Register for Updates | Search | Contacts | Site Map | Member Login

ICRP: Free the Annals!

View Comment

Submitted by Celso Osimani, AIRP = Italian Radiation Protection Association
   Commenting on behalf of the organisation
Document 2005 ICRP Recommendation
 
ICRP has made known the 2005 Draft of its new Recommendations and has encouraged radiation protection Associations and experts to communicate their views to the Commission.

In the context of the revision of the 1990 ICRP Recommendations, the Italian Radiation Protection Association (AIRP) has set up a Working Group with the remit to review the 2005 Draft of the new ICRP Recommendations and to submit comments.

AIRP offers the following views and comments of the Working Group on the 2005 Draft with a view to contributing to the Commission's efforts in ensuring that the future ICRP Recommendations are well understood and applied.

GENERAL COMMENTS

The AIRP Working Group welcomes the Commission's openness in making the Draft of its 2005 Recommendations available for comments by the Associations of professional radiation protection experts, thus initiating an ongoing world wide debate.

The AIRP Working Group also finds commendable that the philosophy and ideas in the 2005 Draft are, indeed, a pragmatic and positive evolution in respect of the 1990 Recommendations that acknowledges the need of stability in regulatory systems.

Radiation protection principles

The Working Group notes that a formal enunciation of radiation protection principles is missing in the Draft and should be provided in the final Recommendations, like in Publications 26 and 60.
In this context ICRP says, rightly, that justification is the province of governmental authorities and radiation protection considerations are only one of the inputs to decisions on justification.

Nevertheless, the Working Group believes that a formal statement of the justification principle is necessary in the new Recommendations, if nothing else as guidance for a correct expression of radiation protection inputs to decision makers.

In this regard the Working Group feels that the role of radiation protection in the justification process should be confirmed by ICRP as regards the assessment of the health aspects in the balancing of detriment and benefit. Accordingly, it is suggested that the Commission should also give guidance on the criteria and methods to apply in order to give correct radiation protection inputs to the authorities responsible for decisions on justification.
For instance, it is up to governmental authorities to decide whether mitigation measures should be applied in an existing controllable exposure situation, e.g. in the aftermath of an emergency, and the decision should be made taking into account all aspects in the prevailing circumstances, radiation protection considerations being obviously included.

In the context of radiation protection principles, the Working Group expresses disagreement with the ICRP recommended use of the Best Available Technology principle for the reasons set forth in the comment on the relevant paragraph of the 2005 Draft.

Scientific bases of the Recommendations

ICRP has announced, e.g. in paragraphs (93), (96), (173), (176) and (203), its intent to publish further documents in which, inter alia, the scientific bases and guidance on the application of the new Recommendations will be expounded.

Indeed, the need for clearly setting forth the scientific bases and the rationale underlying the Commission's choices is particularly felt for those aspects of the system as crucial as the estimates of radiation risk and the related radiation quantities, factors and coefficients (i.e. DDREF, nominal risk coefficients, tissue weighting factors, wT's, and radiation weighting factors, wR's). It goes without saying that this holds above all where significant changes have taken place in today's ICRP thinking in respect of past Publications.

The Working Group is aware that a great deal of work is required of the Commission for such a wealth of information to be published before the new Recommendations.

Nevertheless, the Working Group believes that ICRP ought to strive to ensure that, as a minimum, the scientific bases underlying the new Recommendations and the rationale for changes in scientific approaches should be made known before the Recommendations, with the relevant draft documents being proposed for discussion. In this context, it will be laudable if a new Draft of the Recommendations is also made available for comments before its adoption by the Commission. It is felt that continuing the openness policy will be of great help in ensuring that ICRP thinking is well understood and widely accepted.

LNT Hypothesis and related concepts

Throughout the Draft the Commission emphasises that the Linear No Threshold hypothesis and the whole array of radiation protection quantities, coefficients and factors, like e.g. DDREF, wR and wT, apply from a few mSv to some tens of mSv (i.e., in the low dose and dose rate range).

The Working Group believes that the Commission should clearly indicate what it intends for the dose range and the dose rate, as well as the type of radiations for DDREF, in which linearity and related concepts and quantities are meant to apply.

In this context, the Working Group notes that in paragraphs (36) and (54) in the 2005 Draft the Commission warns against protection quantities and nominal risks coefficients being used for assessments of a retrospective nature. As a matter of fact, ICRP emphasises that, for the reasons stated with clarity in the Draft, such quantities and coefficients are meant only to be utilised prospectively, i.e. for radiation protection purposes.

The Working Group considers in this respect that, on the one hand, an ideal protection system would certainly be the one availing itself of the same array of quantities and methods both for prospective and retrospective assessments of radiation risk; on the other hand, it is to be recognised that expert judgment is called for in defining radiation protection quantities since the present state of knowledge on low dose radiation effects does not allow both for utmost scientific rigour and a practicable and efficient protection system.

The Working Group points out, however, that the Commission is seemingly not too steadfast in heeding its own advice on the scope and use of radiation protection quantities, at least judging from the rather obscure paragraphs (88), (239), (240) in the 2005 Draft, for which see the specific comments; in this respect, see the comment on paragraph (202) as well.

Controllable exposures

The Commission intends its future Recommendations to apply to controllable sources in normal, emergency and existing controllable exposure situations. The framework is clear enough, yet some aspects should be addressed in detail, notably as regards the relationship between practices and normal situations since the Commission seems to use the two terms interchangeably.

The Working Group notes in this context that the Commission has retained the practice concept but has avoided mentioning intervention. The Working Group feels that intervention is a very useful concept to which regulatory authorities and radiation protection professionals have become accustomed, and suggests that there would be merit in retaining it, if nothing else for reasons of clarity and conciseness, thus e.g. avoiding the use of long circumlocutions, such as 'taking action in an existing controllable exposure situation or in an emergency'.

