AW: AW: [ RadSafe ] Excess relative risk

John Jacobus crispy_bird at
Wed Feb 13 21:10:06 CST 2008

  Thank you for the thoughtful reply.  I am not familiar with Sir Hill, but since he died in 1991, i doubt he had much to say about the the controversy that we have discussed.  So, I can only assume that your comments below are your own and not his comment.  I think that is a little misleading, but I assume is not deliberate. 
  Selection of data or studies is not new or really that unusual in science, and may be worse in epidemiological circles.  The proof of any theory or hypothesis is can it be tested.  In many of the studies that you might cite, such as the shipyard study, a reanalysis of the data shows that there may be fundamental flaws.  Of course, it you rely on flawed data, how does that advance the theory or hypothesis?  I think this has a lot to do with the specificity of the data sets used.  Personnally, I don't think any one study, or even tens of studies, can anwer the questions of what biological responses are going on at low dose, low-dose rate levels.  This is probably due to the number of end-points and confounding factors that are involved.  This is one reason that I believe there is no consistancy in hormetic studies.  The other being the use of questionable or flawed studies.
  Of course, some studies will run counter to the "conventional" wisdom of the hypothesis, but if you continually pick studies that bias the hypothesis, then you may not obtain the strength of the association between cause and effect.  In many cases the studies have different end-points, which should affect the consistency between the various studies.  If you look for some factor that you believe supports your hypothesis, you may miss what is really happening, e.g., irradiated cells produce a enzyme that you believe protect the DNA, when it has nothing to do with DNA protection.  You need to first identify those factors that related between the studies.  Your "consistant" results are then irrelevant to your hypothesis.  This is one reason I dislike comparing cell studies with animal studies with human epidemiology.  I believe that these are several of the points Sir Hill was making.
  I am not sure where you get the idea that I or "these experts" hold to the idea that reactions that occur lead conclusively to cancer after many years.  Actually, I do not get the point of your second paragraph, as I have never said it.  I do believe that such events as cancer are statistical in nature in the general population, and cancers may begin to develop many years before they are evident.  It is plausible that ionizing radiation dose cause some additional cancers due to cancers, but probably not as on the basis of one hit equals one cancer.  That is dumb, becasue molecular repair does occur in nature which confounds the study of radation risks.  However, natural cell repair mechanisms must also be imperfect mechanism, otherwise there would not be any cancers occurring naturally.   Yet, there is something to be said for the "target theory" of radiation effects, as cells are more senstive to ionizing radiation during their reproductive cycle, e.g., during M
 (mitotic) phase.  However, this is confounded by the fact that few cells are probably undergoing mitosis in the body compared to their growth state.
  For me,the LNT hypothesis is politically, but is a "short hand" tool for trying to manage or explain radiation risks when we really don't have a good model for these risks.  It is conservative, but provides a reasonable explanation until we have someting better, if ever.  
  Nevertheless, it has been politicalized by those with interests in using it as an excuse for the lack of new nuclear power plants.  However, neither LNT hypothesis nor ALARA were the cause for no nuclear constrution in the US.  I have believed for many years it was an economic issue.  Recently, new plants are planned for, but unless the utilities can make a profit, plant construction will again stop.  And despite your statement, it is clear that you believe in hormesis.  Otherwise, why to always cite studies that support it.
  I always enjoy your correspondence, even if we do not agree.  But I do wish you would not say I believe in this or that without asking.  
Rainer.Facius at wrote:

The selection below of two out of nine desiderata by one of the nestors of epidemiology, Sir Austin Bradford Hill, may provide for another explanation for our controversy.

1) You appear to rely on those experts which cherry-pick (to use one of your favorites) those studies which show positive associations - although often only for gratuitous choices of one-sided significance tests at significance levels of only 90% !!! I assume your position does not prevent you from realizing that well published human studies exist which display negative associations. Some of them have repeatedly presented to you (apart from Shipyards and domestic radon). Given that to my knowledge the number showing negative associations equals or even surpasses those with positive association, the very very least you are entitled to claim is that an essential proviso is violated which would allow you to “cry for causation”, i.e., consistency. By the way, the inconsistency comprises also the “volatility” of organs/sites for which ‘significant’ positive associations between low dose and dose rate exposures to sparsely ionizing radiation and cancer risk.

2) You appear to rely on those experts which still hold that knowledge of the statistics for the induction of initial ionization events in some 10^-15 s in a biologically important molecule suffice to predict conclusively the induction of cancers in an organ 10^8 s after such an event. It was this primitive target theory which guided experts more than half a century ago to formulate the LNT postulate, although already then the original founders of target theory had long ago pointed out its limitation to ‘inert’ systems which are incapable to mount counteractive responses. Half a century after Elkind and Sutton (1959) and a quarter century after Olivieri, Bodycote and Wolff (1984) and in view of the subsequent laboratory research such an assumption is implausible to the utmost in view of “the biological knowledge of the day”. Hence another proviso to “cry for causation” is violated. 

Though I would not claim that hormesis has been conclusively established for human cancer risks after low dose and dose rate exposures, plausibility and consistency (with respect to laboratory work) definitely favour it as compared to the LNT postulate. 

Given that the scientific case for causation is non existent in the relevant exposure regime, I think your occasional characterization of the LNT postulate as ‘political’ is warranted – as is the question “cui bono?”

Kind regards, Rainer 

  Dr. Rainer Facius
  German Aerospace Center
  Institute of Aerospace Medicine
  Linder Hoehe
  51147 Koeln
  Voice: +49 2203 601 3147 or 3150
  FAX:   +49 2203 61970
  Hill A B, 
The Environment and Disease: Association or Causation.
Proc Royal Soc Med 58(1965)295-300 (ISSN 0035-9157)

“Here then are nine different viewpoints from all of which we should study association before we cry causation. 

(2) Consistency: Next on my list of features to be specially considered I would place the consistency of the observed association. Has it been repeatedly observed by different persons, in different places, circumstances and times?

(6) Plausibility: It will be helpful if the causation we suspect is biologically plausible. But this is a feature I am convinced we cannot demand. What is biologically plausible depends upon the biological knowledge of the day.”

  Von: John Jacobus [mailto:crispy_bird at] 
Gesendet: Dienstag, 12. Februar 2008 04:34
An: Facius, Rainer; hflong at; ograabe at; radsafe at
Betreff: Re: AW: [ RadSafe ] Excess relative risk

  One of the purposes of a skeptic is not so much to challenge as to present what is unknown.  
  I have been accussed of being silent.  I am not an epidemiologist, so I have to relie on those who are recognized experts. (If you choose to ignore the conclusion of experts, that is your choice.)   The consensus has been that there are no demonstracted effects below 100 mSv.  Neither harmful or beneficial.  All studies are individual pieces of a puzzle.  To date, the well-known epidemiologists have reached the conclusion stated above.  Individual studies may support your position or mine, but the concensus has always remained the same.   
  Your uncited comments below are interesting, but how do they fit in the overall study of radiation effects?  I have seen some studies that do show negative slopes.  We can all cherry pick the data that supports our positions, but what do the experts say?  

"If history teaches any lesson it is that no nation has an inherent right to greatness.  Greatness has to be earned and continually re-earned."
- Norman Augustine, Chairman of the National Academies Committee 

-- John
John Jacobus, MS
Certified Health Physicist
e-mail:  crispy_bird at
Looking for last minute shopping deals?  Find them fast with Yahoo! Search.

More information about the RadSafe mailing list