Principles and Application of Collective Dose in Radiation Protection

`For moderate increments above background, a linear relationship between the incremental dose and the incremental probability of a deleterious effect will be an adequate approximation' (International Commission on Radiological Protection). This, and similar statements going back over a number of years, validates the collective dose concept. Put simply, collective dose is the sum of the doses to all people in the exposed population and can be converted into the corresponding number of health effects on the basis of the linear dose - response relationship. Taking the argument further, collective dose can be linked to total detriment and, as a final step, a monetary cost can be assigned to unit collective dose representing the cost to society of the detriment. Collective doses find application in the optimisation of protection and estimates of total numbers of health effects have also been inputs to justification decisions. This all sounds intellectually satisfying and watertight until one looks deeper: is the collective dose concept applicable to large populations with very small individual doses and to populations that may exist several generations into the future? This report from the American National Council on Radiation Protection and Measurements provides insights into these issues. The biological assumptions underpinning the collective dose concept are reviewed and issues surrounding the application of the concept to various situations are discussed. The review of the biological assumptions is detailed, covering cellular and animal studies on effects of ionising radiation together with epidemiological data from exposed human populations. The conclusion reached is that it is prudent to assume a linear dose - response relationship in the low dose region. However, NCRP concludes that direct supportive evidence is limited and ultimately confidence in this assumption may derive from mechanistic biophysical studies. In connection with possible collective doses delivered into the far future following, say, disposal of long-lived radionuclides, the report draws attention to the uncertainties in the fertility, size and location of future populations, and their level of medical technology. The report makes a number of recommendations which include: regulatory limits should not be set in terms of collective dose; collective dose is most useful when applied to populations with known characteristics; all doses should be included in collective dose calculations as there is no conceptual basis for excluding any individual doses no matter how small; calculation of collective doses to future populations should be done with care and with recognition of the uncertainties. The report presents the ideas and issues clearly. It is of considerable interest to those who were involved in the discussions surrounding the BNFL Sellafield authorisation three years ago. However, it is not the last word on the subject as further work and ideas are required on the applicability and interpretation of collective dose estimates, particularly in the context of optimisation of protection.