Dose is meaningless
... emerging consensus
[This page from November 2006 is now updated with this new link to extracts from ICRP Publication 103 (the 2007 Recommendations) but its content otherwise remains unchanged. At the foot there is recent material on ICRP's position.]
The 2005 Recommendations of the International Commission on Radiological Protection: Draft for Consultation were published in late 2004. The final version has not been published at the date of writing (early November 2006) and ICRP tells us publication has in fact been set back by the IRSN's report on the European Committee on Radiation Risk (ECRR).
Consultation on a second draft closed in the summer. Our responses can be seen on the ICRP site
The ICRP 2004 draft contains many statements revealing the incomplete state of knowledge of radiation risk. Many of them have been watered down in the 2006 draft or have disappeared altogether.
Here we reproduce extracts from the 2004 draft which confirm the validity of our long-standing concerns about heterogeneity of energy distribution. The ICRP's response to heterogeneity is to employ assumptions. Most are individually questionable and when taken together, as they must be, they are simply not acceptable as a system of radiation protection. The upshot is that "dose" is an effectively meaningless term yet the industry's regulators have no other terms with which to assess and quantify risks. Reassurances about "trivial doses" are revealed as empty."3.2. Summary of health effects caused by ionising radiationIt seems perverse that having admitted so many flaws in the concept of absorbed dose ICRP simply continues to use it.
(37) The relationship between radiation exposures and health effects is complex. The physical processes linking exposure and doses in human tissues involve energy transport at the molecular level. The biological links between this energy deposition and the resulting health effects involve molecular changes in cells. In Publication 60 (ICRP, 1991) , the Commission recognised that the gross (macroscopic) quantities used in radiological protection omitted consideration of the discontinuous nature of the physical and biological processes of ionisation. However, it concluded that their use was justified empirically by the observation that the gross quantities (with adjustments for different types of radiation) correlate reasonably well with the resulting biological effects. It further recognised that more use might eventually be made of other quantities based on the statistical distribution of events in a small volume of material, corresponding to the dimensions of biological entities such as the nucleus of the cell or its DNA. Meanwhile, for practical reasons, the Commission continues to use the macroscopic quantities.
3.3. Absorbed dose in radiological protection
(41) A particular feature of ionising radiations is their discontinuous interaction with matter. The related probabilistic nature of energy depositions results in distributions of imparted energy on a cellular and molecular level that are very heterogeneous at low doses. […]
(42) […] At the low doses generally of concern in radiological protection, the fluctuation of energy imparted can be substantial between individual cells and within a single hit cell. This is the case particularly for densely ionising radiations such as alpha-particles and charged particles from neutron interactions.
(44) Absorbed dose is defined based on the expectation value of the stochastic quantity e, energy imparted, and therefore does not consider the random fluctuation of the interaction events. It is defined at any point in matter and, in principle, is a measurable quantity, i.e. it can be determined experimentally and by computation. The definition of absorbed dose has the scientific rigour required for a fundamental quantity. It takes implicitly account of the radiation field as well as of all of its interactions inside and outside the specified volume. It does not, however, consider the atomic structure of matter and the stochastic nature of the interactions.
(46) For densely ionising radiation (charged particles from neutrons and alpha-particles) and low doses of low LET radiation, the frequency of events in most cells is zero, in a few it is one and extremely exceptionally more than one. The value of energy imparted in most individual cells is then zero but in the hit cells it will exceed the mean value by orders of magnitude. These large differences in the energy deposition distribution in microscopic regions for different types (and energies) of radiation have been related to observed differences in biological effectiveness or radiation quality.
(47) In the definition of radiological protection quantities no attempts are made to specify these stochastic distributions at a microscopic level. Even the quality factor used in the definition of operational quantities is dependent on LET only which also is a non stochastic quantity. Instead a pragmatic and empirical approach has been adopted to take account of radiation quality differences - and therefore implicitly also of the differences in distributions of energy imparted in microscopic regions - by defining radiation weighting factors. The selection of these factors is mainly a judgement based on the results of radiobiological experiments.
