European Residential Radon Pooling

[niton@mchsi.com]

Maury,



I recall you have heard this term a few years ago, see: 

http://www.vanderbilt.edu/radsafe/0201/msg00754.html



Pooling is not a new buzzword as can be seen by its use in the papers below.  

However, it may be a new term for many in the Health Physics community.  Some 

people may have loosely used meta-analysis to describe the use of raw data in 

the past.   See these papers from the early 1990s. Pooling is a much more 

rigorous approach to combining data from various studies. Pooling allows 

greater diagnostic ability - for example - to determine hetereogeneity between 

studies or determine whether one or two studies are driving the results of the 

entire pooling.  These diagnostics will be described in great detail when the 

results of the North American Pooling are published in 2005. Feel free to email 

me directly if you would like more details bill-field@uiowa.edu  Merry 

Christmas!



>From an occupational perspective.

J Occup Med. 1991 Dec;33(12):1257-60. 

Data pooling in occupational studies.

Checkoway H.



Department of Environmental Health, University of Washington, School of Public 

Health and Community Medicine, Seattle 98195.



Summarizing epidemiologic data from multiple studies can be accomplished EITHER 

by a meta-analysis of published findings OR by pooling data and performing new 

analyses of enlarged data sets. Meta-analysis is relatively easy and 

inexpensive to perform but usually is restricted to examination of overall 

risks. Data pooling offers a more direct approach for subcohort analyses and 

for estimation of dose-response relationships. The advantages of pooling data 

from multiple studies of workers with similar exposures pertain primarily to 

enlarged study bases and a greater potential for evaluating statistically 

stable dose-response relationships than can be accomplished by meta-analysis. 

Valid methods for data pooling that are developed for cohort mortality studies 

of relatively large worker populations ultimately can be applied to studies of 

other health outcomes using different study designs.





Int J Technol Assess Health Care. 1990;6(1):5-30. 



Some statistical methods for combining experimental results.



Laird NM, Mosteller F.



Harvard School of Public Health, Boston, MA 02115.



Advances in science and technology are generally the product of multiple 

investigations. This article discusses statistical methods for combining 

empirical results from a series of different experiments or clinical 

investigations. We delineate the steps an assessor might take in combining data 

from different studies and provide references for topics not discussed in 

detail. The article reviews some of the most commonly used statistical 

techniques for combining results in the medical and social sciences.







> To the best of my flawed (rusted) recollection, the meta-studies that I 

> have seen reported pooled the raw data from mutiple more or less related 

> studies -- not risk estimated risk levels or significance levels. I'm 

> sorry, Bill, but I fail to see the difference. Pooling as it is being 

> used in this thread seems to be just a new buzzword. By the same token, 

> there have been "good" meta-studies or meta-analyses, and there have 

> been a lot that were poorly done. This is why I still believe that great 

> caution and detailed review is warranted regarding conclusions from such 

> studies. The designs and analyses of such studies have been fraught with 

> flaws.

> Cheers,

> Maury Siskel         maurysis@ev1.net

> 

> niton@mchsi.com wrote:

> 

> >Barbara,

> >

> >Pooling combines the DATA (not risk estimates) from each study. 

> >

> >

> >"Consider the danger of radon gas.  If there is one environmental problem that 

> is real, it is radon.  ...there is no hysteria over adon....because it's 

> natural...."Rush Limbaugh, 1992 

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

> >

> >  

> >

> 

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

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/