Bioassay Intercomparison Programs

Although, we offer other programs, the intercomparisons constitute our principal raison d'être. Currently, five programs for urinalysis are available once a year for the following radionuclides:

• Total Uranium
• Tritium
• Carbon-14
• Dual spike (Tritium and Carbon-14)
• Fission/Activation products

A urinalysis intercomparison for Pu-239/240 and Am-241 may be offered in the near future.

Analytical Performance Criteria

Measurement Bias

Relative bias is used as the indicator of how closely a reported measurement corresponds to the known radionuclide activity added by NCRC staff to urine samples distributed to participating laboratories for analysis. It is calculated as follows: First, each observation for a spiked sample is corrected for the appropriate urine blank analyzed on the same day. The result obtained is represented by y_jkl where:

j=the laboratory; j=1, 2, ..., n

k=the spike; k=1, 2, ..., 3

and l=the aliquot; 1=1, 2, ..., 5

The relative bias of the result for each aliquot (b_jkl) of a particular spike is calculated thus:

where a_k=the known value for a particular spike.

The relative bias (b_jk) for the mean of measurements on all the aliquots of a given spike is then obtained as follows:

where L=the number of replicate measurements (=5, in this intercomparison).

S-106 sets the minimum testing levels, MTL, for intercomparison participants at five or more times the acceptable lower detection limit stipulated in Table 2 of S-106 for various radionuclides commonly encountered in personnel monitoring. At levels equal to or greater than the MTL, the S-106 acceptable limits for mean relative bias are as follows:

- 0.25 <= mean relative bias and <= + 0.50

Table 2. Minimum Detectable Amounts for Selected Radionuclides

Radionuclide	In vitro (Bq/L)	In Vivo (Bq)
Hydrogen-3 (HTO)	400	N/A
Technetium-99m	N/A	5 x 104
Iodine-125	4	100
Iodine-131	4	100
Carbon-14	70	1.5 x 107 (lung)
Cesium-137	4	400
Americium-214	0.01	20 (lung)
Iron-59	N/A	500
Cobalt-60	5	500
Strontium-90	0.4	N/A
Zirconium/Niobium-95	N/A	400
Cerium-144	100	1 x 104
Natural Uranium	5 µg/L	4 mg (lung)

Relative Precision

For NCRC intercomparisons, the standard deviation of the bias is taken as the measure of the reproducibility of an analysis, and is referred to as "relative precision" in the reports. It is calculated as follows:

S-106 specifies the following performance criterion for Relative Precision for replicate measurements of a given radionuclide concentration:

|Relative precision| <=0.04

Methodology

Urine standards are prepared by Centre staff from human urine. A preservative is added to this matrix: Na - merthiolate to tritium, carbon-14 and dual spike samples, and concentrated hydrochloric acid to a concentration of 1% v/v to uranium and fission/activation products standards. A quenching agent may also be introduced into tritium samples to decrease counting efficiencies to the lower values occasionally encountered in employees' samples.

Most of the aqueous radionuclide solutions of appropriate concentration are prepared from U.S. National Institute for Standards and Technology (NIST) standards. In the case of the C-14 program, the Centre purchases C-14 labeled urea, which is normally found in human urine, from Amersham. Amersham provides the NCRC with a certificate of NIST traceability. Additionally, our staff cross-calibrate this standard against a USNIST C-14 hexadecane standard.

For the Uranium, Tritium, and Carbon-14 Programs, performance is tested at three radionuclide concentrations which are unknown to the participants. Precision is tested by providing five identical samples of each concentration. The vials are randomly labeled. For the Fission/Activation Products Program, the participants are given in randomly labeled triplicate samples, a combination of several (at least three) radionuclides which they have to both identify and measure. Radionuclides are chosen from the list given in Table 2 of Regulatory Standard S-106. When elements other than those in the list below are added to samples, feedback on bias and precision is provided, but participants cannot be passed or failed on performance for these particular elements. Surrogates may be used where radionuclides have extremely short half-lives (e.g., Ba-133 for I-131).