Dose constraints

It is the Working Group's understanding that the Commission wishes to present the values of dose constraints in Chapter 6 of its new Recommendations as having been chosen with reference to the world wide average of background radiation exposures, accordingly avoiding the need to 'justify' its choices in terms of risks that the society at large holds as tolerable.

However, there is room for doubts on the bases for the choice of new constraints (individual risk or average value of background) considering that some of past ICRP Publications (77, 81 and 82) avowedly continue to represent the Commission's thinking (paragraph 29 of the Draft). This aspect should be clarified.

The Working Group feels that, in any case, a discussion of recommended dose constraints in terms of risk is unavoidable for both artificial and primordial radionuclides, and the Commission should comment on the risks entailed by exposures incurred when applying the recommended values of dose constraints.

An aspect that ought to be expanded upon in the new Recommendations is the relationship and related guidance between dose constraints and dose limits, the latter being maintained in the forthcoming ICRP Recommendations since the Commission has acknowledged, rightly, the need of stability in regulatory systems.

Other aspects on which the Commission should give further guidance are the use of the 'single source' and 'critical group' concepts, perhaps by way of examples; simple as the concepts may deceptively seem, doubts are surely bound to arise in their application in connection with the use of the new constraints.

New names for radiation effects and quantities

An aspect that causes considerable doubts in the Working Group is the adoption of new names for radiation effects and quantities. It is felt that names of well known quantities and radiation effects ought not to be changed unless strictly necessary, in that such changes are likely to cause confusion; furthermore, some of them are downright questionable, as in the case of 'deterministic effects' being renamed to 'tissue reactions'. The new name is also felt to be of debatable merit since cancer resulting from radiation exposure might be considered a tissue reaction.

In this respect the Working Group notes that deterministic effects will be renamed for the third time since the 1977 Recommendations if the Commission goes through with the intended change in name, and that there is no perceivable reason in the Draft for such a renaming.

The Working Group feels that people will be brought to think that the so called 'tissue reactions' are something entirely different from 'non stochastic effects' or 'deterministic effects' of previous Recommendations, which is not, at least judging from the Draft. In this respect it must be pointed out that no radiation protection expert is supposedly so naïve as to think that thresholds for deterministic effects hold one and the same value for each individual exposed.

The same argument applies to the Commission's renaming 'equivalent dose' to 'radiation weighted dose' because of possible misunderstanding with the term 'dose equivalent' used in operational quantities and/or difficulties in translation to some other language than English.

As a matter of fact, the Working Group feels that renaming equivalent dose is likely to cause confusion, in that a reader will think that 'radiation weighted dose' is a conceptually new quantity with respect to equivalent dose, which is not.

Protection of non human species

The Working Group shares the view in the 2005 Draft that "it is probably true that the human habitat has been afforded a fairly high level of protection through the application of the current system of protection.", in this respect the Commission also acknowledges, inter alia, that "there are now also other demands upon regulators, in particular the need to comply with the requirements of legislation directly aimed at the protection of wildlife and natural habitats".

Indeed, the Working Group is well aware of the reasons why ICRP has decided to address the protection of non human species and believes that, all things considered, this is an unavoidable development.

In this respect the Working Group commends the Commission's effort to develop the scientific bases for a common approach to the protection of human beings and of non human species.

Nevertheless, the protection of non human species is one of several instances in the 2005 Draft where the Commission indicates an intent to issue recommendations rather than formulating them.

In view of that, the Working Group suggests that ICRP should limit Chapter 11 to issue recommendations proper and leave Annex B out.

Text of the Recommendations

The Working Group realises that the 2005 ICRP document is still in the drafting stage, and notes that there is certainly the need of refining touches.

The language used should be made homogeneous throughout the document and highly technical aspects, e.g. the discussion on the derivation of wT's, should be made much clearer than they are, beside the fact noted above that they need to be expanded upon.

There are minor instances of inconsistencies to eliminate, e.g. in paragraph (90) there is an overlined symbol, DT, for the average absorbed dose in an organ or tissue never used before or after; in paragraph (A30) the lethality fraction is termed 'qT', while it is given as 'q' in the footnote on page 72.

Above all, the Working Group believes that ICRP ought to go to great lengths in order to make the text of its Recommendations clear and to avoid being ambiguous or vague; e.g., expressions like "high accidental doses" in paragraph (176) should not be used in the Recommendations.

In order to attain the objectives above and to make the new Recommendations easy to understand and to use the
Working Group suggests that:
- the document should be revised with a view to ensuring clarity as well as consistency in the terms used;
- a Glossary of terms should be provided so as to render the meaning of concepts consistent and unambiguous;
- an Index should be compiled for ease of finding.

Final Consideration

Scientific excellence, clarity and consistency are expected of all ICRP Documents, particularly as regards ICRP Recommendations, which are milestones in the theory and practice of radiation protection all over the world.
The Working Group believes that the Commission should avoid haste and take all the time necessary to finalise the new Recommendations in order to reach those objectives.

SPECIFIC COMMENTS

Paragraph 13

"(13) The primary aim of the Commission is to contribute to the establishment and application of an appropriate level of protection for the human population and, where necessary, for other species without unduly limiting the desirable human actions and lifestyles that give rise to, or increase, radiation exposures." (Bold is the AIRP Working Group's)

The Working Group notes that in the paragraph above the Commission does not mention as its primary aim an appropriate level of protection for individual members of the human population; in this respect it is suggested that providing an adequate level of protection for individual human beings should be a primary aim of the radiation protection system.

Section 2.3 Exclusion and authorisation of exposures

The philosophy set forth in this Section is certainly both rational and pragmatic and is unreservedly commended; none the less, in the following paragraphs (24) (28) there are issues that it is felt should be further addressed by the Commission.

