[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

TRAB



Michael,



First, do your views represent the Texas Radiation Advisory Board (TRAB)?



It is interesting that you failed to post the Billet site for Dr. Field's 

response.  Dr. Field indicated his letter had not been answered even though 

he says he brought up his concerns about the letter a long while before 

September 11, 2001.



I think his response in the Billet more than answered the TRABs extremely 

erroneous allegations.  Please post the site so others can see his response.



And how long does it take to write a letter, you just did it.  The TRAB 

could have easily responded before now if they felt they had a valid 

response.



Don

>I've made "more misstatements?"  Where?

>

>I make things up as I go along to justify my own paradigms?  Care to 

>elaborate?



Unfortunately, I deleted your post right after reading it, so I no longer 

remember the specifics of what you said. I will have to wait until Feb 1 to 

see the archives.





We took issue with several aspects of the study that, we felt, gave reason 

to question the validity of the results.



The scientific method would be to write a letter-to-the-editor, rather than 

bashing it in a letter to the governor.



>

>Although we acknowledge respectful disagreement with Dr. Field on this 

> >matter, we will "recant" nothing



Why am I not surprised?



---------------------------------------->

>First, the IRLCS was only able to establish ³statistical significance² by

>omitting cases that died during the study period.



If you read the paper again, you will see a statistically significant dose 

response (positive finding) using the categorical analysis and ALL subjects 

so you are clearly wrong.





--------------------------

You said, they only found significance by excluding cases that died during 

the study.  They did NOT exclude any subjects that died during the study. SO 

you are obviously wrong again.  HOWEVER, they did a sub analysis of subjects 

who did not fill out a questionnaire themselves because they were deceased 

prior to contact, not during the study.  The percent of live cases in the 

Iowa Study is extremely impressive and is the highest of any radon case 

control study I know of.  They said they used rapid reporting to get out to 

the subjects as soon as possible since 50% of lung cancer subjects died the 

first year after diagnosis.



>Second,  the data set was "carved up into non-uniform intervals.  Again, 

>Dr.

>Field disagrees with this.  This may be a fine point, but they are not

>uniform.  If the authors chose to define the intervals down to the

>"common-place" (or not) 4.23 WLM, they did not succeed.   The intervals 

>were

>4.23, 4.23, 4.22, 4.23 and >16.95.  They may have made an a priori decision

>on what the intervals would be, but the decision to break down the 

>intervals

>to the 23/100 of a WLM is not well-described and for no apparent reason.

>Why?







Michael wrong again, read the paper.  Are you really serious going out to 

that many decimal places?  I'll tell you - that looks pretty darn even to 

me. It says in the paper that they a priori decided they needed 15% of the 

subjects in the highest exposure category to allow a sufficient number for 

statistical analysis.  After they did that they broke the categories up into 

equal categories.  I don't really think it is relevant how many places you 

carry it out to since it would not make any difference in the findings.





----------------------------



>

>Third, the disparities in smoking histories between subjects and controls

>were "poorly corrected."  Perhaps "corrected" was not the best word to use

>in this instance.  A more appropriate term would have been "controlled" (or

>"poorly controlled").  Again, Dr. Field disagrees.





Michael wrong again, The modeling they did for smoking was very well done 

and in fact it you read the paper they even tested for residual confounding 

from smoking.



As Dr. Field has said, you can not match completely on smoking smoking.  How 

do you match, pack-year rate, numbers of packs per lifetime, year started, 

year ended, etc?  If you pick one, then the other variables will not be 

matched.  It is obvious you do not understand the process of adjusting for 

smoking in a mutivariate analysis.



Dr Field has already wrote a letter to the Billet and has not had a response 

from you or the TRAB in the Billet. If you have the time to post this 

response you have the time to respond to the Billet.  It may not be a 

response from the TRAB to his letter, but it is better than no response at 

all.



Don



_________________________________________________________________

Send and receive Hotmail on your mobile device: http://mobile.msn.com



************************************************************************

You are currently subscribed to the Radsafe mailing list. To unsubscribe,

send an e-mail to Majordomo@list.vanderbilt.edu  Put the text "unsubscribe

radsafe" (no quote marks) in the body of the e-mail, with no subject line. You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/