This literature review surveys research on workplace substance abuse conducted in accordance with methodologically sound principles. It evaluates individual studies for the appropriateness of their designs, methods of substance detection, generalizability of findings, and the appropriateness of their conclusions. An attempt is made to determine the prevalence of substance use, the characteristics of substance abusers, the types of jobs associated with high rates of substance abuse, and the consequences of using substances in the workplace. Findings are presented from both self-report and urinalysis studies in order to develop a range of prevalence estimates. The review concludes that nationally representative self-report studies largely agree on a five percent prevalence estimate for on-the-job use of marijuana and that large-scale urinalysis reports generally concur with self-report findings. In another study, the U.S. Department of Labor produced an overall drug use estimate of nine percent by compiling the detection rates of over 7,500 U.S. companies with drug testing programs. Demographic studies are cited which showed that age is the strongest predictor of substance use, followed by sex, education, race, and geographic region; the most likely drug users in the workplace are white male high school dropouts 18-34 years of age in the western United States. Finally, it is concluded that the consequences of substance use appear to include increased absenteeism, tardiness, and possibly accidents on the job. (NB)