Comments on the draft report for consultation: The Use of Effective Dose as a Radiological Protection Quantity
Joint comments from the Dutch National Institute for Public Health and the Environment (RIVM) and the Dutch Authority for Nuclear Safety and Radiation Protection (ANVS):
RIVM: H Slaper, T van Dillen;
ANVS: P Görts, M van Bourgondiën, F van de Put, M Farahmand.
Part 1: General remarks
The effective dose as a risk-adjusted dosimetric quantity and its proper use is highly relevant. We appreciate the extensive description of (1) the scientific concepts and basic elements underpinning the radiological quantity, (2) its application in the management of protection against stochastic effects in each exposure situation (planned, existing and emergency) and for all exposure categories (public, occupational and medical), and (3) the further guidance on the limitations of its use. In addition, we value the proposed discontinuation of the “equivalent tissue/organ dose” as a separate protection quantity against tissue reactions (deterministic effects) and we agree with the fact that setting limits on the absorbed dose would be a more appropriate approach to prevent such effects.
Given the fact that detailed information on age, sex and population differences are available (and used) in the derivation of the “averaged” quantity effective dose, it would be welcome to also provide further guidance on the use of these underlying data for more detailed risk-estimates for instance for specific groups at risk.
This publication is concise, well written and, due to its broad scope, it will be of great value to the entire radiation-protection community. Consideration of the comments and issues raised in part 2 of this review (see below), may improve the draft report.
Part 2: Specific remarks
Section Main Points
Lines 221-226
The use of E above 100 mSv in emergency situations (short-term, high exposure at high dose rates) requires considerations of two issues: (1) the possibility that tissue reactions occur, and (2), a consideration that risks for specific groups might be higher than derived using the standard detriment of the Sievert. Both issues are mentioned and elucidated in lines 970-982. In the summary lines 221-226 only the first of these points is mentioned. We think the second issue should also be indicated in the summary (main points).
Lines 227-230
ICRP distinguishes the use of the term “dose coefficients” for internal exposures and “conversion coefficients” for external exposures. In this summary item the term “dose coefficients” is used for both.
Lines 267-270
E is an approximate indicator of risks and should not be used for individual risk estimates. It would be of great interest to also have some more ICRP-guidance on how to assess risks more specifically. In other words, how to account for age, sex and population-group specific characteristics. The present document does not specify the preferred approaches. We suggest that the ICRP work this out in a future publication.
Lines 285-287:
The meaning of the following sentence is not clear: “For public exposures, components of dose integration in time and space should be considered in estimating collective does…large populations over very long periods of time.” It could be misinterpreted as an encouragement to use collective doses over very large populations and long periods of time. This could be avoided if it is clearly stated that the “boldface highlighted” arguments just above (lines 282-285) apply to both the public and occupational exposures.
Section 1. Introduction
Lines 337-338:
“the risk of thyroid cancer and leukemia is greater at younger ages at exposure.”
This is correct, but it could easily be misinterpreted that the cancer risk is greater at younger ages, which especially for thyroid cancer is not true. Our suggestion is to rephrase to: “exposures received at younger ages lead to higher lifetime excess risks for thyroid cancer and leukemia compared to similar exposures received at a later age.”
Section 2. Health Effects
Lines 412-417:
Should mental retardation also be mentioned here?
Lines 673-675
“… showing greater nominal risk coefficients and detriment values for females by a few tens of percent.” This only holds for the total nominal risk and total detriment values, since for individual cancer detriments the differences can be much larger than a few tens of percent (for example differences in thyroid cancer and lung cancer as can be seen in figure 2.1, and breast cancer). We suggest adding “total” to the “nominal risks” and “detriment values”. Also it is not fully clear where the few tens of percent refer to: the comparison with male or with the averaged risk factor.
Line 737 Fig 2.1
The X-axis refers to the ages at exposure, therefor X-axis titles should be: Female age at exposure (y) and Male age at exposure (y) respectively.
Section 3. Dosimetry
Line 830
Typing error in word “fundamental”.
Lines 870-880
The radiation weighting factor wR is largely selected from and based on measurements of RBE values determined in biological end-point studies, as mentioned in line 878. A few examples of relevant end-points could be given with the comment that these end-points are related to stochastic effects.
Lines 910-917
The value of wR=20 for heavy ions is indicated to be conservative. For alpha particles the conservatism is not mentioned; does this mean that the value of 20 is considered a best estimate? (could a range be indicated?)
Line 915
Should Section 2.1 not be Section 2.7?
