Re: "Overmatching" is an Oxymoron. DBPC May Force LNT Out

[epirad@mchsi.com]

A new study of interest -



Am J Epidemiol 2002 Sep 15;156(6):548-55 

  

Exposure to Residential Radon and Lung Cancer in Spain: 

A Population-based Case-Control Study.



Barros-Dios JM, Barreiro MA, Ruano-Ravina A, Figueiras A.



Department of Preventive Medicine and Public Health, 

School of Medicine, University of Santiago de 

Compostela, Santiago de Compostela, Spain.



Although high radon concentrations have been linked to 

increased risk of lung cancer by both experimental 

studies and investigations of underground miners, 

epidemiologic studies of residential radon exposure 

display inconsistencies. The authors therefore decided 

to conduct a population-based case-control study in 

northwest Spain to determine the risk of lung cancer 

associated with exposure to residential radon. The study 

covered a total of 163 subjects with incident lung 

cancer and a population sample of 241 cancer-free 

subjects since 1992-1994. Odds ratios for radon were 

estimated using logistic regression adjusted for sex, 

age, lifetime tobacco use, family history, and habitat. 

The adjusted odds ratios for the second, third, and 

fourth quartiles of radon (breakpoints: 37.0, 55.2, and 

148.0 Bq/m(3)) were 2.73 (95% confidence interval (CI): 

1.12, 5.48), 2.48 (95% CI: 1.29, 6.79), and 2.96 (95% 

CI: 1.29, 6.79), respectively. An additive synergic 

effect between radon and tobacco was found. The results 

from this study suggest that, even at concentrations far 

below official guideline levels, radon may lead to a 2.5-

fold rise in the risk of lung cancer. Furthermore, the 

synergy found between smoking and radon may prove useful 

when it comes to drafting public health recommendations.

> Bill, ANS and HPs

> "Overmatching" is an oxymoron - for definitive study or clinical trial, such

> as Cameron or I would propose. Double-blind,  neither subject or investigator

> even knows who are the controls.

> 

> Even with Double Blind, Placebo Control (DBPC), application must be only to

> 1, like population (not your 1% outlier Iowa),

> 2, like sub-population (only ever-smoking women over 50 have more CV risk with

> Prempro, they will find),

> 3 like individual (allergy, disease, etc make medicine both art and science).

> 

> The Prof. Field referenced by Am Cancer Soc for "Radon Causes Cancer", could

> help prove whether or not, "Radon at 2-3pCi/L, 74-111Bq(m-3), in living area,

> is associated with less c-reactive protein, more T cells, less lung cancer

> and/or greater longevity than when less than 1pCi/L"

> 

> The more "multivariate" the more probibility of error. I won't trust

> statistical correction

> for smoking, a factor 10x the one in question, when we can both study

> non-smokers with lung cancer (few as they are) and placebo control study

> (difficult as it is).

> The Am Cancer Soc  applies 1% outlier Iowa data to the 99%, although it

> conflicts with animal and ecologic results.

> 

> LNT must be dumped, says the precautionary principle, if we prove that BENEFIT

> from low dose radiation is being withheld.

> 

> Howard Long

> 

> epirad@mchsi.com wrote:

> 

> > Howard,

> >

> > Please read the study Mr. Muckerheide posted concerning

> > the problems with over matching in a case-control study.

> >

> > Howard, it is really not my position to accept the

> > ethics, that would be up to the Human Subjects Internal

> > Review Board at the facility where such a study would be

> > performed.  I would not have to sign on for the study to

> > take place.  It may be helpful to identify the rate

> > limiting steps that prevent the funding of such a

> > study.

> >

> > My criticism of your proposal is more a matter of

> > reality.  It would be impossible to perform a placebo

> > controlled study of radon in the general population.

> >

> > The multivariate analyses performed in case-control

> > studies has tremendous power to adjust for smoking and

> > other potential confounders since data is collected at

> > the level of an individual.

> >

> > Bill

> > > Bill, Bernie, rad scientists and HPs,

> > > Radon, like all other MEDICINES, must be tested with maximum matching

> > > possible (placebo-controlled studies, double blinded), to definitively

> > > identify most beneficial dosage, frequency of bad results (unusual

> > > sensitivity), etc.

> > >

> > > Why does Field refuse to accept the ethics of such a study? In Iowa,

> > >  37% smoking "matched controls" (for 87% smoking lung cancer cases),

> > > elicits incredulity about the validity of  "statistical correction".

> > > True controls must be identical to cases in everything except the

> > > chemical being tested, even to placebo effect (usually 30%).

> > >

> > > Cohen is correct here. Field refuses for radon, a test required of every

> > > other medicine.

> > >

> > > Howard Long

> 

> ************************************************************************

> 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/

> 

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

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/