The accuracy of simultaneous testing for two or more drugs of abuse is analyzed. Probability theory and drug-testing accuracy concepts applicable to the testing for multiple drugs are reviewed, these concepts are applied to laboratory proficiency and transportation drug usage data, and accuracy levels are estimated that could occur intransit agency drug testing programs that simultaneously test for five different abused substances. The finding of this analysis is that as the number of drugs tested for increases, the probability that a positive test result is erroneous (the false accusation rate) increases significantly. For example, the false accusation rate when testing for five drugs is about 4 times the false accusation rate when testing for one drug. Therefore, it is suggested that if transit system decision makers wish to obtain certain maximum false accusation rates at their own organizations, they must adapt laboratory sensitivity and specificity rates for the number of drugs actually being tested for. This paper appears in transportation research record no. 1297, Public transit research: management and planning 1991 .
Abstract