1990-01-01 1990 RECOMMENDATIONS OF THE ICRP
1990 RECOMMENDATIONS OF THE ICRP
5.5.1. The optimisation of protection in public exposure
(186) In practice, almost all public exposure is controlled by the procedures of constrained optimisation and the use of prescriptive limits. It is often convenient to class together individuals who form a homogeneous group with respect to their exposures to a single source. When such a group is typical of those most highly exposed by that source, it is known as a critical group. The dose constraint should be applied to the mean dose in the critical group from the source for which the protection is being optimised. Occasionally, the same group will also be critical for other sources, or, if the critical groups are different, each group may incur some dose from the sources for which it is not critical. If the exposures in any critical group are likely to approach the dose limit for public exposure (see Section 5.5.2), the constraints applied to each source must be selected to allow for any significant contribution from other sources to the exposure of the critical group
(187) The main aim of constrained optimisation in public exposure should be to develop practical restrictions on the sources of exposure, e,g. in the form of restrictions on the release of radioactive waste to the environment.
5.5.2. Dose limits in public exposure
(188) With the widespread use of source-related dose constraints and practical restrictions on the sources of public exposure, generally applicable dose limits are rarely limiting in practice. However, because the constraints are source related they might, at least in principle, fail to take adequate account of the exposures from other sources. Although the Commission does not believe that this occurs to a significant extent, it continues to recommend dose limits for public exposure, if only to provide a limit on the choice of constraints
(189) The Commission defines the scope of its dose limits for public exposure by confining it to the doses incurred as the result of practices. Doses incurred in situations where the only available protective action takes the form of intervention are excluded from the scope of the dose limits. Separate attention has to be paid to potential exposures. (See Section 5.6.) The intended emission of radionuclides from installations, inluding the emission of naturally occurring radionuclides from installations such as mines and waste disposal sites, should be treated as practices. The resulting doses should be subject to the dose limits. Radon in dwellings and in the open air and radioactive materials, natural or artificial, already in the environment, are examples of situations that be influenced only by intervention. Doses from these sources are therefore outside the scope of the dose limits for public exposure. Other exposures to natural sources are also outside this scope. Radon in both existing and new dwellings is dealt with in Section 6.2.1. The conduct of intervention involves occupational exposure an should be treated accordingly.
(190) At least two approaches are possible in choosing limit for public exposure. The first is the same as that used for choosing occupational limits. Assessing the consequences is no more difficult than in the occupational case, but judging the point at which these consequences can reasonably be described as unacceptable is much more difficult. The second approach is to base the judgement on the variations in the exlsting level of dose from natural sources. This natural background may not be harmless, but it makes only a small contribution to the health detriment which society experiences. It may not be welcome, but the variations from place to place (excluding the large variations in the dose from radon in dwellings) can hardly be called unacceptable.
(191) The consequences of continued additional exposure giving annual effective doses in the range from 1mSv to 5 mSv are presented in Annex C. They provide no easy basis for a judgement, but do suggest a value of the annual dose limit not much above 1 mSv. On the other hand, the data in Figure C-6 of Annex C show that, even at a continued exposure of 5 mSv y-1 , the change in the age specific mortality rate is very small. Excluding the very variable exposures to radon, the annual effective dose from natural sources is about 1 mSv, with values at high altitudes above sea level and in some geological areas of at least twice this. On the basis of all these considerations, the Commission recommends an annual limit on effective dose of 1 mSv. Averaging over time is discussed in the next paragraph.
( 192) In deriving restrictions on sources of public exposure, some allowance is made for variations in the environmental pathways to man, but there will always be the possibility of larger transient changes. There will also be variations in the effectiveness of control procedures applied at the source and the Commission recommends that the transient increases in dose resultlng from such variations should be included in the doses subject to the dose limits. Doses due to major accidents are not subject to the dose limits because they can be dealt with only by interventlon. Since the detriment is a function of the accumulation of dose over many years, it would be unduly restrictive to require the controls to be related rigidly to annual dose limits. Some flexibility in the limits is desirable. The Commission's previous recommendations provided for a principal limit on the annual effective dose, with a subsidiary limit on the effective dose in some years, provided that the average effective dose over a lifetime did not exceed the principal limit.
This recommendation is still sound in principle, but the Commission has concluded that the very long averaging period in the subsidiary limit gives excessive flexibility. It now recommends that the limit for public exposure should be expressed as an effective dose of 1 mSv in a year. However, in special circumstances, a higher value of effective dose could be allowed in a single year, provided that the average over 5 years does not exceed l mSv per year. Because this represents only a slight change from the previous recommendation, the Commission recommends that the 5-year period should be applied retrospectively when the new recommendation is being implemented. For this purpose, values of effective dose may be added to earlier values of effective dose equivalent. It is implicit in this limit that the constraints for the optimisation of protection in the design of new Installations should be smaller than 1 mSv in a year.
(193) In selecting the limit on effective dose, the Commission has sought a value that would be only just short of unacceptable for continued exposure as the result of deliberate practices the use of which is a matter of choice. This does not imply that higher doses from other sources, such as radon in dwellings, should be regarded as un-acceptable. The existence of these sources may be undesirable, but it is not a matter of choice. The doses can be controlled only by interventlon which wlll also have undesirable features.
