View Report Details
Development of probabilistic models for prediction of the biokinetics of radionuclides in animals
Project Code: R01053
Kelly, M ; Thorne, M
The models used to represent the biokinetics of radionuclides in various animal species that are currently used by the Food Standards Agency (FSA) were developed in the early 1980s, based on reviews of literature published between the mid-1950s and the late 1970s. Since those models were developed, a substantial body of new literature has been produced and there is therefore a need to take account of this new information in updated versions of these models.
Furthermore, when the models currently used by the FSA were formulated, limitations of computing power and the methodological basis for assessments then in place resulted in an emphasis being placed on the development and application of deterministic models, in which best estimates of parameter values were used. Also, the models currently used by the Agency were developed using software coding techniques and quality control procedures that were appropriate at the time, but would not satisfy current guidance on good practice.
The basic objectives of the project were to develop, parameterise and implement a new set of biokinetic models that addressed the issues raised above. In a change to the original project specification, it was agreed that the biokinetic models should be implemented within the framework of PRISM, a computer code developed by Quintessa. PRISM provides facilities for probabilistic simulations of compartment models, and the implementation of the biokinetic models was undertaken under a separate contract between the Food Standards Agency and Quintessa. As a consequence, the objective to implement the biokinetic models fell outside the scope of the project reported herein.
Overall, the following was achieved during this study:
1. A literature review of issues pertaining to the transfer of radionuclides in animals was undertaken. This constitutes a record of the most recent data relating to radionuclide uptake and transfer by farm animals.
2. The literature review demonstrated that the reviews undertaken by Coughtrey and Thorne still provide, for many elements, definitive advice regarding uptake and transfer by animals that has not been superseded by later studies.
3. A set of biokinetic models were developed for the elements of interest. These models take account of the latest data on animal biokinetics, as set out in the literature review report. The biokinetic model for iodine is based on a detailed review of iodine biokinetics undertaken by one of the authors in the context of another study.
4. The individual biokinetic models were structured in an integrated framework, thus enabling the construction of an integrated computer model of animal biokinetics.
5. A complete parameter database was developed, including estimates of uncertainty on model rate coefficients and other parameters. The models were specifically designed to ensure that, as far as possible, uncertain parameters could be treated as independent.
6. The models were implemented individually within AMBER, to check that the models led to plausible output values and to act as a means of verifying the output from the implementation within PRISM.
7. Sensitivity studies were undertaken, to identify the key parameters (and hence those for which greatest efforts are required to reduce levels of uncertainty) for the new models.
Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: