Keeping the ICRP Recommendations Fit for Purpose

The science behind radiological protection is complex. It starts with the physical interaction between ionising radiations and the material that composes the human body (and other biota), then considers the potential changes induced by these interactions, including the response of different issues within the body and the whole body implications of those local responses, and tries to quantify the harm that might be done (detriment). It then considers the acceptability of this harm in terms of the tolerability of risks and by putting them into perspective with other risks.  It also considers how different groups and individuals might respond if exposed, recognising workers, the public and patients as different groups with different potential gains and losses and planned, existing and emergency exposure situations. The field thus encompasses physics, biology, sociology, ethics and politics.

The International Commission on Radiological Protection was “established to advance for the public benefit the science of radiological protection, in particular by providing recommendations and guidance on all aspects of protection against ionising radiation. The Main Commission is the governing body, setting policy and giving general direction” (Ref).  The recommendations of the ICRP form the basis of radiological protection around the globe. A useful review of the effects of this last major restatement of the recommendations can be found in a PHE paper “Application of the 2007 Recommendations of the ICRP to the UK”. (Ref)

ICRP have now pre-released a major discussion document as an early step in the consultation process for the next round of recommendations. Christopher Clement et al 2021 J. Radiol. Prot. in press. On line (updated version) available at This article is based on the accepted manuscript 20 July 2021.

In this we are reminded that the objective of the system of radiological protection described in ICRP-103 is “to contribute to an appropriate level of protection for people and the environment against the detrimental effects of radiation exposure without unduly limiting the desirable human actions that may be associated with such exposure”. The review that preceded this document started 20 years ago and took 10 years. “While it is safe to conclude that the System is robust and has performed well in relation to the protection objectives, the System must adapt to address changes in science and society to remain fit for purpose.

It is suggested that the ICRP-103 objective to prevent tissue reactions (deterministic effects to us oldies) should be modified to recognise that there are medial situations, emergency situations (and space exploration) where tissue reactions may be tolerated to achieve the desirable benefit of a particular activity. This seems sensible but is going to add, rather than remove, complexity.

A review of the lifetime risk estimates imbedded in the concept of detriment is due a review to reflect the evolution of scientific knowledge of risks and expert judgement. “In addition, explicit recognition of differences in detriment with age at exposure and between males and females could improve the clarity of application of the System, showing, in particular, that risks to young children are greater than risks to adults, and that risks to older individuals are low.” This could be useful, for example, in removing the perceived need to evacuate elderly people from their homes during a radiological or nuclear emergency to avert radiation doses of as little as 30 mSv which are of no real threat to them.

The discussion paper points out that the current system “principally deals with health effects resulting directly from exposure to radiation” which is not entirely in line with the WHO definition of health as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity”. Including mental and social wellbeing in the system can only make it much more complicated, situation dependent and subjective (but pretty much removes the need to worry about the physics of the fundamental interactions).

There is a section (Section 2.3) on the protection of the environment and non-human biota. I have always considered this a job creation scheme for radiobiologists and mathematical modellers (of which I used to be one) and of little practical value in the real world of radiation safety. I realise that this view sees me ejected from the moral high ground.

In Section 2.5 the paper reports that “There have been many requests for more guidance on how to balance societal, economic, and other factors in the optimisation of protection and safety, requiring input from many fields of expertise” and summarises the work that ICRP have published in this general area. This includes ICRP Task Group 114 which seems to suggest that there are occasions where the lowest exposures or risks are sought when a better balance could be achieved to advantage.

Also in this section is a discussion of a more holistic approach in safety assessments of medical facilities. Again, this seems to be running the risk of making the system more complex and maybe asking too much of one stream of work.

Section 2.6 discusses dose limitation and worries that, since it only applies to planned exposure situations, it fails the “ethical obligation to protect individual people under all circumstances”. This seems to me to be a bit like worrying that the seat belt and airbags in my car don’t provide me with any protection when I am walking around a shopping centre.