Footnote 1 in Paragraph 24

"1 ... For example, some radionuclides that are primordial and therefore considered ‘natural’ can be produced artificially. Others that are produced by humans and therefore considered ‘artificial’ are in fact also produced in nature by incoming solar neutrons or natural fission processes such as that at Oklo, Africa."

The Working Group believes that the adjective 'solar' should be cancelled in the sentence in the footnote.

Paragraph 28

"(28) The practical application of the concept requires derivation of exemption levels in terms of activity concentration." (Bold is the AIRP Working Group's)

The Working Group points out that both IAEA Basic Safety Standards of 1995 and Directive 96/29/Euratom, which lays down the Basic Safety Standards for the European Union, provide for exemption in terms of two sets of threshold values for each radionuclide to be exempt: activity and activity concentration, with values for each radionuclide being identical in the two regulatory requirements (IAEA ad EU Directive).

Paragraphs 38, 48, 101, 117, 200 The LNT issue

"(38) Radiological protection in the low dose range is primarily concerned with protection against radiation-induced cancer and hereditary disease. These diseases are termed stochastic effects, as they are probabilistic in nature and are believed to have their origins in damage in single cells. For protection purposes, it is assumed that these effects increase with increasing radiation dose, with no threshold, and that any increment of exposure above the natural background produces a linear increment of risk."

"(48) ... The averaging of absorbed dose and the summing of mean doses in different organs and tissues ... is only possible under the assumption of a linear dose-response relationship with no threshold (LNT)."

"(101) Although there are recognised exceptions, for the purposes of radiological protection the Commission judges that the weight of evidence on fundamental cellular processes supports the view that in the low dose range up to a few tens of mSv, it is scientifically reasonable to assume that in general and for practical purposes cancer risk will rise in direct proportion to absorbed dose in organs and tissues. This view accords with that given by UNSCEAR (2000)"

"(117) ... However, even in the absence of a true dose-threshold, any effects on IQ following in utero doses of a few tens of mGy would be undetectable and therefore of no practical significance."

"(200) ... For making decisions, a large dose to a small number of people is not equivalent to a small dose to many people, even if the two cases correspond to numerically equal collective doses." (Bold in the paragraphs above is the AIRP Working Group's)

The Working Group is aware that the Commission plans to publish an ad hoc document, now made available on the web for comments, on the complex and controversial low dose issue in which, inter alia, LNT is discussed.
None the less, the Working Group believes that issues of principle should be addressed in the new Recommendations clearly and consistently. Accordingly, it is felt that the Commission should clarify its thinking on LNT and review the wording in the paragraphs quoted, since they give rise to the doubts below.
- Is LNT merely a computational tool that radiation protection simply could not do without if a viable system is to be recommended and acted upon, as paragraph 48 seemingly implies, or is LNT the most prudent scientific hypothesis in the present state of scientific knowledge on low dose radiation effects, as implied in paragraphs 38 and, partially, 101?
- The statement in bold in paragraph 117 is felt to be somewhat inconsistent with other statements, e.g. the conclusions on LNT in Annexe G of the UNSCEAR 2000 Report and in the recently published Draft on Low Dose Extrapolation (pages 12, 40): if existence of a threshold is excluded and linearity is assumed as the most scientifically reasonable and/or prudent hypothesis in the present circumstances, then it is hardly arguable that undetectable effects are of no practical significance. As a matter of fact, were this line of reasoning (what is not detectable has no significance) followed through to its ultimate implications, the low dose problem would vanish from sight outright.
- The statement quoted in paragraph (38) ("any increment of exposure above the natural background produces a linear increment of risk") should be worded more carefully; as it stands, it seemingly implies that background radiation doses cause no risk. The same remark holds for the sentence in paragraph (200) which might be understood as a departure from the LNT hypothesis.

Paragraph 54

"(54) It must be stressed that effective dose is intended for use as a principal protection quantity for establishment of radiation protection guidance. ... Its main use is to enable external and internal irradiation to be added as a means to demonstrate compliance with the Commission’s quantitative restrictions on dose, which are expressed in effective dose." (Bold is the AIRP Working Group's)

There is an important use of effective dose which seems to have been overlooked: i.e. to enable non-uniform internal and/or external exposures in the various tissues or organs to be added and thus give a 'global measure' of stochastic radiation risk.

Paragraphs 55, 84

"(55) Effective dose... is in principle as well as in practice a non-measurable quantity."

"(84) ... The protection quantity adopted by ICRP for the control of stochastic effects is the effective dose. This quantity is by its definition related to doses in the human body and generally is not measurable." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group points out that in the two paragraphs above there is a seeming inconsistency in the statements on the measurability of effective dose. Since effective dose is a quantity that can certainly be assessed but obviously cannot be measured directly, it is felt that in paragraph (85) ICRP should delete the adverb "generally" and use the same terms as in paragraph (55).

Section 3.4.1 Radiation weighting factors

The Working Group points out that there are issues in this Section, notably in paragraphs (61), (73) and (74), that need clarification by the Commission.

Paragraph 61

"(61) The evaluation of wR values is based mainly on RBE data from in vivo investigations with animals." (Bold is the AIRP Working Group's)

This statement is of questionable accuracy for the reasons given below.
- As concerns neutrons, paragraph (70) of the Draft states that wR values for neutrons are essentially based on the Q(L) relationship ("The calculation of the energy dependence of the radiation weighting can be based on the Q(L) relationship defined in Publication 60 (ICRP, 1991a) and the calculation of a human body averaged mean quality factor qE").
- According to paragraph (75), the wR value for protons seems to be based mainly on calculations.
- Paragraph (76) states that the wR value for alpha particles is based on calculations using the Q(L) function.
- According to paragraph (78), the wR values for heavy ions seem to have been chosen from in vitro experiments.
The Working Group feels that the statement in paragraph (61) should be re-formulated; furthermore, it would be very appreciated having a table with a summary of the RBE estimates from the cited experiments with animals.

