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

Re: A question of statistical significance vs operational significance



Maury,



With detailed individual level data, you can account for 

the noise through multivariate regression.  As with most 

case control studies, the largest uncertainty in many of 

the case-control studies has been obtaining good 

retrospective exposure information.  In a case-control 

study, unlike an ecologic study, the noise (or random 

misclassification) generally makes it more difficult to 

find an assocaition if one exists.  In an ecologic study 

such as Dr. Cohen's random misclassification often 

drives the assocaition toward either a positive or 

negative assocaition.  So when people ask to explain his 

findings, it is like asking someone to explain 

randomness.



See:  http://www.cheec.uiowa.edu/misc/rd_review.pdf



The recent paper we published in the Journal of Exposure 

Analysis and Environmental Epidemiology demonstrates how 

less sophisticated dosimtery methods for case-control 

studies reduce the risk estimtess because of problems 

with obtaining an accurate retrospective exposure 

assessment.  Please remember, the 50% increased risk we 

noted for the Iowa Radon Lung Cancer Study was for an 

average exposure of 4 pCi/L (EPA action level).  We 

noted higher risks for higher exposures. These findings 

were not dependent on LNTT extrapolations.   



SEE:

J Expo Anal Environ Epidemiol 2002 May;12(3):197-203

Residential radon exposure and lung cancer: Variation in 

risk estimates using alternative exposure scenarios.

Field RW, Smith BJ, Steck DJ, Lynch CF.

Department of Epidemiology, College of Public Health, 

University of Iowa, Iowa City, Iowa, USA.



The most direct way to derive risk estimates for 

residential radon progeny exposure is through 

epidemiologic studies that examine the association 

between residential radon exposure and lung cancer. 

However, the National Research Council concluded that 

the inconsistency among prior residential radon case-

control studies was largely a consequence of errors in 

radon dosimetry. This paper examines the impact of 

applying various epidemiologic dosimetry models for 

radon exposure assessment using a common data set from 

the Iowa Radon Lung Cancer Study (IRLCS). The IRLCS 

uniquely combined enhanced dosimetric techniques, 

individual mobility assessment, and expert histologic 

review to examine the relationship between cumulative 

radon exposure, smoking, and lung cancer. The a priori 

defined IRLCS radon-exposure model produced higher odds 

ratios than those methodologies that did not link the 

subject's retrospective mobility with multiple, 

spatially diverse radon concentrations. In addition, the 

smallest measurement errors were noted for the IRLCS 

exposure model. Risk estimates based solely on basement 

radon measurements generally exhibited the lowest risk 

estimates and the greatest measurement error. The 

findings indicate that the power of an epidemiologic 

study to detect an excess risk from residential radon 

exposure is enhanced by linking spatially disparate 

radon concentrations with the subject's retrospective 

mobility. DOI: 10.1038/sj/jea/7500215

 

Regards, Bill Field

> 

> 

> Bill, thanks for taking the trouble to refer me to relevant data. My

> impressions, however, are that you folks are suggesting with an

> epidemiological risk factor of 0.5, that out of an annual total lung

> cancer

> incidence of 157,400 cases, 18,600 or about 12% are attributable to

> radon. I

> don't believe those radon cases could hope to be distinguished from the

> noise

> or error variance.  When thinking about the increasing weight of

> evidence

> favoring beneficial health effects from exposure to low level radiation

> (such

> as household radon), I cannot bring myself to get very concerned about

> the EPA

> radon campaign -- except for some of my darker suspicions which already

> have

> been well-fed over the years by the performance of EPA. I just cannot

> view

> radon as a threat and I suspect it might even be beneficial to us.

> Perhaps my

> ignorance, but time will tell after I'm long gone.

> 

> Thanks again for your response to me.

> Sincerely,

> Maury                   maury@webtexas.com

> ================================

> epirad@mchsi.com wrote:

> 

> > Maury,

> >

> > Our direct observations

> > http://www.cheec.uiowa.edu/misc/radon.html are in

> > agreement with the BEIR VI

> > (http://www.epa.gov/iaq/radon/beirvi1.html)projections

> > which estimate that approximately 18,600 lung cancer

> > deaths each year in the United States are associated

> > with prolonged radon progeny exposure.

> >

> > Bill Field

> > > Bill,

> > >

> > > Would you select an objective, reliable measure of the impact of

> > > cancer on human health, e.g., mortality, morbidity, longevity, etc. and

> > > tell me what that observation is today in perhaps the US, or the world,

> > > or Iowa, or whatever? Then, if we could suddenly cause all radon and its

> > >

> > > progeny to disappear completely from the earth while all other

> > > conditions remain unchanged, what do you see in any hard data, or

> > > believe would be the observed effect or change in that selected cancer

> > > measurement in, say, 20 years or so?

> > > Cheers,

> > > Maury Siskel           maury@webtexas.com

> 

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

> It is the soldier, not the campus organizer, who has given us the

> freedom to demonstrate.                       Charles M. Province

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

> 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/