Meta article on radiation and health effects

Cyril R
Posts: 5037
Joined: Jul 14, 2008 3:12 pm

Meta article on radiation and health effects

Post by Cyril R » Nov 04, 2014 2:27 pm

An anti-nuclear person provided the following reference in support of LNT.

http://www.columbia.edu/~djb3/papers/pnas4.pdf

It has a very ambitious title, "Cancer risks attributable to low doses of ionizing
radiation: Assessing what we really know"

Despite the ambitious title, it is full of holes.

It clearly disproves LNT by the graphs alone, and it makes many statements that are not supported by the raw data. It also has much fuzzy wording such as this is probably caused by this and that.

Graph number 4 shows decreasing cancer risk above 200 mSv or so, and higher cancer risk near the zero exposure point than the point near 50-100 mSv. This clearly is no proof of LNT.

The so called "protracted doses" aren't really chronic at all, more like semi-prompt, such as bomb fallout where most of the dose is in the short term from short lived FPs, so this doesn't help much to explain say 1 mSv/day of exposure for years.

I'm looking for a more detailed critique of this article. Does it exist or can someone pull it apart for me.

Joshua Maurice
Posts: 186
Joined: Sep 15, 2011 7:58 pm

Re: Meta article on radiation and health effects

Post by Joshua Maurice » Nov 06, 2014 7:50 pm

From the paper:
The epidemiological data suggest that it is [approx] 10–50 mSv for an acute exposure and [approx] 50–100 mSv for a protracted exposure.
Nothing new here. IIRC, that's right (for a proper definition of "protracted exposure").

(Emphasis added:)
Second, what is the most appropriate way to extrapolate such cancer risk estimates to still lower doses? Given that it is supported by experimentally grounded, quantifiable, biophysical arguments, a linear extrapolation of cancer risks from intermediate to very low doses currently appears to be the most appropriate methodology. This linearity assumption is not necessarily the most conservative approach, and it is likely that it will result in an underestimate of some radiation-induced cancer risks and an overestimate of others
Wait what?

My understanding of dose-response curves of various toxic and dangerous "substances" is that they are linear or sigmoidal - in other words the initial decrease has a small slope, the slope increases as the dose gets larger, until it hits an inflection point, and then the slope begins to fall off and plateau. Further, my understanding of all of the available evidence is that at larger dose rates, the dose-response curve for radiation harms is linear or sigmoidal.

A charitable reading is merely that a linear model underestimates risk of a sigmoidal dose-response curve past the inflection point. I do not believe that is the intent of the bolded section of the paper in the above quote. I do not believe that this is compatible with a reasonable-person reading with context.

For what the paper says in bold to be true, that means we would need to see some rather novel shape in the dose-response curve for radiation around zero. I need to follow up if any other substance has such a strange dose-response curve. I suspect not. IMHO, this paper is already sounding like alchemy or homeopathy.

Post Reply