Paragraphs 73, 74

"(73) ... If, however, the strong correlation between qE and wR as defined in Publication 92 would be applied, in the energy range between 5 and 150 MeV this would result in an increase of wR for neutrons between 22% and 39% relative to the data of the continuous function as defined in Publication 60. Such an increase is not supported by any experimental data."

"(74) It is therefore recommended to stay with the continuous function of Publication 60 at neutron energies equal and above 1 MeV and to change this function at low energies only." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group notes that the wR values for neutrons of energy above 1 MeV (Figures S1 and 1) in the 2005 Draft are different from those put forward both in Publication 92 (Figure 4.4) and in a paper by Kellerer et al. (Radiation Protection Dosimetry, 109, 181-188, 2004), in that the current continuous function for wR has been retained by ICRP for neutrons with energy above 1 MeV.

Thus, the linkage between wR and an LET dependent weighting factor proposed in Publication 92 has been ignored in the ICRP Draft. It is worth noting in this context that the linkage proposed was the same in Publication 92 and in the above quoted paper by Kellerer et al., while the neutron wR values above 1 MeV are different.
In this respect, given the statements on radiobiological findings in paragraphs (267) and (272) of Publication 92 and in the above quoted paper, it is questionable that "(73) .... the strong correlation between qE and wR as defined in Publication 92" can be accepted below 1 MeV and not above.

The Working Group feels that the ICRP motivation for change in this respect should be supported by ad hoc references since the last sentence in paragraph (73) of the 2005 Draft is seemingly insufficient to justify such a change with respect to the wR(E) function proposed in ICRP Publication 92, as well as in the above paper by Kellerer et al.

In this context it must also be pointed out that the value wR = 5 for neutron energies above 100 MeV is not appropriate (see Radiation Protection Dosimetry, 80, 371-378, 1998).
Paragraphs 84, 85

"(84) The definitions of the operational quantities take account of the common situation in which the individual dose assessment is performed with dosemeters worn on the body. The personal dose equivalent is, therefore, defined by the dose at a specific depth in the body below the point where the dosemeter is worn. The protection quantity adopted by ICRP for the control of stochastic effects is the effective dose."

"(85) ... for external sources ... in most practical situations the values of the operational dose quantities provide an assessment of effective dose that is sufficiently accurate for radiological protection applications."
The Working Group points out that the recommended dual system of protection and operational quantities is not applicable for high energy radiation. For such kind of radiation, in fact, the operational quantities do not provide an assessment of effective dose.

The Working Group feels that there is a lack of explanation and of significant discussion of this problem that should be addressed by the Commission in an ad hoc Document.

Paragraph 88

"(88) Although dose records are for individuals the dose coefficients on which they are based are derived for reference individuals. If doses approach or exceed the dose constraints, then investigations may need to be undertaken to address workplace and individual specific characteristics in the dose assessment." (Bold is the AIRP Working Group's)

The statement above needs clarification by the Commission. Given that the Commission is proposing, inter alia, a maximum dose constraint of 20 mSv/y for workers, if one takes the statement at its face value this would seem to imply that an individual specific assessment has to be made, for instance in terms of RBE and absorbed dose distribution in the body, for workers' doses e.g. of 18 mSv or even less.

Apart from this requirement being next to unthinkable in practice as there are applications where such relatively high doses are not unheard of (e.g. medical interventional procedures or industrial radiography in cases of fairly frequently occurring minor mishaps), it must be pointed out that a similar statement has obvious merits for those cases where exposures approach or exceed the 100 mSv constraint recommended by the Commission in case of emergencies. In fact, a 100 mSv exposure approaches the border between low and high doses where the Linear No Threshold assumption and the concepts of effective and equivalent dose no longer hold.

Paragraphs 91, 92

"(91) In Publication 60, the operational quantity Annual Limit on Intake (ALI) was defined"

"(92) ... to define as a further operational quantity the Derived Air Concentration (DACE)." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group points out that, contrary to what the Commission states in the two paragraphs above, neither ALI's nor DAC's are operational quantities. Indeed, ALI's and DAC's are the secondary and derived limits, respectively, for the operational quantities 'intake' and 'air concentration'.

Paragraphs 105, A32, A33

The Commission does not indicate for what type of radiation and for what dose and dose rate range it intends DDREF to apply, e.g. for low LET radiation, absorbed doses below 0.2 Gy and absorbed dose rates below 0.1 mGy/min (i.e. the low dose and dose rate range according to UNSCEAR 2000, Annex G, paragraph 29), nor does the new ICRP Draft on Low Dose Extrapolation shed much light on this issue either.

The Working Group believes that these important concepts and the rationale for the choice of a value of 2 for DDREF should be discussed in detail in the Recommendations.

Paragraph 111

"(111) ... the Commission proposes nominal probability coefficients for lethality adjusted cancer risk as 6.2 10-2 Sv-1 for the whole population and 4.8 10-2 Sv-1 for adult workers aged 20-64." (Bold is the AIRP Working Group's)

It is not clear why nominal probability coefficients for lethality adjusted cancer risk are given with the values above for the whole population and workers. Taking the sum of the unrounded lethality adjusted cancer risk factors for single tissues or organs from the data in Table A1 (q.v.), the lethality adjusted cancer risk would seem respectively to be 6,35 10-2 Sv-1 for the public and 4,89 10-2 Sv-1 for workers, which values do not match with the numbers in the paragraph above.

As a matter of fact, there does not seem to be a traceable criterion for rounding; please see also the comment on paragraph (112).

The Working Group also points out that ICRP Publication 60 considered a working lifetime to be 47 years, and wonders whether the age range of adult workers should be 18 to 65 instead of 20 to 64. To what extent, if any, does the indicated age range for workers (20 to 64) influence risk estimates in Annex A?