3.3.2. Radiological protection quantities: Averaging of dose
(48) While absorbed dose is defined to give a specific value (averaged in time) at any point in matter, averaging of doses over larger tissue volumes is often performed when using the quantity absorbed dose in practical applications, as in radiological protection. It is especially assumed for stochastic effects at low doses that such a mean value can be correlated with the risk of a detriment to this tissue with sufficient accuracy. The averaging of absorbed dose and the summing of mean doses in different organs and tissues of the human body, as given in the definition of all the protection quantities, is only possible under the assumption of a linear dose-response relationship with no threshold (LNT). All protection quantities rely on these hypotheses.
(49) Protection quantities are based on the averaging of absorbed dose over the volume of a specified organ or tissue. The extent to which the average absorbed dose in an organ is representative of the absorbed dose in all regions of the organ depends on a number of factors. For external radiation exposure, this depends on the degree of penetration of the radiation incident on the body. For penetrating radiation (photons, neutrons) , the absorbed dose distribution within a specified organ may be sufficiently homogeneous and thus the average absorbed dose is a meaningful measure of the absorbed dose throughout the organ or tissue. For radiation with low penetration or limited range (low-energy photons, charged particles) as well as for widely distributed organs (e.g. bone marrow) exposed to non-uniform radiation flux, the absorbed dose distribution within the specified organ may be very heterogeneous.
(50) For radiations emitted by radionuclides residing within the organ or tissue, so-called internal emitters, the absorbed dose distribution in the organ depends on the penetration and range of the radiations and the homogeneity of the activity distribution within the organs or tissues. The absorbed dose distribution for radionuclides emitting alpha particles, soft beta particles, low-energy photons, and Auger electrons may be highly heterogeneous. This heterogeneity is especially significant if radionuclides emitting low-range radiation are deposited in particular parts of organs or tissues, e.g. plutonium on bone surface or radon daughters in bronchial mucosa and epithelia. In such situations the organ-averaged absorbed dose may not be a good dose quantity for estimating the stochastic damage. The applicability of the concept of average organ dose and effective dose may, therefore, need to be examined critically in such cases and sometimes empirical and pragmatic procedures must be applied. ICRP has developed dosimetric models for the lungs, the gastrointestinal tract and the skeleton that take account of the distribution of radionuclides and the location of sensitive cells in the calculation of average absorbed dose to these tissues.
3.3.3. Radiation weighted dose and effective dose
(51) The definition of the protection quantities is based on the mean absorbed dose …
The 1991 assertion (see ICRP para. 37 above) that the use of macroscopic quantities is justified empirically is not acceptable. In the ensuing 15 years developments in cell biology and epidemiology, particularly following Chernobyl, have rendered it unsafe. The European Committee on Radiation Risk (ECRR) has recently developed weighting factors to compensate for some of the shortcomings of the ICRP approach. IRSN's 2005 report on ECRR states:"Various questions raised by the ECRR are quite pertinent and led IRSN to analyze this document with a pluralistic approach.IRSN's report concludes:
a. Besides natural and medical exposures, populations are basically undergoing low dose and low dose rate prolonged internal exposures. But the possible health consequences under such exposure conditions are ill-known. Failing statistically significant observations, the health consequences of low dose exposures are extrapolated from data concerning exposures that involve higher dose rates and doses. Also, few epidemiologic data could be analyzed for assessing inner exposure effects. The risks were thus assessed from health consequences observed after external exposure, considering that effects were identical, whether the exposure source is located outside or inside the human body. However, the intensity, or even the type of effects might be different.
