Abstract
The object of the paper is to provide the basis to decide whether testing for drug use should be used, taking into consideration individual and social benefits and costs that the results of the tests could originate. In the analysis, special attention is paid to the fact that the actual costs of the tests themselves are not the most important elements of the social costs of a testing program. The most important costs are those generated by the defective results of the tests. They can provide false positive results, meaning that a nonuser is identified as a user, or false negative results in which a user is not identified as such. Social costs of the false positive results range from lowered worker morale to legal suits, while the false negative results eliminate any benefit that the identification of drug users could have. Combining all these elements, Correa and Woods specify a mathematical procedure as a decision tool to be used for determining whether a testing program should be implemented in specific industries or groups of industries. A complete implementation of the model is not carried out due to the lack of the required statistical information. However, preliminarytrials showed that the conceptual and mathematical framework is operational, and that the model prepared could be used as a guideline for collecting the most appropriate data for decision making.