In this study, we sought to evaluate a modification of the Michigan Alcohol Screening Test designed to include problems associated with other drug abuse/dependence besides alcohol. Scores of the lifetime Michigan Assessment-Screening Test/Alcohol-Drug (MAST/AD) were compared to other lifetime measures of substance abuse and dependence and to psychiatric scales reflecting current or recent symptoms. Two university medical centers with alcohol-drug programs located within departments of psychiatry hosted 520 patients with alcohol-drug-related diagnoses. Patients completed their own MAST/AD using a paper-and-pencil format. Based on interviews with the patient, a research associate rated the patients' substance-related problems on the Minnesota Substance Abuse Problem Scale (M-SAPS) and obtained information on lifetime treatment for substance abuse. An addictions psychiatrist determined abuse or dependence and made a current diagnosis of alcohol abuse/dependence only, drug abuse/dependence only, and alcohol plus drug abuse/dependence. The MAST/AD was highly correlated with the M-SAPS and several other measures of substance abuse morbidity. Patients with alcohol-only and drug-only diagnoses did not differ from one another on the MAST/AD, although both groups had lower scores than those with alcohol plus drug diagnoses. Current psychosocial morbidity as assessed by the patient and the psychiatrist was associated with the MAST/AD, although less strongly than with lifetime substance abuse measures. The lifetime MAST/AD demonstrates reliability as a severity measure for alcohol and/or drug abuse. With minor modification, this standard measure can be expanded from alcohol diagnoses to all substance diagnoses. This study in a clinical population did not demonstrate its utility as a screening instrument; additional work is needed to reveal its utility for this purpose.