The report suggests a broader principle which would apply in all situations and encompass the concepts of limits, constraints and reference levels, possibly combining the latter two concepts into one. This is an interesting thought, worth thinking about provided the ICRP are willing to back peddle if it does not work out as hoped.

Dose constraint

A prospective and source-related restriction on the individual dose from a source, which provides a basic level of protection for the most highly exposed individuals from a source and serves as an upper bound on the dose in optimisation of protection for that source. For occupational exposures, the dose constraint is a value of individual dose used to limit the range of options considered in the process of optimisation. For public exposure, the dose constraint is an upper bound on the annual doses that members of the public should receive from the planned operation of any controlled source.

Reference level

In emergency or existing controllable exposure situations, this represents the level of dose or risk, above which it is judged to be inappropriate to plan to allow exposures to occur, and below which optimisation of protection should be implemented. The chosen value for a reference level will depend upon the prevailing circumstances of the exposure under consideration.


The report states that “defining a fundamental principle to protect the individual would result in a System where all three fundamental principles apply under all circumstances regardless of the exposure situation or category. This change would require the re-examination and clarification of the distinctions between limits, constraints, and reference levels”. (I’m not sure how you can put a useful dose limit on an accident or malicious use of radiation).

Section 2.7 suggests that the experience of using the three exposure situations introduced by ICRP-103 has led to the opportunity to update, clarify and expand the guidance. This seems reasonable and the application to space travel interesting.

ICRP identify ethics, communications and stakeholder involvement and education and training as important overarching considerations and briefing discuss each in turn in Section 3.

It is proposed to use absorbed dose (in gray) for the control of doses to individual organs and tissues for the avoidance or minimisation of tissue reactions leaving equivalent dose (in sievert) as an intermediate step in the calculation of effective dose. “Radiation weighting could then be considered separately for tissue reactions and stochastic effects for the calculation of radiation-weighted absorbed dose in Gy and effective dose in Sv, respectively.” This is intended to “apply scientific knowledge more appropriately and simplify radiological protection, with a clearer distinction between organ/tissue doses in absorbed dose in Gy and effective dose in Sv”. This does seem more transparent than the switch we currently make from seiverts to grey at (an arbitrary) high dose as at low dose you are concerned about stochastic effects and high doses with tissue reaction.

The paper reports discussions with ICRU with the intention that “the measured quantities for the estimation of effective dose would be related directly to effective dose in the reference phantoms, renamed as ‘dose quantities’ (ambient and personal dose) rather than ‘dose equivalent quantities’. Operational quantities for the measurement of doses to the skin and lens of the eye will become ‘absorbed dose quantities’”. Another episode where those of us who work around radiological protection are required to forget the hard learned jargon we work with and replace it with a different set.

In Section 4.2, discussing effective dose, the paper discusses the increased use of more accurate and differentiated anthropomorphic phantoms leading to more accurate and transparent values of detriment and relative detriment separately for males and females of different age groups. The report suggests that “Revisions to the methodology of calculation of effective dose could improve its suitability for the assessment of risk. Best estimates of health risk should be calculated using estimates of absorbed doses to organs/tissues and age- and sex-specific risk models for individual types of cancer, but risk estimates at low doses will still be subject to the uncertainties inherent in risk projection models”. The question this raise in my mind is “are the age and sex differences larger than the uncertainties in the estimates?”

It is suggested in Section 4.4 that the revision of dose per unit intake values in the light of the new recommendations should be more rapid than previous experiences due to preparatory work and experience gained. This seems to be a reasonable hope.

Section 5 suggests a review of the classification of radiation effects as either stochastic effects or harmful tissue reactions to ensure that it remains fir for purpose, suggesting that “For example, for protection purposes, it may be useful to distinguish between severe and other tissue reactions, or between short-term and long-term health effects”. This seems reasonable. There are occasions where the gain from a process may be worth suffering a mild or temporary tissue effect.

Since the last recommendation were made there has been considerable research and epidemiological study of the impacts of low exposures to ionising radiation. A Task Group is currently reviewing the linear no threshold assumption in the light of this work. It looks likely to survive.