Paragraph 112

"(112) ... the nominal probability coefficient for fatal cancer in the whole population that may be projected from the cancer incidence data of Table A1.a of Annex A is 4.4% per Sv ... " (Bold is the AIRP Working Group's)
It is not clear why the nominal probability coefficient for fatal cancer is given as 4.4% Sv-1 for a whole population since taking the sum of the unrounded products of the nominal risk coefficients for cancer incidence times the respective lethality fractions found in the a) section (Whole population) of Table A1 of Annex A the unrounded value would seem to be 4.57% Sv-1. Please also see the comments about Table A.1. of Annexe A of the Draft.

Paragraph 113, Table 6

"(113) An additional point relating to the lethality adjusted cancer risk of Table 6 is that during the period that these recommendations are likely to apply, the survival rates for many cancers are expected to rise. In this respect, the nominal risk coefficient given will tend to be an over-estimate of risks in the future." (Bold is the AIRP Working Group's)

The Working Group feels that the statement above (estimates of nominal risk coefficients for lethality adjusted cancer risk being conservative because of expected improvements in survival rates in the future) has overtones of cynicism. The suggestion is that this paragraph should be left out, although it is thought to be correct.

Table 6: Nominal probability coefficients for stochastic effects (10-2 Sv-1)
Exposed population---Lethality adj.---Lethality adj.---Detriment----Detriment
----------------------------cancer risk----heritable eff:s--------------------Pub.60
Whole population-------6.2--------------0.2-----------------6.5------------7.3
Adult workers------------4.8--------------0.1-----------------4.9------------5.6

For reasons of clarity, it is suggested that the Table in paragraph (113) should be completed with a column giving the nominal probability coefficients for unadjusted fatal cancer in the whole population and in workers.
Moreover, it must be pointed out that the sum of lethality adjusted cancer risk (6.2) and lethality adjusted heritable effects (0.2) for the whole population does not equal the indicated detriment (6.5).

Paragraph 135, Figure 2

"(135) Figure 2 illustrates the differences in concept between individual dose constraints for protection from a single source in all situations and the use, in normal situations only, of individual-related dose limits."
The Working Group feels that the figure is of help in that it renders the difference between individual dose limits and constraints; yet, there are unclear aspects in the types of situations shown on the left side as regards dose limits for both workers and the public. The suggestion is that the left portion of Figure 2 should be revised and made clearer.

Paragraph 160

"(160) Doses above the natural background will entail an increasing need for action. Individual doses of several tens of millisieverts, whether they are received either singly or repeatedly, require that action be considered. Exposures that are within the natural background range are legitimate matters for concern, sometimes calling for significant action." (Bold is the AIRP Working Group's)

The rationale underlying the statement in capitals is not easily understandable and the Commission should clarify it since, inter alia, it is seemingly inconsistent with a statement in paragraph (38): "any increment of exposure above the natural background produces a linear increment of risk.". Please also see the next comment.

Paragraph 161, Paragraph 164, third bullet

"(161) Provided that the additional sources come from practices that have not been judged to be frivolous, the need for action should be low for doses less than about one hundredth of background dose."

"(164) The value of an effective dose of 0.01 mSv/year is the minimum constraint that should be considered for application in any situation. This value corresponds to a low need for action (Figure 3), giving rise to trivial risk to the exposed individuals." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group is fully aware that the statements above are in accordance with the Commission's approach to the setting of constraints, i.e. the Commission considers that optimisation should be carried out in order to determine an optimised value below the maximum value of constraints, thus, accordingly, even below this very low dose level.

Nevertheless, the statement gives rise to marked perplexity, to say the least: one hundredth of background dose corresponds to an effective dose of 10 µSv/y, i.e. the dose level that was judged to be below radiological concern and chosen as the basis to derive exemption levels by both IAEA and the European Union in their Basic Safety Standards.

It is legitimate to ask to what (lower) optimised dose level the Commission thinks values of exemption should correspond, given that one tenth of 10 µSv/y is an exceedingly small value.

Moreover, it is felt that in the last part of the 3rd bullet of paragraph (164) there appears to be a contradictio in adiecto: it is difficult to see how any need for action, however low, can be thought conceivable in presence of risks that are explicitly deemed trivial by the Commission.

The Working Group suggests that the Commission should indicate that there is no need for action in case of trivial risks such as those addressed in the paragraphs above.

Paragraph 163

"(163) Table 7 presents the Commission’s recommended maximum values of dose constraints. ... They should be considered as giving the upper restriction that is to be applied by the appropriate authorities to determine the most applicable constraints for the situation under consideration. The Commission expects that the resulting values normally will be lower than the maximum value recommended by the Commission, but probably not by as much as a factor of ten." (Bold is the AIRP Working Group's)

It is felt in this context that the Commission ought to make explicit mention of the use of optimisation when values
of dose constraints for a given situation are chosen.

Paragraphs 168, 211

"(168) It is necessary to deal separately with at least three types of exposed individual. These types can be called informed individuals, patients, and general individuals. They can, essentially, correspond to individuals whose exposures fall into the three classes of exposure defined in Chapter 5.3, i.e. occupational, medical and public."

"(211) These exposures of aircrew and couriers in the operation of commercial jet aircraft should be dealt with as occupational exposure in the general system of protection and thus of informed individuals." (Bold in both paragraphs is the AIRP Working Group's)

Apart from exposures of informed and consenting helpers and carers in paragraph (225), the Working Group notes that the Commission uses the term "informed individuals" in the two paragraphs above essentially to refer to workers.

The Working Group suggests that it should probably be clearer and simpler to refer to occupational exposures outright.

Paragraph 169

"(169) Workers in controlled areas of workplaces are nor strictly volunteers, ... thereby forming a group of informed individuals." (Bold is the AIRP Working Group's)
Besides controlled areas, international requirements such as IAEA and the European Union's Basic Safety Standards also provide for supervised areas.

