| |
 |
Labs Analyze Radioactive Samples in Emergency Response Tests Working under severe time pressure, government laboratories can analyze radioactive samples fairly quickly—in a matter of hours—but with variable accuracy, and sometimes relaxed quality control procedures, according to a published analysis of two federally organized tests of the labs’ emergency preparedness. The exercises, described in a recent paper,* were organized by the National Institute of Standards and Technology (NIST) and the Environmental Measurements Laboratory (EML), a Department of Homeland Security Science and Technology laboratory. Six laboratories, including federal, state and one foreign organization, participated in the NIST tests, and 17 in the EML tests. The tests were conducted to let government officials know how quickly data can be obtained from radioactive samples after an accident or attack, and how accurate the results are likely to be, as well as to identify opportunities for improvement. Decisions concerning the protection of emergency workers and public health, as well as the provision of food, shelter and medical care may depend on timely, reliable data. The tests evaluated the labs’ measurement capability and capacity, effect of time spent counting radioactivity, measurement accuracy (or traceability to standards), measurement confidence (or uncertainties) and operational issues. The NIST and EML tests involved different types of samples and analyses, but the results were consistent: The average measurement across all matrices and measurement types was within 30 percent of the expected value with 99 percent confidence. This compares to 2-4 percent in the regular NIST Radiochemistry Intercomparison Program (NRIP), which has a 60-day turnaround time. “The question is, is this good enough?” says NIST research chemist Kenneth Inn, lead author of the paper. A consensus has not yet been reached among laboratories and the government regarding what would constitute adequate radioanalytical measurements under emergency circumstances. The paper notes that some laboratories, in order to respond quickly, had to modify standard operating procedures, reducing quality control over the data. In addition, some labs underestimated the precision of their measurements, while others overestimated it. As experience is acquired in future exercises, improvements are expected in the labs’ capabilities. Based on the lessons learned so far, the paper lists a number of recommendations for improving response to a radiological emergency:
NIST is already carrying out some activities consistent with these recommendations. For instance, NIST hosts a yearly workshop to discuss the results of the exercises and techniques used by the participating laboratories. NIST also has workshops with participating laboratories focusing on measurement uncertainties. In addition, NIST is working with other federal agencies to help them build networks of "go to" laboratories and develop related certified reference materials and performance testing programs, and is seeking funding for a national emergency radioanalytical response preparedness exercise. NIST Tests The National Air and Radiation Environmental Laboratory (Environmental Protection Agency); the Waste Isolation Pilot Project Laboratory (Department of Energy); the Waste Sampling and Characterization Facility (DOE: Fluor Hanford); the New Jersey Department of Health; the Washington Department of Health; and the Institute of Nuclear Energy Research (Taiwan) participated in the NIST round of tests, which were open to federal, state, private and foreign laboratories. The laboratories received five duplicate samples of up to five matrices, specifically soil, water, air filters, urine and fecal samples spiked with radioactive elements. Participants were instructed to analyze and report within eight hours their measurements of gamma-ray emitting elements (cobalt, barium, cesium, europium), as well as measurements of alpha radiation (heavy particles) and beta radiation (light particles such as electrons) from other elements (strontium, thorium, uranium, plutonium, and americium). The results were variable. For instance, for gamma-ray emitters in water, all labs reported results within traceability limits. For air filters and soil samples, alpha radiation activity was reported within the traceability limit, but beta radiation measurements exceeded the limit. One laboratory performed radiochemical analyses of water within 8 hours; another laboratory returned results after 35 days. In another case, a calculation error on the urine samples resulted in values that were 60 percent too low, exceeding the traceability limit. Overall, the average percent difference from the NIST values across all matrices and measurement types was about 17 percent, while the average traceability limit for emergency response tests was about 32 percent. These results indicate that the average over many measurements from many laboratories would fall within about 32 percent of the expected value 99 percent of the time. NIST set statistical confidence levels and levels of acceptable bias and precision for radiobioassays of biological samples in accordance with existing ANSI standards, including ANSI N42.222and ANSI N13.30. DHS Tests The Environmental Measurements Laboratory (EML), the EPA’s National Air and Radiation Environmental Laboratory, and the Idaho National Engineering Laboratory (DOE) participated in the DHS test round, along with state laboratories in California, Colorado, Georgia, Indiana, Kansas, Louisiana, Missouri, New Jersey, New York, North Carolina, South Carolina, Texas, Washington and Wisconsin. EML asked laboratories to measure gamma-ray emitting elements and report their results “as soon as possible.” The labs were given two sets of air filters spiked uniformly with a combination of cobalt, cesium and americium. The labs responded in three to 28 hours. While the results for cobalt and cesium were generally “acceptable” (defined as most laboratories agreeing with each other), most labs could not detect americium because the activity levels were near the limits of detection capability. While most labs were fairly competent in their analyses, they reported having difficulty processing samples quickly and presenting the results in the required format. The reported values for the cesium and cobalt measurements ranged from 31 percent below to 26 percent above the EML value. These results are within the traceability limits for emergency response tests, and are consistent with the overall results from the NIST study. * Kenneth G.W. Inn, Iisa Outola, Svetlana Nour, Hiromu Kurosaki, Lynn Albin, Anna Berne. Radioanalytical Emergency Response Exercise. Journal of Radioanalytical and Nuclear Chemistry. August 2006. See also: A-Z Subject Index, NIST Home Page Created:
09/29/06
|