It is almost certain that different people have different susceptibility to harm from ionising radiation but likely that there is insufficient information to include this in a system to protect workers and the public. “However, there are already efforts to individualise radiological protection of patients which should be considered in the review of the System, taking into account scientific, ethical, and practical aspects”.

Similarly, there is now more information on heritable effects that should be considered.

Likewise, there is more data on relative biological effectiveness and it is likely that a more sophisticated approach may now be appropriate.

The idea that “detriment could be calculated separately for males and females and at different ages at exposure, and the corresponding values of relative detriment could be used directly in the calculation of effective dose, rather than the current use of simplified age- and sex-averaged tissue weighting factors” sounds good. As does “explicit treatment of detriment from irradiation in utero could also be re-evaluated”.

There will also be a conversation about the replacing detriment with other proposed measures of harm such as fatality or disability-adjusted life years.

The discussion paper concludes that “The last review of the System of Radiological Protection was initiated 23 years ago, and the current General Recommendations (ICRP, 2007) were published 14 years ago. The System has performed well and remains robust, and there are significant practical benefits to stability in the System. Nonetheless, it must progress to remain fit for purpose as society evolves, scientific understanding advances, and new uses of ionising radiation emerge”.

The ICRP and others continue to research the effects of ionising radiation on people, biota and the environment. A time comes when the strengths and weaknesses of the current system should be discussed and new knowledge should be systematically reviewed incorporated into a revised international system of radiological protection. It appears that that time is approaching. This paper is one step in the consultation process. An ICRP Digital Workshop on 19 – 20th October is another step (Ref).

I shall watch this process develop with interest and get involved if I deem it good use of my time.






What is the case for the nuclear emergency planning community’s snobbishness about improvised respiratory protection for the general public?

In my many years as an emergency planner in the nuclear industry I’ve never heard a real debate about respiratory protection as a public protective action in the event of an accidental atmospheric release of radioactivity. It has always been dismissed because without proper masks and fit testing the protection factors offered are compromised.

In 1981 the IAEA [1] identified that respiratory protection was one way to reduce dose uptake in workers and members of the public. It recognised that high levels of protection require properly designed and fitted devices and realised that these would only be available to those with planned roles in a response. They accepted that if any use is to be made of such measures by the public, the simple equipment and techniques to be employed can only be of a very rudimentary nature.

They provided a quite extensive table of filtration factors for common materials. This included the finding that 16 layers of man’s cotton handkerchiefs provides a geometric mean efficiency of 94.2% against aerosols of 1-5 μm particle size – a protection factor not to be sneezed at. At eight layers the efficiency drops to 88.9%. A single bath towel is worth 73.9%.

The public, they said, “can be advised to use such simple items while proceeding to take shelter, and possibly during sheltering. Similar precautions could be recommended while members of the public were being evacuated from a contaminated area”.

In 2002 the US NCR published a document [2] which suggested that improvised respiratory protection can be used as a secondary protective action that can be used to provide a nontrivial level of additional protection. They also provide a table of protection factors.

In 2007 IAEA stated that [3] “Improvised respiratory protection (e.g. a wet cloth over the mouth and nose) has been shown to be effective but it has not been demonstrated that the public will apply it effectively during an emergency. Improvised respiratory protection should not be assumed to provide adequate protection from an inhalation hazard and therefore its implementation should not be allowed to interfere with evacuation or sheltering”. This does not say that improvised respiratory protection should not be recommended under any circumstances; it just says it should not be used instead of shelter or evacuation.

The latest advice on the protection of the public in the event of a nuclear accident from PHE [4] makes no mention at all of RPE, improvised or otherwise. This publication suggests a dose reduction factor of 0.6 for inhalation dose from shelter in place over the period of a release.

So why have we taken improvised respiratory protection out of our tool box of techniques to reduce public dose? It seems to offer protection factors at least comparable with shelter in place for particulate activity.

We worry about golf courses and caravan parks within our DEPZs where shelter in place is considered likely to be ineffective. Could we at least provide a supply of half decent face masks with the stable iodine tablets we store at these establishments as a secondary protection while thinking about evacuation?

