|SOFT - TIAFT 1998||Poster Session 4||Friday October 9, 1998|
"BLIND" CONTROL SAMPLES IN A PROFICIENCY TESTING PROGRAM|
Grete H. Wethe, Per Holmgren*
National Institute of Forensic Toxicology, P.O.Box 495 Sentrum, N-0105 Oslo, Norway
|Most of the proficiency testing programs have an obvious disadvantage in using control samples which are not concealed as real samples. The participants know that the sample represents a matter of control. Some control samples should therefore be sent from one of the ordinary requisitioners to the laboratory as if they were regular samples.
A collaboration making this possible has been established between the institutes of forensic medicine and forensic chemistry/toxicology in Sweden and Norway. By agreement, the Institute of Forensic Toxicology in Oslo sends an aliquot of an autopsy blood sample already analysed in the routine testing at the Institute in Oslo to the Institute of Forensic Medicine in Linköping for further transmission as if it was one of their own autopsy cases to the Department of Forensic Chemistry in Linköping. Similarly, the Department of Forensic Chemistry in Linköping sends a sample to Institute of Forensic Medicine in Oslo for transmission to Institute of Forensic Toxicology in Oslo.
The scheme was started in 1992, with 12 cases being processed up to now. To date 48 drugs (metabolites not included) were analysed and detected by both laboratories. 12 drugs were analysed by both laboratories, but detected by only one. The concentrations of these 12 drugs were near the cut-off-value. 8 drugs were analysed and detected by one laboratory only.
Fairly good agreement was found with respect to quantitative results. A lower result when performing the last analysis was seen in some cases where degradation of the drug is well known. Up to now the discrepancies between the two laboratories have been acceptable and would not have had important consequences for the interpretation given in the individual cases.