Taking into account the above described UNSCEAR estimates for the effects that can be attributed to the normal operation of NPPs, and its own findings, the ICRP recommended the use of ‘detriment-adjusted nominal risk coefficients’ for the only purpose of radiological protection at low doses. These coefficients are numerals expressed in % per unit dose, which — mul­tiplied by dose — aim at quantifying the plausibility or ‘degree of belief’ of latent effects as a result of radiation exposure. The coefficients are nominal, in the sense that they do not necessarily correspond to a real value, since they relate to hypothetical (not real) people who are averaged over age and sex. Since the different possible effects may cause distinct detriment to people, the coefficients are multidimensional, quantifying the plausible expectation of harm, and including among other factors the weighted plau­sibility of fatal and non-fatal harm, and life lost should the harm actually occur.

The detriment-adjusted nominal risk coefficients recommended by ICRP are:

• for malignancies,

— 5.5% Sv-1 for a whole population

— 4.1 % Sv-1 for an adult population

• for hereditable effects,

— 0.2% Sv-1 for a whole population

— 0.1 % Sv-1 for an adult population

which result in a combined value of

• 5.7% Sv-1 for a whole population

• 4.2% Sv-1 for an adult population.

The risk coefficients imply a central assumption of a linear dose-response relationship for the induction of cancer and heritable effects, according to which an increment in dose would induce a proportional increment in risk even at low doses. This assumption is essential for the practical implementa­tion of the system of radiation protection (see hereinafter) in order to provide the basis for the summation of doses of various levels, from differ­ent sources, and from external exposure and from intakes of radionuclides (Beninson, 1996).

It is again emphasized that the risk coefficients are nominal, i. e. artificially constructed using average phantoms. Figure 11.4 indicates how the effective dose is constructed using these phantoms (ICRP, 2007b).

International radiation safety standards have taken the UNSCEAR esti­mates and ICRP recommendations into account, rounding an overall nominal risk coefficients to ~5% Sv-1, as the basis of the requirements for limiting radiation risks. This is because, while it is not demonstrable, it is considered plausible that risks be attributable to radiation exposures, even at low doses, and therefore for reasons of social duty, responsibility, utility, prudence and precaution, it is ethically required that regulatory bodies do ascribe such nominal radiation risks to prospective exposure situations. However, both UNSCEAR and ICRP had made clear that while nominal risk coefficients can be used for attributing risk for purposes of prospective planning, they cannot be used for attributing factual health effects retrospectively.