Ben is a commenter on YouTube and wrote:
"And we're in the thick! I truly appreciate a substantive discussion. This is where I believe the conversation should be. What is the rate, then, of mutation accumulation? LNT-logic would produce a far-greater number than a non-linear low-dose response, which (while incredibly noisy data) epidemiology of higher env. background level populations seems to support. (It's so noisy, though, that I'm not advocating abandoning LNT as a regulatory tool. But I want to know what the reality is.) More data!"
Ben - please read some of my other webpages on the right hand side of this blog. If you have further questions, I'll be glad to answer them.
Ah - YouTube comment limits, indeed. Interesting blog! I will definitely poke around, and I suspect a truth that we're in general agreement belies this debate!
ReplyDeleteAs for my own blocked YouTube comment, "@TheHealthPhysicist (and for the curious), there is plenty of experimental data that suggests non-linearity in human radiation dose-response. See: http://www.pnas.org/content/109/2/443 , http://www.ncbi.nlm.nih.gov/pubmed/17149977 for studies I found most compelling, in add'n to population studies, e.g.,: http://www.ncbi.nlm.nih.gov/pubmed/19066487 "
Feel free to visit my own blog, (should be linked via my commenter name here), and likewise, feel free to contact me or question away!
I addressed your first link here, though I was probably too brief. If you understand what the DDREF is within LNT, then that is all that the study is confirming: http://ribjoint.blogspot.com/2011/12/whats-rif-that.html
ReplyDeleteYour second link deals with adaptive response (AR). There are many examples of AR but these are observed under very specific human-created situations.
The third link suffers from the ecological fallacy. Instead of measuring the dose people actually get, they used gross indoor/outdoor readings. If you look at the CI, it goes from <0 to >0, meaning that the statistical power is very low. The A-bomb survivor study is over 60 years old and over $1 BILLION. It has the best statistics.
Bob,
DeleteLet me check into the first to make sure I understand you fully.
The second, yes, is very controlled. However, that does not make it any less a legitimate identification of cellular response. Enough, indeed, to lend direct support for the idea of non-LNT (not conclude, but definitely suggest a legitimate possibility). I always find it curious that these studies are arm-waved away as though they have nothing to contribute...
The third, yes I agree with you that (obviously) the Japan cadre is the best source of data we have. However, they suffered large, acute doses, so actually reveal very little about actual low-dose response. This is where a cadre targeted more for dose and dose rate they received is more useful (zeroing in on the bottom of the graph). The statistics with the Japan cadre are excellent, yes, but again, you have to *pick* your initial assumptions (linear/non-linear) before performing your regressions. (And as a modeler, I know that there are an infinite number of mathematical models that can fit the data but are not accurate representation of reality.) e.g., the statistics *proving* that at the speed of sound friction on an aerodynamic surface goes to infinity was excellent, and therefore breaking the sound barrier was impossible... until we identified the proper boundary conditions (dare I say, thresholds?) and went and did it.
So, too, do I see the trimmings of non-linearity lurking in controlled studies (#2) and population studies (#3)...
Regarding the second study...it is a legitimate ID of cellular response. But that's not the issue. LNT is about excess dose versus excess cancer. There is a baseline radiation dose (background) and a baseline cancer rate in the absence of excess radiation above background.
DeleteIn the 1960's the epidemiology of the A-bomb study could only discern excess cancer at 100 rem. Then, people were bantering about non-LNT below 100 rem (morons!). With time, more people in the study got cancer and the statistics improved. Today, we can discern down to 10 rem.
The LNT deniers have employed "shifting of the goalposts". If we could get down to 1 mrem, they'd use the same tactics below that.
When you take the data and run a Maximum Likelihood Estimate (look it up), you get a most likely threshold of zero.
When you spend over a decade and $100M to find some biological response differential between say 15 rem and 1 rem (ie, straddling the 10 rem), one can't be found.
An analogy...we can only measure gravity down to a kg today. In the past it was higher. Is it equally plausible that gravity doesn't exist or is repulsive below 1 kg? Not without extraordinary evidence it's not.
Epidemiological studies follow a power law. If you want to see LNT to 1 rem (10x lower than current), then you estimate that you need about 10^2 times the number of people currently being studied in order to have the statistical power to get there. There are currently about 120,000 people in the study, so expect to need 12,000,000. Not too likely to happen. There is an ongoing One Million Worker Study, which might do better than the a-bomb one.
No one is picking linear/non-linear with the a-bomb study. It can be modeled as either linear or linear-quadratic. There is no statistical difference, so for parsimony we just go with linear.
If you want to continue on your quixotic quest of non-linearity, I wish you luck! People have been trying for decades!
I visited your blog...much nicer than mine. I do things very quickly here. I see your interested in space radiation. I've done a few posts on that, just search "space". You might find this particularly interesting:
ReplyDeletehttp://ribjoint.blogspot.com/2012/10/prediccs-radiation-fields-in-space.html
Very kind of you to say, but you have more content than I do, so we're even. ;)
DeleteHave to get back on target here, so I may not be able to reply for a bit, but I appreciate your very reasonable take on the issue, and I look forward to reading your thoughts on space radiation. It seems we are on two sides of an arbitrary line in the midst of the gray area between prevailing camps, which is where I believe the promise of making true progress actually exists.
Cheers,
Ben
Thanks...but extraordinary claims require extraordinary evidence. All of the evidence supports LNT, which is why it's been the scientific consensus for more or less as long as relativity (gravity) has. (See my earlier remark on gravity...a measurement threshold should not be confused with an effect threshold).
DeleteCheers back at ya',
Bob