The quantification of uncertainties in internal doses assessed from monitoring measurements

A novel method is described for the assessment of total uncertainties in intakes and internal doses assessed from in vivo and bioassay monitoring measurements. Using the information on uncertainties in intake patterns, measurements and biokinetic model parameters, the probability distribution functions for assessed intake and dose were generated using Monte-Carlo simulations. The method was implemented using software written in MS Visual Basic 6.0. Preliminary results are presented for the example of routine tritium-in-urine monitoring. It was shown that the uncertainty in the assessed dose first decreases with increasing monitoring interval, reaching a minimum at approximately 14 d, and then increases as the monitoring interval is increased beyond 14 d. The distribution describing the ratio of assessed-dose-to-true-dose becomes very asymmetric at longer monitoring intervals. In principle, this method should allow realistic uncertainties to be placed on assessed internal doses.