Some laboratories use the liquid scintillation spectrum of employees' urine samples to screen for internal contamination with fission products, the majority of which are also hard beta-emitters. These participants are allowed to report to the Centre their results for Gross Beta screening, using the Fission/Activation Products samples that we provide. Although, they are not graded on a pass-or-fail basis for this screening technique, feedback is given regarding reproducibility of their results and how these compare with those of their peers.

For the Dual Spike Program, participants receive randomly labeled triplicate samples which contain both tritium and carbon-14 in ratios that are varied from one sample distribution to the next.

The Centre provides urine blanks for each set of samples. Participants are asked to analyze one set of standards per day, (i.e., the urine blank plus the spiked sample or samples prepared from the urine matrix per day).

Participant results are recorded on a Report Form that is distributed with the samples. This form is returned to the NCRC within 30 days of sample distribution date. Feedback regarding actual spike concentrations are faxed to participants within a week following receipt of results Report Form.

Final performance reports are produced by staff and mailed to participants within 90 days of sample distribution. Starting next fiscal year, final reports will be available in electronic format upon participant request. With the exception of copies sent to the CNSC where licensees are identified, anonymity is maintained in the reports by assigning a code to each laboratory. Except for shipping costs, programs are offered free to Canadian laboratories.

Participants - Our Client Base

Our client base include the following groups which as seen on the attached map are fairly well distributed throughout Canada.

Uranium mines
Uranium mills and refineries
Nuclear power generating stations
Universities
Hospitals
Radiopharmaceutical companies
Governmental laboratories
Private laboratories

At the present time, the number of participants in each program are as follows:

Uranium (9)
Tritium (16)
Carbon-14 (9)
Dual Spike (10)
Fission/Activation Products (7)

BENEFITS OF PARTICIPATION

To the Regulator

Assurance of accuracy and reproducibility of measurements used for internal dose assessment is provided by an agency independent of the nuclear industry or commercial interests;

Homogeneity of samples warrantied by NCRC ensures that information provided regarding comparability of licensees' results was obtained using the same "yardstick" for tests;

Regularly offered programs provide information on the continuing capability of internal dosimetry services.

Comparability of Results

Shown on this graph is the relative bias performance of 12 uranium bioassay laboratories. Laboratories 1 to 6 use the Jarrell-Ash fluorometer; laboratories 21 to 24, either the Scintrex or the KPA laser phosphorimeter; and laboratories 31 to 32, ICP/MS. In addition to assessing whether a particular licensee is within S-106 acceptable bias limits, the figure demonstrates how the regulator can conduct an easy appraisal of how each licensee's results compares with it peers - regardless of technique or instrument used. Further, because the NCRC warrantees the homogeneity of the urine matrix used for preparing the intercomparison urine standards, the regulator can be certain that each participant is tested with the same 'yardstick' as the other participants.

To the Participants

• Independent agency provides external verification of accuracy of measurements;
• Allows laboratory to demonstrate measurement capability to employees, the regulator, or to customers;
• Urine blanks can be used to measure lower limits of detection (MDA) for different radionuclide analyses;
• Some laboratories have used the NCRC-spiked urine standards to calibrate their instruments and establish calibration curves.

Figure 1

To the Regultor, Participants, and Potential Customers of Commercial Laboratories

A multi-year analytical track record can be produced upon request for a particular laboratory from the NCRC's historical database.

Figure 2

A Laboratory's Track Record

A historical database is maintained for each participant of all intercomparison results submitted to the Centre to date, as well as information on measurement performance. The figures above shows an in-house uranium bioassay laboratory's track record for relative bias (Fig. 1) and for relative precision (Fig. 2). This type of information can be of great value to the regulator or a prospective client of a bioassay laboratory.