Results of Radiobioassay Blind-Spike Pilot

*Thomas R. La Bone, Harriet P. Priester

Westinghouse Savannah River Company

From March to October 1999, 45 urine samples having a nominal concentration of 0.05 dpm/L of Pu-238 and Am-241 were submitted to the SRS Bioassay Laboratory. The samples were submitted in the names of real workers and the Bioassay Laboratory did not know which samples were spiked or the amount of the radionuclides present in the samples. The observed false negative rate of 16.7% for Pu-238 and 4.9% for Am-241 at a nominal concentration of 0.05 dpm/L corresponded to the MDCs for these radionuclides were 0.071 dpm/L and 0.064 dpm/L, respectively. Probit plots of the ratio of the reported result to the spike value were provided for Pu-238 and for Am-241. The results from both analyses are normally distributed and have almost identical population variances. The Pu-238 did not exhibit any bias as indicated by the median ratio of essentially 1 (1.06) with 2 high results and three low results. The Am-241 analyses had a median ratio of 1.32 with 5 high results and 0 low results. This indicates that there may be a slight positive bias for Am-241 analyses. A review of a limited number of higher-level spikes (0.15 to 1.0 dpm/L range) for the same time period does not indicate as large a bias as observed for the low-level spikes. The blind spike program has been adopted on a permanent basis for the SRS Bioassay Laboratory and may be expanded to include other radionuclides for which the laboratory performs analyses.

^{*This paper was prepared in connection with work done under Contract Number DE-AC09-96SR18500 with the United States Department of Energy.}