Abstract
Frequency and results of urine toxicology screens in a state-guided system of emergency rooms in a large city were compared with frequency and results of the same tests in the same population at a university-run intensive treatment psychiatric unit. Differences both in prevalence and in results were noted, with the emergency rooms testing significantly less patients than the intensive treatment unit and documenting a significantly lower rate of active substance abuse comorbidity to psychiatric disorders. Possible explanations for these differences are discussed, including clinical, attitudinal, and cost containment factors. These findings have a definite impact on treatment decision making, diagnostic precision, and validity of epidemiologic reports.