Controllable dose - COMARE letter

Professor R H Clarke
Director, NRPB
Chilton
DIDCOT
Oxon OX11 0RQ

14 July 2000

Dear Roger

CONTROLLABLE DOSE
PRESENTATION BY PROF ROGER CLARKE, 4 FEBRUARY 2000, NRPB CHILTON

Thank you for your interesting presentation on controllable dose and the opportunity this allows us to comment on the proposals it contains. It was also useful to be able to discuss various points with John Dunster following your presentation.

As you know, COMARE is not constituted to give advice on the acceptability of radiation exposures. Its terms of reference require it to "assess and advise Government and the Devolved Authorities on the health effects of natural and man-made radiation in the environment and to assess the adequacy of the available data and the need for further research". We recognise that health effects are but one of the many factors involved in judging the acceptability of radiation exposures; our task is to present the best estimate of health effects that can be made, for the information of Ministers and the general public. Our comments are, therefore, restricted to general points and also to our specific remit.

Necessity of changes to the current system

In the past, ICRP have undertaken reviews to their current advice only when significant new data on the risks of radiation exposure have become available. This is not the case for the present consultation. In your presentation you gave some explanation of the reasoning behind this and John Dunster gave some further insight into the perceived need for ICRP advice to receive more credibility. Nevertheless, it is not apparent to the uninformed reader why these new proposals are necessary. Whilst we accept that ICRP 60 is a complex document and would benefit from clarification, we are concerned that the concept of controllable dose may be perceived as a change in the underlying principles of radiological protection, which would need to be scientifically justified and also transparent. We felt that this would be particularly so regarding the proposed change from a dose limit for the general public to a reference level.

General Protection of the Environment. You noted in your presentation the inherent ICRP premise that the protection of the general public ensures the protection of the environment more generally, that is the protection of other species from harm. This is an admirable concept and we were pleased to hear that any proposed changes would strengthen this ideal.

Justification. This topic was well discussed by the audience following your discussion and to some extent is outside of our remit but we feel it may be helpful if we noted our support for the retention of the principle of Justification as argued by those representing the Authorising and Licensing authorities.

Collective dose

The proposals being considered by ICRP envisage restricting the use of the concept of collective dose. Collective dose is central to much of the work of COMARE and the Committee's position is, therefore, set out in detail below.

In almost all cases, the doses received by members of the public are at or below levels associated with increased risk detected by epidemiological studies. We do not, however, accept the argument that because an effect cannot be measured with any confidence it is, therefore, of no importance. The potential radiation effects that are of most concern are cancer and germ line (inherited) mutations. In the Japanese atomic bomb survivors, the frequency of cancer increases with increasing radiation dose but as you noted, for reasons of statistical power these effects cannot be demonstrated below around 80 mSv. Nevertheless, there are good biophysical reasons to expect that the types of radiation damage to DNA, believed to be responsible for cancer and mutations, occur in proportion to the radiation dose and that there is no threshold below which adverse radiation effects do not occur. Thus, we consider that it is prudent to assume that the induction of such effects can be extrapolated to very low doses.

There are those who would disagree with this view and who hold that there is a threshold and that very low doses may even be beneficial. Others are persuaded that low doses are much more effective than are currently assumed. Neither of these positions has, in our view, enough supporting evidence to seriously challenge the consensus position that is consistent with the bulk of the available evidence. We acknowledge that estimates based on the linear low dose extrapolation model are uncertain, they may over- or under-estimate the real levels, but our opinion is that they are not likely to greatly underestimate the real level (for example, by more than a factor of 2).

COMARE is frequently asked to assess the health effects of radiation exposures, both historic (eg, by releases from a nuclear plant) and future (eg, as the outcome of an authorisation application). To do this using the linear no-threshold model requires first the estimation of collective dose, ie, the integrated average dose to the exposed population, and then the application of a risk factor to those doses.

With historic exposures the various exposed populations can frequently be well defined and modelling (and in some cases measurement) can give a reasonably secure estimate of the radiation doses that they have received (we prefer to consider high and low LET doses separately). We also routinely use sensitivity analyses to test the effect of the least secure assumptions made in the dose modelling. An example of COMARE's use of a narrowly defined collective dose and its health consequences as applied to young people in Seascale can be found in COMARE's 4th Report. Such uses of collective dose are an essential part of COMARE's investigation of the effects of historic releases.

A much less certain use of collective dose has to be applied in respect of authorisation applications for future releases from nuclear plants about which COMARE's advice is sought. There are serious modelling difficulties when dealing with large populations and these increase as projections are made into the distant future. We have truncated our own estimates to 500 years (see, for example, our advice in connection with the THORP authorisation). These collective doses, and the estimated premature deaths due to cancer resulting from them, need to be considered in the context of (a) these uncertainties, (b) the very much greater collective dose and premature deaths resulting from inescapable background radiation, and (c) the risks associated with other activities that are regarded as tolerable by society. It may be argued that such estimates are of limited value, and we would not disagree. Nevertheless, when considered in the context stated above, they provide Ministers and the public with an order of magnitude of the health effects that may ensue. Such information enables some sort of comparison to be made with the risks from alternative technologies and becomes part of the equation leading to regulatory decisions. We would, therefore, be very reluctant to dispense with the use of collective dose in these situations. However, we do not consider that collective dose can be meaningfully estimated over geological time spans. The use of the collective dose concept in connection with the storage of radioactive waste does not seem to us to be valid.

In conclusion, we would prefer to see a clarification of the current system, as contained in ICRP 60, rather than the major change envisaged in your presentation. We feel that the use of sensitivity analyses would help to explain and quantify the uncertainties in the data used to draw up the current risk estimates. Having said that, we welcome the fact that ICRP is trying to make its advice more available and usable, and we hope that our comments will prove helpful in this context.

Yours sincerely

Professor BRYN BRIDGES
COMARE Chairman