(194) Limits are also needed for the lens of the eye and localised areas of skin since these tissues will not necessarily be protected against deterministic effects by the limit on effective dose. Because the total period of exposure may be nearly twice as long as for occupational exposure, and because the exposed individuals may show a wider range of sensitivity than the more limited population of workers, the recommended annual limits (non-occupational) for the equivalent dose in these tissues are lower than those for workers. The Commission has adopted an arbitrary reduction factor of 10, leading to annual limits of 15 mSv for the lens and 50 mSv averaged over any 1 cm2 area of skin, regardless of the area exposed. The recommended limits are summarised in Table 6.
Table 6. Recommended dose limits
Dose limit (1)
Application Occupational Public
Effective dose 20 mSv per year, 1 mSv in a year(3)
averaged over defined
periods of 5 years(2)
Annual equivaient dose in
the lens of the eye 150 mSv 15 mSv
the skin(4) 500 mSv 50 mSv
the hands and feet 500 mSv -
(l) The limits apply to the sum of the relevant doses from external exposure in the specified period and the 50-year committed dose (to age 70 years for children) from intakes in the same period (see paragraph 143).
(2) With the further provision that the effective dose should not exceed 50 mSv in any single year. Additional restrictions apply to thc occupational exposure of pregnant women, which is discussed in Section 5 .3 3.
(3) In special circumstances, a higher value of effective dose could he allowed in a single year, provided that the average over 5 years does not exceed 1mSv per year.
(4) The limitation on the effective dose provides sufficient protection for the skin against stochastic effects. An additional limit is needed for localised exposures in order to prevent deterministic effects (sec paragraphs 173 and 194).
5.6. Potential Exposures
(195) The initial treatment of potential exposures should form part of the system of protection app]ied to practices, but it should be recognised that the exposures, if they occur, may lead to intervention. At this stage, there should be two objectives, prevention and mitigation. Prevention is the reduction of the probability of the sequences of events that mav cause or increase radiation exposures. It involves maintaining the reliability of all the operating and safety systems and of the associated working procedures. Mitigation is the limitation and reduction of the exposures if any of these ,sequences do occur. It involves the use of engineered safety features and operational procedures to control each sequence of events with the aim of limiting its consequences, should it occur. The arrangements for mitigation should not be restricted to plans for intervention. A great deal can be accompiished at the stages of design and operation to reduce the con-sequences of accident sequences so that intervention may not become necessary. It is difficult to compare, and to combine, the benefit of a reduction in probability (pre-vention) with that of a reduction in dose (mitigation) because a reduction in probability by a factor is not usually seen as equivalent to a reduction in dose by the same factor.
(196) In order to maintain a strict coherence in the treatment of actual and potential exposures, it would be necessary to extend the concept of detriment to include the probability of occurrence of the situation giving rise to the detriment. Techniques for achieving this are still being developed. Meanwhile, emphasis has to be placed on one part of the detriment, the probability of an attributable death. It must also be recognised that the uncertainties in estimating the probability of occurrence will usuaily be much greater than the uncertainties in estimating the probability of the consequences should the dose occur.
(197) The simplest way of dealing with the potential exposure of individuals is to consider the overall (a priori) individual probability of attributable death from cancer, rather than the effective dose, as the quantity to be used in the system of protection. For this purpose the probability is defined as the product of the probability of incurring the dose and lifetime conditional probability of attributable death from the dose if it were to have been incurred. A restriction corresponding to a dose limit can then be expressed in the form of a risk limit, i.e. a limit on the fatality probability. (See Section 5.6.3.) If the risk limit is derived from the probability of death attributable to exposure at the relevant dose limit, a corresponding level of protection will also be provided against non-fatal cancer and against deterministic effects.
(198) This use of the overall individual radiation risk is an adequate starting point for use in the system of protection, but it is not sufficient. This is because the situation will change if the event giving rise to the potential exposures actually occurs. At low prob-abilities of the potential event, an overall individual risk limit might imply doses when the event occurs that would be large enough to call for intervention or might result in deter-ministic effects. These undesirable outcomes should be borne in mind at the planning stage. They may call for lower risk constraints (analogous to dose constraints) than would be needed for high probability, Iow dose situations. When assessing the individual risk, it should be remembered that the conditional probability of deleterious effects if a dose is, in fact incurred may be higher than the nominal probability because the doses and dose rates may be higher than those for which the nominal probability coefficients have been selected and because deterministic effects may become important at these higher doses.
(199) The specification of collective detriment from potential exposures is difficult and controversial even if the consideration of detriment is limited to attributable deaths. It is not appropriate to depend on the use of the product of the probability of an event and the number of attributable deaths should it occur- the expectation value of the number of deaths-because this conceals the fact that the outcome will be either no consequences if the event does not occur, or the full consequences if it does. It also involves an implicit assumption of reciprocity between reductions in probability and reductions in the scale of consequences: i.e. the assumption that a frequent event with small consequences and a rare event with large consequences are equally detrimental if the expectation values of the consequences are the same.
(200) A more comprehensive approach to the collective detriment from potential exposures is that of multi-attribute analysis. Each characteristic (attribute) of the avail-able options has to be identified and quantified. It is then given a weighting factor judged to represent its importance. The weighted attributes can then be aggregated to provide a figure of merit or compared individually with the weighted attributes in other options. Either method leads to a quantitative, or semi-quantitative, basis for choice between options.