Lines 961-982
Paragraph 46 on the upper limit of applicability of the effective dose:
The applicability of effective doses above the 100 mSv can be highly relevant in emergency situations, and this also holds for the two factors that need consideration when 100 mSv is exceeded: possible tissue reactions and the applicability of the DDREF. As indicated earlier, the DDREF should also be mentioned in the main points, and it could use some more guidance in this section. We propose mentioning at least what is meant by high dose rate also here (we assume dose rates exceeding 5 mGy/hr, as implied by lines 327 and 328 in the introduction). Furthermore, could “somewhat greater” in line 981 be quantified here (factor of 2 increase?).
Lines 1004-1007 in relation to the previous paragraph 48:
The implementation of revisions regarding ICRP recommendations and models take a long time. The ICRP-103 recommendations were already issued in 2007 and require new revisions which are now underway. In the meantime, ICRP Publication 119 from 2012, with dose and conversion coefficients based on ICRP Publication 60 (1991), is still widely used. A table with most recent sets of dose and conversion coefficients and the associated recommendations and models on which they are based would be helpful. This table could best be hosted and updated on ICRP’s website).
Lines 1013-1014
In evaluating population exposure doses over time, changes in dose/conversion coefficients can lead to trend variations and hence to a (possible) misinterpretation of changes. A warning on this issue, or guidance on how to deal with that, would be welcome.
Line 1038 in paragraph 52
For external exposure, ICRP uses the term “conversion coefficients” instead of “dose coefficients” (the latter term is used for internal exposures, e.g. by inhalation and ingestion). Our suggestion is to replace “dose coefficients” by “conversion coefficients” in line 1038.
Line 1084
Dose limits are not discussed in section 2.1, probably section 2.2 is meant here.
Lines 1119-1126 Paragraph 56 on the stochastic risks of skin cancer
Reference is made to ICRP-60 (1991) and ICRP-103 (2007), but no argumentation is given on a large change in approach between ICRP-60 (and ICRP-59) on the one hand and ICRP-103 on the other hand. ICRP-60 (and ICRP-59) show large differences with respect to the excess-risks for skin areas exposed to UV-radiation and for skin areas that are (usually) not-exposed to UV. The nominal risk in ICRP-103 is taken from ICRP-60, however the risk-distinction between UV-exposed and non-UV-exposed is not included in the ICRP-103 publication. This reflects an important change in approach between ICRP-103 and ICRP-60, and should at least be explained. The reference made to Charles et al 2003 does not give an explanation: this study only supports averaging over skin areas, but the interaction with UV-exposure is not studied in that analysis. The ICRP-103 approach leads to a significant underestimate of excess risks due to ionizing radiation for the UV-exposed areas and an overestimate for non-UV-exposed areas.
Lines 1158-1161
We suggest to add: “for various exposure geometries” on line 1160. The sentence then becomes: ”..the maximum of the conversion coefficient curves for effective dose for various exposure geometries as function of particle energy for all particles considered in Publication 116.”
Lines 1214-1236, section 3.8 on collective dose
Some further guidance how to avoid misuse of the collective dose concept would be welcome. The suggestion is made to restrict dose-range and/or time period (lines 1234-1235) but no guidance on this is given.
Lines 1237-1238
dN/dE is the number of individuals … between E and dE, but per unit effective dose. We suggest to delete dE from the description and leave the rest as is: thus “… where dN is the number … between E and E+dE and …”
(Note: the equation of the collective dose between lines 1236 and 1237 is correct)
Section 4. Occupational and Public Exposures
Line 1254
“… exposure situations” (please add “s”)
Line 1269
Please change “… system of protection …” to “… system of radiation protection …” (add radiation).
Line 1428 and further, paragraph 83
It should be mentioned that the reference levels in emergency exposure situations (as well as in existing exposure situations, para 82) should be compared with residual effective doses, which are the remaining effective doses after implementation of countermeasures.
Lines 1452-1460, paragraph 84
Please indicate how to account for the natural background.
Section 5. Medical Exposures
Line 1602
“Surveys are made to establish …”.
Should this be “Surveys are carried out to establish …” ?
Lines 1637-1656, Table 5.1
It would be good to explain the choice for these 3 countries (UK, USA and Russian Federation). Furthermore, the values for the UK are relatively low when comparing e.g. to other European countries, and they may therefore not be representative.
Line 1710
“… five hundred lower.” Please add a reference here.
Lines 1842-1849, Table 5.2
See previous comment for “lines 1637-1656, Table 5.1” on the values for the UK.
Moreover, the proposed term for dose level “Minimal” (as well as moderate) is not mentioned in the text.
Section 6. Summary and Conclusions
Lines 2018-2022 in relation to lines 2000-2008
For low doses to the public and workers, the use of the sex- and age-averaged effective dose in combination with optimization below dose constraints and reference levels should offer sufficient protection for all groups within a population (lines 2000-2008). However, for effective doses exceeding 100 mSv (lines 2018-2022), the latter may not be valid anymore and a differentiation between sex and age group may become more significant. It would be appreciated if this were elucidated.