Protection of works and members of the public from ionising radiation is dependent on the assessment of radiation doses received. For radionuclides that have entered the body , doses must be determined indirectly using appropriate mathematical models that describe the distribution of radionuclides and the delivery of doses to body over time. For this purpose the ICRP publishes effective doses coefficients (effective dose per unit intake, Sv per Bq) for radionuclides that enter the body by inhalation and ingestion
Consideration of uncertainties on doses can provide numerical estimates of the reliability of the protection quantities (dose coefficients) used in radiation protection to assess exposures to radionuclides that enter the body by ingestion or inhalation (‘internal emitters’). The ICRP system of radiological protection is reviewed and the meaning of reliability is clarified. It is argued that the reliability of an effective dose coefficient as a protection device can best be determined by comparing the nominal detriment adjusted cancer risk associated with the dose coefficient, with a best estimate of risk for the exposure pathway and exposed population group, taking into account uncertainties in biokinetic, dosimetric and risk parameters.
The present work describes the application of parameter uncertainty analysis to quantify uncertainties resulting from internal exposures to 238U, 226Ra, 239Pu, 241Am, 137Cs, 90Sr, 131I, 129I, and 3H by members of the UK public, confining consideration to uncertainties in biokinetic models and parameter values. The report does not consider uncertainties in risk directly, but derives uncertainties in the biokinetic models that are used to calculate the retention and excretion of radionuclides in the body, in order to calculate distributions of effective dose per unit intake.
The central values and ranges of the distributions are used to inform the derivation of uncertainty factors (UF) for the different dose coefficients. A UF indicates a 95% probability of the risk coefficient being within a factor, UF, of the nominal risk associated with the appropriate ICRP dose coefficient, E50, with respect to uncertainties in the biokinetic model and parameter values. The inferred UF values are around 2 to 3 for ingestion and 2 to 6 for inhalation for all age groups. It is instructive to consider these ranges alongside the likely levels of exposure that are expected from the radionuclides considered (micro-sievert range) and the dose limit for planned exposures for members of the public (1000 micro-sieverts).