[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Glass-Based Radon Measurements
Friends,
For those interested in Radon, I have calculated the odds ratios from the
data given in the first 2 rows of Table 4 of the Lagarde et al. paper and
graphed them. These rows give the number of cases and controls as a function
of radon concentrations as assessed by air based and surface based
measurements respectively. The graph can be seen at
http://members.shaw.ca/eic/odds.pdf
Also shown on the graph (dashed lines) are the relative risks as reported by
the authors.
To me at least, it seems that the raw odds tell a different story than the
RR values.
Please let me know if I made any mistakes in calculating the odds ratios.
Kai
http://www.eic.nu
and http://www.gammawatch.com/
----- Original Message -----
From: "Kai Kaletsch" <info@eic.nu>
To: "RadSafe" <radsafe@list.vanderbilt.edu>
Sent: Sunday, September 08, 2002 7:07 PM
Subject: Fw: Glass-Based Radon Measurements
> Friends,
>
> As a result of my request, a radsafer has sent me a copy of the Lagarde et
> al. paper. I still seem to have a problem wrapping my head around this
> case - control stuff. I would very much appreciate it if someone could
take
> the time to help me out (if you want to see the paper, you can get to a
pdf
> version from
> http://www.nature.com/jea/journal/v12/n5/index.html ):
>
> According to the first row of Table 4 of the Lagarde et al. paper there
are
> 33 cases and 58 controls in the 50 to 80 Bq/m3 category ( i.e. 1.76
controls
> per case) and 24 cases and 55 controls in the 80 to 140 Bq/m3 category (
> i.e. 2.29 controls per case). Relatively speaking, there are 30 % more
> healthy people (controls) in the higher dose category than in the lower
dose
> category. Yet, the authors calculate a higher relative risk for the 80 to
> 140 Bq/m3 category than in the 50 to 80 Bq/m3 category (1.42 vs. 1.28).
>
> To me, that doesn't make any sense. If you asked me if I wanted to belong
to
> a group that is full of sick people or to a group that is full of healthy
> people, I would answer: "I want to belong to the group that has lots of
> healthy people, because that means my risk of becoming sick is lower."
>
> Could someone please try to explain this in common sense terms?
>
> Thanks in advance,
> Kai
>
> ----- Original Message -----
> From: "Kai Kaletsch" <info@eic.nu>
> To: <epirad@mchsi.com>; <radsafe@list.vanderbilt.edu>
> Sent: Saturday, August 31, 2002 9:42 AM
> Subject: Re: Glass-Based Radon Measurements
>
>
> > Bill and Friends,
> >
> > I'm still trying to wrap my head around what the actual observables are
in
> a
> > case - control study. The way I understand it, case - control studies
work
> > like this:
> >
> > 1. Find people with a disease (cases).
> > 2. Match people to the cases that are similar (controls).
> > 3. See if there is a difference in exposure to a suspected causal
> agent.
> > 4. If there is a difference in exposure, you calculate an excess
> relative
> > risk (ERR) per unit exposure.
> >
> > The only observables are related to step 3, i.e. differences in
exposure.
> >
> > My question is: Is there any way that we can infer the observables
> > (differences in exposure) from the reported calculated results? (I don't
> > have ready access to J Expo Anal Environ Epidemiol 2002
Sep;12(5):344-54)
> >
> > For example, could the statement:
> >
> > "...the excess relative risk (ERR) of 75% (-4% to 430%) per 100 Bq m(-3)
> > obtained when using a continuous variable for surface-based average
radon
> > concentration estimates, were about twice the size of the corresponding
> > relative risks obtained among these subjects when using air-based
average
> > radon concentration estimates."
> >
> > mean:
> >
> > "surface-based average radon concentration estimates showed only 1/2 as
> much
> > excess radon concentration of cases vs. controls than using air-based
> > average radon concentration estimates."?
> >
> > In physics, the authors of studies usually put their raw data somewhere
so
> > that anyone can download it, re-analyze it, check for errors in
arithmetic
> > etc etc. Is this not usual in epidemiology? (Obviously one would have to
> > hide the personal info of the participants, but that shouldn't be a
> > problem.)
> >
> > To me, this would add more credibility to the studies. If the same
people
> > are designing the study and collecting the data and analyzing the data
and
> > writing conclusions about the data, would that not tend to amplify any
> > biases these people might subconsciously have by a factor of 4?
> >
> > Best Regards,
> > Kai
> >
> > ----- Original Message -----
> > From: <epirad@mchsi.com>
> > To: <radsafe@list.vanderbilt.edu>
> > Sent: Thursday, August 29, 2002 3:14 PM
> > Subject: Glass-Based Radon Measurements
> >
> >
> > > Two recent papers that may be of interest.