Does the Covid-19 experience that shows that large fractions of the population will wear face masks when advised and has made them far more available to the public change our current attitude?


  1. IAEA, Safety Series No. 55, Planning for Off-Site Response to Radiation Accidents in Nuclear Facilities
  2. US NCR, Perspectives on Reactor Safety, NUREG/CR-6042, Rev. 2 SAND 93-0971
  3. IAEA Safety Standards Series No. GS-G-2.1, Arrangements for Preparedness for a Nuclear or Radiological Emergency (2007),
  4. PHE, Public Health Protection in Radiation Emergencies, PHE-CRCE-049, (2019).

Learning for nuclear emergency planning from COVID-19?

The model of the UK response to a nuclear emergency that results, or may result, in a plume of radioactive material spreading across populated areas of the countryside is to get the local responders, notably the local authority, emergency services and health services, in one place to discuss, decide, coordinate and respond.

Within the model is a unit called the Science and Technical Advise Cell (STAC) with the mission “to ensure timely coordinated scientific and technical advice during the response to an emergency”. We were told that “The STAC should bring together technical experts from those agencies involved in the response and who may provide scientific and technical advice to the Gold Commander. The purpose of the cell would be to ensure that, as far as possible, scientific or technical debate was contained within the cell so that the SCG (and others involved in the response) received the best possible advice based on the available information in a timely, coordinated and understandable way.”

Implicit in this process is the assumption that in any event there is an objective truth and that if scientists chat about it for a while they will determine and understand that truth and be able to explain it to the decision makers who have been too busy on other aspects of the response to explore the science for themselves. The decision makers will be jolly grateful to the STAC and, armed with the scientific consensus, will go on to make the right decisions. They might even say time after time that they are being driven by the science.

In this ideal world, these decisions will be reported to the public and to the media, will be implemented and the crisis will be bought to a close. Also in this ideal world the decisions turn out to be the “right” decisions and the only decisions that could be considered to be “right”, all other options, explored and unexplored, being “wrong”.

One thing we have definitely seen with the coverage of Covid-19 is that the media will not just forward your advice to the public as you might hope. Instead they will turn out an army of interpreters who, fearful that Mrs Miggins and her neighbours will not understand that those in a defined area are being asked to shelter and those outside the area are not, will explain at length what they think “shelter” means and why it has been recommended. They will then find a talking head to explain it again and then another to say that the previous interpretation was wrong and that the public should be being advised to do something else entirely. They will summarise by saying that there is a lack of clarity in what the advice is and who it applies to before cutting to a member of the public who will confirm, in response to a loaded question, that they don’t understand the advice and that they are very worried.

Returning to the decision making, the major issue is that the science does not give all the answers. We may be able to estimate radiation doses to the public, with and without protective actions, but these will be educated guesses rather than accurate. The amount of dose saved (benefit) that makes a protective action (with a cost) worthwhile is debatable and probably different for different people in different situations. The decision to interrupt the lives of people and ask them to stay in their homes knocking back stable iodine tablets is therefore a judgement call not the outcome of a neat equation. This is particularly true when you realise that the estimates of future doses are horribly dependent on assumptions made about what is happening, and what is going to happen, in the bowels of a damaged nuclear facility, what the weather will be when the activity gets out and where the members of the public will be and what they will be doing. Again the media will bring out an army of “experts” to discuss the technology, the science and the decision making process and will argue that the science is debatable, the process flawed and that any of the decisions made are dubious.

Maybe what should happen is that advisors advise and decision makers decide. The spokesperson issuing the advice should state that the decisions have been taken by the Strategic Coordination Group who took into account scientific advice, advice about the incident and how it could develop and the concerns of and for the people affected.

The media should be asked to transmit the advice as given and to resist reheating and reinterpreting it.

That will work.

What are the lessons for the nuclear industry from Covid-19? How do we ensure that our protective action decision making process is robust, transparent, unambiguous and trusted to ensure a high level of public compliance and optimum dose reduction?