b. The pertinence of dosimetric values used for quantifying doses may be questioned. Indeed, the factors applied for risk management values are basically relying on the results from the Hiroshima and Nagasaki survivors' monitoring. It is thus not ensured that the numerical values of these factors translate the actual risk, regardless of exposure conditions, and especially after low dose internal exposure.
c. Furthermore, since the preparation of the ICRP 60 publication, improvements in radiobiology and radiopathology, or even in general biology, might finally impair the radiation cell and tissue response model applied to justify radioprotection recommendations. It was thus justified to contemplate the impact of such recent observations on the assessment of risk induced by an exposure to ionizing radiation.""The phenomena concerning internal contamination by radionuclides are complex because they involve numerous physico-chemical, biochemical and physiological mechanisms, still ill-known and thus difficult to model. Due to this complexity, the behaviour of radionuclides in the organism is often ill described and it is difficult to accurately define a relationship between the dose delivered by radionuclides and the observed consequences on health. This led the radioprotection specialists to mostly use the dose/risk relationships derived from the study of the Hiroshima/Nagasaki survivors, exposed in conditions very different from those met in the cases of internal contaminations.and
This fact raises numerous questions, which should be considered with caution because a wide part of the public exposure in some areas of the world is due to chronic internal contaminations and very few data concern these situations.
[…] the questions raised by the ECRR are fully acceptable, … ""… we do not possess, in the current state of knowledge, the elements required to improve the existing radioprotection system."We realise that we are inviting the rejoinder that IRSN also says:[however] "the fact is that the [ECRR's] arguments stated to justify this doctrine modification are not convincing, as the demonstration as a whole does not meet the criteria of a strict and consistent scientific approach."and"the existing radioprotection system corresponds to the best tool being available at present for protecting human from the deleterious effects of ionizing radiations."and"… a significant improvement of the radioprotection system in the field of internal contamination [can be] conceivable only by development of studies and research. "See this link for ECRR's response to various points made by IRSN, and for the IRSN report itself.
IRSN's statements are a bizarre double standard; they have agreed with ECRR's criticisms of the ICRP system, which on that basis can itself be described as "not meet[ing] the criteria of a strict and consistent scientific approach" (as IRSN demands of ECRR). IRSN's subsequent call for more research may be only what is expected of scientists, but such research would take years. Policy makers and stakeholders engaged in decommissioning have to make decisions now.
CERRIE: DOSE IS “MEANINGLESS”… There are important concerns with respect to the heterogeneity of dose delivery within tissues and cells from short-range charged particle emissions, the extent to which current models adequately represent such interactions with biological targets, and the specification of target cells at risk. Indeed, the actual concepts of absorbed dose become questionable, and sometimes meaningless, when considering interactions at the cellular and molecular levels.
from CERRIE (Government's Committee Examining Radiation Risks of Internal Emitters) Majority Report Chapter 2 Risks from Internal Emitters Part 2 paragraph 11. See www.cerrie.org for full report.
See this site for the Minority Report
And the Department of Health's Radiation Protection Research Strategy July 2006 - could be LLRC's shoppping list.
ICRP throws in the towel
At a meeting in Stockholm, 22 April 2009, Dr Jack Valentin, Scientific Secretary Emeritus of the ICRP admitted that ICRP's risk model could not be applied to post-accident exposures because the uncertainties were two orders of magnitude. (see transcript)
The next day, Deputy Director of Strålsäkerhetsmyndigheten, Carl-Magnus Larsson also said the ICRP model could not be used to predict the health consequences of accidents. He added that for elements like Strontium and Uranium which bind to DNA national authorities would have the responsibility to assess the risks. Another SRM member said that the Secondary Photoelectron Effect was well recognised, also that in 1977 the ICRP had considered a weighting factor ”n” for elements which bind to DNA but had not implemented it.
If you are seeing this page full screen (i.e. without a navigation bar on the left) you can't see how the rest of the site is organised.
This Home page link takes you to the index page, which has links to all the topics we discuss on the site [only use it if this page is full screen]