> > >
> > > Regards, Bill Field
> > > bill-field@uiowa.edu
> > >
> > > J Expo Anal Environ Epidemiol 2002 Sep;12(5):344-54
> > >
> > > Glass-based radon-exposure assessment and lung cancer
> > > risk.
> > >
> > > Lagarde F, Falk R, Almren K, Nyberg F, Svensson H,
> > > Pershagen G.
> > >
> > > Institute of Environmental Medicine, Karolinska
> > > Institutet, Stockholm, Sweden.
> > >
> > > Lung cancer risk estimation in relation to residential
> > > radon exposure remains uncertain, partly as a result of
> > > imprecision in air-based retrospective radon-exposure
> > > assessment in epidemiological studies. A recently
> > > developed methodology provides estimates for past radon
> > > concentrations and involves measurement of the surface
> > > activity of a glass object that has been in a subject's
> > > dwellings through the period for exposure assessment.
> > > Such glass measurements were performed for 110 lung
> > > cancer subjects, diagnosed 1985 to 1995, and for 231
> > > control subjects, recruited in a case-control study of
> > > residential radon and lung cancer among never-smokers in
> > > Sweden. The relative risks (with 95% confidence
> > > intervals) of lung cancer in relation to categories of
> > > surface-based average domestic radon concentration
> > > during three decades, delimited by cutpoints at 50, 80,
> > > and 140 Bq m(-3), were 1.60 (0.8 to 3.4), 1.96 (0.9 to
> > > 4.2), and 2.20 (0.9 to 5.6), respectively, with average
> > > radon concentrations below 50 Bq m(-3) used as reference
> > > category, and with adjustment for other risk factors.
> > > These relative risks, and the excess relative risk (ERR)
> > > of 75% (-4% to 430%) per 100 Bq m(-3) obtained when
> > > using a continuous variable for surface-based average
> > > radon concentration estimates, were about twice the size
> > > of the corresponding relative risks obtained among these
> > > subjects when using air-based average radon
> > > concentration estimates. This suggests that surface-
> > > based estimates may provide a more relevant exposure
> > > proxy than air-based estimates for relating past radon
> > > exposure to lung cancer risk.
> > > --------------------------------------------
> > >
> > >
> > > Health Phys 2002 Aug;83(2):261-71
> > >
> > >
> > > 210Po implanted in glass surfaces by long term exposure
> > > to indoor radon.
> > >
> > > Steck DJ, Alavanja MC, Field RW, Parkhurst MA, Bates DJ,
> > > Mahaffey JA.
> > >
> > > Physics Department, St. John's University, Collegeville,
> > > MN 56321, USA. dsteck@csbsju.edu
> > >
> > > Recent epidemiologic investigations of the relationship
> > > between residential radon gas exposure and lung cancer
> > > relied on contemporary radon gas measurements to
> > > estimate past radon gas exposures. Significant
> > > uncertainties in these exposure estimates can arise from
> > > year-to-year variation of indoor radon concentrations
> > > and subject mobility. Surface implanted 210Po has shown
> > > potential for improving retrospective radon gas exposure
> > > estimates. However, in previous studies, the ability of
> > > implanted 210Po activity to reconstruct cumulative radon
> > > gas exposure was not tested because glass was not
> > > available from homes with known radon-gas concentration
> > > histories. In this study, we tested the validity of the
> > > retrospective radon gas reconstruction using implanted
> > > 210Po surface activity by measuring glass surfaces from
> > > homes whose annual-average radon gas concentrations had
> > > been measured almost every year during two decades.
> > > Regression analysis showed a higher correlation between
> > > measured surface activity and cumulative radon gas
> > > exposure in these homes (R2>0.8) than was observed in
> > > homes where only contemporary radon gas measurements
> > > were available. The regression slope (0.57 ky m(-1)) was
> > > consistent with our earlier retrospective results.
> > > Surface activity measurements were as reliable for
> > > retrospective radon gas exposure reconstruction as
> > > yearlong gas measurements. Both methods produced
> > > estimates that were within 25% of the long-term average
> > > radon gas concentrations in a home. Surface measurements
> > > can be used for home screening tests because they can
> > > provide rapid, reliable estimates of past radon gas
> > > concentrations. Implanted 210Po measurements are also
> > > useful in retrospective epidemiologic studies that
> > > include participants who may have been exposed to highly
> > > variable radon concentrations in previously occupied or
> > > structurally modified homes.
> > >
> > >
> > >
************************************************************************
> > > 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/
>
************************************************************************
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/