Comments of the Swiss Federal Office of Public Health
This is an important publication giving access to the details of the radiation detriment calculation used in ICRP Publication 103, which is the basis of the system of radiological protection. The sensitivity analysis represents a major added value providing new elements that indicate on the one hand the stability of the methodology and on the other hand the directions for its improvement. The availability of an open-source software would be welcomed. This would enable members of the radiation protection community to become more directly involved in the estimation of radiological risk/detriment and would contribute to improve the transparency of the ICRP's work.
The publication should discuss in more detail the results of the detriment calculation, given in Table 3.6, and provide some considerations concerning the limitations of the methodology. It can be seen that the nominal risk coefficient (R) and the detriment (D) are close for most tumors/organs (see Table 3.6). This is due in part to the fact that the factor k for lethality adjustment and the factor q for quality of life adjustment are also close to each other. It is questionable whether this makes the current methodology to severity adjustment truly effective and other approaches may be considered, as pointed out in paragraph 127.
Paragraph 6 states that the nominal population is regarded as a mixture of people with non-modifiable factors, e.g. age and sex distribution, and modifiable factors, e.g. lifestyle. It is also noted that the nominal risk coefficient for radon exposure applies to a mixed population of non-smokers and smokers. This example is not adequate and should be removed, because in this case the difference in sensitivity of the two populations can be significant, of the order of 20 if the relative risk model is applicable.
Several organs (bone surface, bone marrow and skin) have been excluded from the sensitivity analysis (see paragraph 107). This is regrettable and the reasons for this exclusion should be better described.