The Working Group suggests that it should be better to speak in terms of occupational exposure.

Paragraph 173

"(173) ... the Commission is considering the use of age-averaged effective dose coefficients and age-averaged habit data for the individual in the case of continuing exposures of the public." (Bold is the AIRP Working Group's)
Given the wide variability of inhalation and ingestion dose coefficients for the six age classes currently recommended by ICRP and of related habits of the individuals concerned, it is difficult to see how such age averaged coefficients and habits might be established without being unduly conservative or otherwise unbalanced. However, the prospect looks promising in that it would somewhat simplify assessments.

One aspect unclear is the term 'continuing': is reference possibly being made to normal situations as opposite to acute exposures, e.g. in emergencies, or is ICRP thinking in entirely new terms that the Working Group fails to understand?

Paragraph 174

"(174) It has never been possible to reach simple formal definitions of a single source or of the total group of relevant sources. In the application of constraints, the term ‘single source’ should be used in a broad sense, such as the x ray equipment in a hospital, or the releases of radioactive materials from an installation. Most situations will give rise to a predominant source of exposure for any single individual, or critical group, making it possible to treat sources singly when considering actions." (Bold is the AIRP Working Group's)

The Working Group is aware of the difficulty to give a satisfying definition of single source, yet, it is felt that ICRP should give guidance on that because the single source concept plays a central role in the application of new constraints.

For instance, in case of radiation from x-ray equipment and releases of radioactive material from the same hospital, constraints should be applied to the sum of both sources of radiation or x ray equipment and releases should be considered separately? The sentence on the critical group concept playing a role in the definition of single source does not shed much light on this complex issue either.

Paragraph 185

"(185) The Commission has concluded that the existing limits on effective dose that it recommended in Publication 60 continue to provide an appropriate restriction on total regulated doses in normal situations. ... the Commission now emphasizes the use of constraints on single sources which are more restrictive than limits in normal situations;" (Bold is the AIRP Working Group's)

It is not clear how dose constraints can always be thought to be more restrictive than dose limits in normal situations since constraints apply to exposures from a single source while dose limits are intended to apply to exposures from all sources in normal situations.

There also is the obvious fact e.g. that, in accordance with the system of dose limits for workers recommended at present by ICRP, workers' exposures may well reach up to 50 mSv in a single year, thus exceeding the maximum value of the relevant dose constraint, which is 20 mSv/y.

The Working Group feels that the Commission should clarify this statement.

Paragraph 194

"(194) For exposures, the principle to be considered is whether these are as low as reasonably achievable. For the control of emissions to the environment, the ‘best available technology not entailing excessive costs’ principle may be used with due consideration to social and economic factors." (Bold is the AIRP Working Group's)
The part in bold is the most disquieting feature in the Draft of the new Recommendations; in fact, this constitutes a considerable departure from the well established principle of optimisation for the control of exposures, in particular as concerns radioactive releases.

Till now the use of the "Best Available Technology" principle for control of releases has never been recommended by the Commission, rightly it is felt, since the "Best Available Technology" principle has strong overtones of minimisation, as opposed to the well known concept of optimisation.

It might be argued that the Commission emphasises the need to keep due account of social and economic factors when applying the "Best Available Technology" principle but decades of experience in trying to get the message through that optimisation is a concept far apart from minimisation do engender scepticism on the viability of introducing such a new concept in radiation protection philosophy and practice.
Indeed, it will be something closely resembling a miracle if people pay attention to the finer points of the ICRP formulation of controlling releases by application of BAT.

What the Working Group does not understand in the least is the underlying rationale in introducing the "Best Available Technology" principle and departing from the well tried optimisation principle as concerns control of releases.

Now that at long last people had begun to see the significance of optimisation, the radiation protection community is regaled with the novelty of ICRP's recommending Best Available Technology; if the reason for this is to keep account of some country's practice, this might be accomplished by simply stating the fact that some countries think it best to apply BAT as regards control of radioactive releases.

One thing is felt to be sure: if ICRP does not reconsider on this, application of the Best Available Technology principle is going to stir up no end of discussions for decades in the radiation protection community and, what matters most, in the courts.

Section 7.2 Distribution of exposures in time and space

In the context of distribution of exposures in time and space, previously termed collective dose, the Working Group considers the basic approach used by the Commission to be sound. Let it be said, incidentally, that this is one case where a new name, e.g. 'dose matrix' or 'group dose', might be appropriate since the Commission is recommending a new approach in order to take account of such exposures.

The Working Group points out, however, that the issues in paragraphs (198) and (201) as well as (202) below need to be further addressed by the Commission.

Paragraph 198

"(198) ... However, the integral of low individual exposures over large populations, large geographic areas, and over large periods of time is generally not a useful tool for decision aiding because this may aggregate information excessively." (Bold is the AIRP Working Group's)

The Working Group concurs with the Commission's view that the doses referred to in the paragraph above are not a useful tool for decision aiding; however, the use of the adverb 'generally' makes the statement somewhat ambiguous.

The Working Group hopes that the Commission will give guidance unambiguously on whether low doses (e.g., 10 uSv/y) should be kept account of.

Paragraph 201

"(201) The Commission now recommends the maintenance of the distribution of individual doses related to a given source in components reflecting the characteristics of the exposed individuals and the time and space distributions of exposures, relevant for the decision making process considered.". (Bold is the AIRP Working Group's)

The Working Group points out that there is no guidance in the 2005 Draft on the methodology to apply when considering the time and space distribution of individual doses. These important and often controversial issues are left unresolved in the 2005 Draft and should be addressed by the Commission. As for the characteristics of the exposed individuals, please see the comment on paragraph (202) below.
Paragraph 202

"(202) ... Aspects to be considered when establishing the importance of each matrix element in the decision-making process may include:
... Age and gender dependent risks as modifiers to dose distributions ... " (Bold is the AIRP Working Group's)
Repeatedly throughout the Draft, and especially so in Chapter 3, the Commission has seen fit to emphasise that the radiation protection concepts and quantities (nominal risk coefficients, wR, wT, radiation weighted dose and effective dose) are obviously affected by large uncertainties and are the result of judgments made by ICRP and, as such, must only be used prospectively for controlling stochastic risks.

