effect 
>> (remember, cigarette smoking has a 10-fold effect; even Cohen 
>> shows only a predicted 2-fold effect from radon, and without 
>> controlling for smoking on an individual case-by-case basis, 
>> you're not going to see anything).
>> - Dan Strom <dj_strom@pnl.gov>

>-----I must say it bothers me greatly to have people criticize my work in 
>off-hand ways. The way science works is for anyone finding fault with a 
>paper to send a letter-to-the-editor of the journal where it was 
>published pointing out what they object to, or at least let the author 
>know about the objection and discuss it with him. I keep hearing (usually 
>second or third hand) about people saying or implying that they don't 
>believe my paper on testing the linear- no threshold theory, but they 
>never say specifically why. I have tried everything to get people to come 
>out and say why--- including offering to pay them as consultants---but 
>with one trivial exception, no one has done so. Don't they believe in the 
>way science operates?

Science is never having to say "I can't tell you why, I just 
believe it."  The questioning of methods and results is a 
cornerstone of science.  Questions and criticisms should be 
directed at ideas, methods, and results, not people.  Nothing in 
my comments should be interpreted as a personal attack; they are 
a part of an ongoing, online scientific debate.  I do not regard 
them as off-hand, although perhaps I should avoid informal words 
like "hosed."  If RADSAFE isn't the place to express ideas such 
as mine, then perhaps it is time for me to UN-SUBSCRIBE. 

Obviously Bernie Cohen and I disagree on the level of credibility 
that his study has.  Regarding communication directly to him as 
opposed to discussing his work with others (that's permitted, 
right?), I must remind Dr. Cohen that, during the 8 years I was 
on the faculty at the University of Pittsburgh, he and I held 
appointments in the same department and communicated often.  
(BTW, I am most greatful to Dr. Cohen for his tireless and 
generous support of the students and faculty of the former 
Department of Radiation Health during that period, with his time 
and with his laboratory facilities, and for teaching us all so 
much about radon.)  Every single objection I hold to his 
epidemiology studies I have personally told him over the past 
decade, either at Pitt, at Health Physics Society chapter or 
national meetings, or other scientific venues; many more than 
once.  I have also published on the topic in the HPS Newsletter.  
I suppose I could dig through the files and find all the letters 
I have sent to Dr. Cohen over the years, including Austin 
Bradford Hill's association-causation paper, criteria for 
epidemiology studies, publication bias, and on and on.  That, 
too, is "how science operates."  

I have personally asked two members of the National Academy of 
Sciences National Research Council BEIR VI committee to ensure 
that that committee reviews Dr. Cohen's work and the linear, 
no-threshold dose-response hypothesis in general, including its 
application for the risk management activity we call radiation 
protection.  I'll await their response.  I hope that they will 
once again articulate why some scientists don't put much credence 
in studies that do not associate exposures with individuals over 
time, regardless of the controls that are made.

This message is my own opinion, and has not been approved by 
Battelle or the U.S. Department of Energy.

- Dan Strom dj_strom@pnl.gov