Eighty-nine of 178 consecutively admitted inpatients were administered the substance abuse sections of the Structured Clinical Interview for DSM-III-R--Patient Version (SCID-P). Patients also provided a urine sample for toxicologic analysis. In addition, primary clinicians assigned admission and discharge diagnoses. Test characteristics (e.g., sensitivity) for confirming a current diagnosis of psychoactive substance abuse were calculated for each measure and compared. Urine toxicology analyses and admission and discharge diagnoses were significantly less accurate in diagnosing psychoactive substance abuse than the SCID-P. This finding indicates that substance abuse is frequently not noted during routine admission and discharge assessments.