The Commission's emphasis in saying this is perhaps excessive in that by now every radiation expert should recite by rote the well known limitations inherent in the radiation protection quantities and concepts in everyday use.

This said, it is difficult to understand why the Commission indicates in the matrix above that age and gender dependent risks should specifically be considered as modifiers to dose distributions.
In the case where exposures are in the low dose range and, as such, well below thresholds for deterministic effects, it is not clear why age and gender should be considered as modifiers to dose distributions. Even given the well known approximations inherent in averaging absorbed dose in the organs and the judgments used in averaging over age and sex to obtain estimates for the definition of wR's, wT's, the Working Group has considerable difficulty in understanding the Commission's intent.

We already keep account of age when evaluating effective dose from intakes by the public and workers by making use of ICRP recommended committed effective dose coefficients, e(g), for activity intakes by inhalation and/or ingestion. What is further required by the statement above remains obscure and the Commission ought to clarify this important aspect.

Furthermore, the Working Group points out that there seems to be an inconsistency between the use of age and sex dependent risks suggested in paragraph (202) and the intent stated by ICRP to propose age averaged effective dose coefficients and age averaged habit data in paragraph (173).

Section 8.2 Natural radioactive substances in environmental materials

The Working Group welcomes the Commission's choice of recommending exclusion levels both for artificial and natural radionuclides in that it is a pragmatic development that will make for easier regulatory choices.
None the less, the Working Group feels for the reasons stated below that the ICRP considerations in the two paragraphs (209) and (210) need further thought by the Commission since the choice of U238 and Th232 numerical levels in Table 10 of the 2005 Draft needs to be further addressed.

Paragraphs 209, 210

"(209) The Commission proposes a set of exclusion values for the activity concentrations of natural radionuclides in materials. These levels were established from consideration of the distribution of concentrations of natural radionuclides in natural materials, representing a value towards the higher end of the generally observed range. In the UNSCEAR (2000) report, activity concentrations of the naturally occurring radionuclides in food range from less than 0.001 up to about 0.1 Bq g-1. The exception is shellfish where 210Po, in the decay series of 238U can have activity concentrations of the order of 1 Bq g-1. Exposures from environmental materials and intakes of food and water, at these activity concentrations, would lead to individual annual effective doses of no more than about 0.2 millisieverts, which does not in the Commission's opinion imply an unacceptable level of exposure."
The Working Group notes that Table 18 in Annex B of the 2000 UNSCEAR Report presents an estimate of annual age weighted committed effective dose of 0,144 mSv (with doses ranging from 0,110 mSv for adults to 0,260 for infants) from intakes of natural radionuclides in food and drinking water.

However, according to paragraph 76 of Annex B, the UNSCEAR evaluations were made using the consumption rates for food and water together with the reference values of concentrations given in Table 15 of the Annex.
In this respect the Working Group points out that in most cases the reference concentration values, used by UNSCEAR 2000 in its estimates, are several orders of magnitude lower than those (0.001 up to about 0.1 Bq g-1) cited by ICRP in the paragraph above.

Paragraph 210

"(210) The Commission notes the recent work undertaken by the IAEA in the production of its report DS161 in which the exclusion levels for the uranium and thorium series and for 40K have been agreed internationally. These activity concentrations are shown in Table 10 and are recommended by the Commission as the levels below which materials do not enter the scope of its recommendations." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group notes in this respect that the European Union issued guidance (European Commission, Radiation Protection 112, Radiological Protection Principles concerning the Natural Radioactivity of Building Materials, 1999) on the use of building materials containing natural radionuclides.

Using the model and the parameter values set forth in the EU RP112 publication cited above and assuming a room made of concrete having the U238, Th232 and K40 concentrations of Table 10 of the 2005 Draft, the Working Group points out that an individual would be likely to incur an annual effective dose of about 13,6 mSv in excess of background.

Furthermore, the Working Group points out that paragraph 5.1 of the IAEA guidance (now published as IAEA Safety Guide no. RS-G-1.7) referred to in the 2005 Draft sounds a note of warning in respect of the indiscriminate use of the exclusion concentration values of natural radionuclides, notably as regards building materials.

Paragraph 226

"(226) Also, medical exposures are incurred by those volunteering for research involving exposures to radiation and insurance companies may require individuals to receive medical exposures. In these cases again, the public constraints are not appropriate and national authorities should use higher values similar to those quoted in the paragraph above." (Bold is the AIRP Working Group's)

The Working Group points out that exposures required of individuals by insurance companies should also be subject to the justification process.

Paragraphs 239, 240

"(239) ... fragments may become very sparsely distributed in the environment, usually as ‘hot particles’. For these situations, the Commission continues to recommend the derivation of protection criteria based on the principles described in this section, namely of the unconditional probability that members of the public would develop fatal stochastic health effects attributable to the exposure situation.

(240) In these situations, such a probability should be assessed by combining the following probabilities: the probability of being exposed to the hot-particle residues; the probability of incorporating a hot particle into the body as a result of such exposure; the incurred average radiation weighted dose as a result of such incorporation; and, the probability of developing a fatal stochastic effect from that dose. These probabilities should be integrated over all the range of situations and possible doses. In establishing such criteria, consideration should be given to the possibility that localised tissue reactions may also occur as a result of the incorporation of hot particles." (Bold in both paragraphs is the AIRP Working Group's)

The Working Group notes that the Commission indicates the average radiation weighted dose as the proper quantity to use in order to evaluate the unconditional probability that members of the public would develop fatal stochastic effects; however, the Commission does not see fit to indicate effective dose as the quantity to use in order to make an estimate of the probability of a fatal stochastic effect in the context of hot particles.
The Working Group does not understand the Commission's intent: is effective dose not seen fit to be used because only fatal stochastic effects are of interest and effective dose is assessed also taking account of those effects termed 'quality of life detriment'?

