Lu Natalie T, Taylor Bruce G
National Institute of Justice, Department of Justice, 810 Seventh Street, NW., Washington, DC 20531, USA.
Forensic Sci Int. 2006 Mar 10;157(2-3):106-16. doi: 10.1016/j.forsciint.2005.03.022.
Drug screening through urinalysis is a widely accepted tool for rapid detection of potential drug use at a relatively low cost. It is, therefore, a potentially useful method for detecting and monitoring drug use in a variety of contexts such as the criminal justice system, pre-employment screening and a variety of treatment centers. This article explores the efficacy of two commercially available drug-screening assays: Online KIMS assay (Roche) and EMIT II assays. First, we evaluate the sensitivity and specificity of two immunoassays. A total of 738 urine samples were collected among adult arrestee populations from Chicago, New Orleans and Seattle through the Arrestee Drug Abuse Monitoring (ADAM) program. Partial samples were split within one laboratory and analyzed by both enzymes multiplied immunoassay technique (EMIT) II and kinetic interaction of microparticle in solution (KIMS) assays for a 10-drug panel (amphetamine, barbiturates, benzodiazepines, marijuana, cocaine, methadone, methaqualone, opiate, phencyclidine and propoxyphene). Gas chromatography-mass spectrometry (GC-MS) was used as a confirmation method for all positives from either EMIT II or KIMS for all experiments. Second, the paper examines whether using different testing laboratories plays a role in the final results. The same experiments were repeated at two different testing locations: one in California and one in London and England. Third, the paper studies whether drug testing results vary between two laboratories when each of them had used their own routine screening method: the Forensic Science Service (FSS) at Birmingham, United Kingdom with KIMS assay and Medscreen Limited at London, United Kingdom with EMIT II. In summary, both EMIT II and KIMS assays generate fairly consistent results. The concordance rate against each of the 10 drugs tested is relatively high (97.4-100%). The discrepancies, in most cases, occurred at drug concentrations near the cut-off levels. There were more discrepant results between two laboratories compared to when specimens were analyzed at the same laboratory using two different assays.