One might be brought to argue that ICRP might be thinking in terms of a site specific cancer incurred as a result of hot particle incorporation; nevertheless, the following sentence ("These probabilities should be integrated over all the range of situations and possible doses") does not support this view and one is left fraught with doubts.
The Working Group feels that the Commission should clarify these aspects, e.g. in Annex A, and give ad hoc guidance for these assessments to be adequately addressed in practice.

COMMENTS ON ANNEX A

The conceptual framework in the Annex is clear enough, if exposed too succinctly in that the Annex in its present form is more of a summary than the in-depth discussion the subject requires. In fact, there are a great deal of aspects that should be enlarged upon because these crucial subject matters need to be addressed as extensively and as clearly as possible, as pointed out in the General Comments.

One specific aspect (heritable risk) is pointed out in the comment on paragraph (A17) below; in general, the Working Group suggests that the following questions should be addressed more clearly.
- Use of ERR and EAR models, with a discussion of the findings of the UNSCEAR 2000 Report in this regard; the part concerning transfer of risk between populations should also be made clearer.
- Quality of life detriment: the rationale underlying the analytical formula for the adjusted lethality fraction, qT, and the 0.1 value chosen for qmin in Section A.3.5.
- Derivation of wT's values; the considerations on thyroid, salivary glands, brain and kidney risk should be enlarged upon. ICRP contends in paragraph (A39) that additional judgment was used this respect, and the contention should be supported by a review of available data.

Paragraph A17, fourth bullet

"New genetic risk coefficients recommended by ICRP consider exposure and genetic risk for two generations only – the equilibrium value used in ICRP60 is judged to be of questionable scientific validity because of the unsupported assumptions necessary on selection coefficients, mutation component and population changes over hundreds of years." (Bold is the AIRP Working Group's)

As a matter of fact, the Working Group points out that ICRP took into consideration genetic risks for all generations in Publication 60 as well as in statements subsequent to Publication 26. Thus, considering genetic risks for the time span of two generations only is certainly one of the aspects that deserve in-depth discussion by the Commission since it is a considerable departure from previous ICRP postures.

More in general, a discussion of the findings in the UNSCEAR 2001 Report in this respect is absent and the line of reasoning and references according to which the present estimates were reached are obscure.
In this context, the Working Group again emphasises that aspects so crucial to the system as estimates of risk coefficients need adequately to be expanded upon by ICRP, all the more so that the Commission has seen fit to use rather strong language in respect of its own previous evaluations ("the equilibrium value used in ICRP60 is judged to be of questionable scientific validity").

Section A.3.1 Uncertainty and sensitivity analyses

The title of the Section gives rise to doubts in that no mention of sensitivity analyses on radiation risk estimates can be found in the five paragraphs of the section.

Table A1

Two minor points need to be made as regards Table A1a, and similar considerations also apply as concerns Table A1b.

Firstly, referring to the a) section (Whole population) of Table A1, the (incidence) nominal risk coefficients are shown as integers, and as such they add up to 1811, not 1812. If the (incidence) nominal risk coefficients and the lethality fractions, which are also shown as integers, do not really have implicit decimal digits there does not appear to be a single, traceable criterion adopted for the rounding of decimal figures shown in subsequent columns.

For instance, the unrounded product of the nominal risk coefficient in column 2 times the lethality adjusted nominal risk in column 3 should equal 66.1 (not 67) for the breast, 10.8 (not 10) for the ovary, 23.5 (not 23) for the bladder, 5.3 (not 7) for the thyroid.

Secondly, the nominal risk coefficient for cancer incidence is given in 10-4 P Y Sv in the second column of Tables A1 and A2; ICRP should clarify this as one would think that lifetime probabilities of incidence are addressed here, and not yearly probabilities.

Table A1 Unrounded Values
Tissue------Nom. risk coeff.----Lethality----Lethality----Relative cancer----Detriment----Relative
--------------per PYSv-------------------------adj.nom.risk--free life risk------------------------detriment
Oesophagus----17-----------------0,93----------16,9----------0,87----------------14,7----------0,023
Stomach---------91-----------------0,83----------88,6----------0,88----------------78,0----------0,121
Colon-----------101-----------------0,48----------76,4----------0,97----------------74,1----------0,115
Liver--------------19-----------------0,95----------19,0----------0,88----------------16,7----------0,026
Lung------------100-----------------0,89----------98,9----------0,8------------------79,1----------0,122
Bone surface-----7-----------------0,45---------- -5,1----------1----------------------5,1-------- -0,008
Skin---------- -1000-----------------0,002----------4,0----------1----------------------4,0----------0,006
Breast-----------121--------------- -0,29----------66,1----------1,29----------------85,3----------0,132
Ovary-------------13-----------------0,57----------10,8----------1,12----------------12,1----------0,019
Bladder-----------43-----------------0,29----------23,5----------0,71----------------16,7----------0,026
Thyroid-----------24-----------------0,07------------5,3----------1,29------------------6,9--------- 0,011
Bone marrow----41-----------------0,67----------37,0----------1,63-----------------60,3----------0,093
Other solid-----214-----------------0,49---------163,9----------1,03---------------168,8----------0,261
Gonads- .--------20-----------------0,8------------19,3----------1,32-----------------25,4----------0,039
..hered.
Total-----------1811--------------------------------634,9------------------------------647,3----------1,000