We assessed the practicality of using the Short Michigan Alcoholism Screening Test (SMAST) and the Alcohol Use Disorders Identification Test (AUDIT) in screening adolescents for alcoholism in a primary care setting. In addition, we sought to determine the prevalence of alcohol use among adolescents, 16-21 years of age, presenting to a private Family Medicine practice for medical care. A consecutive sample of 67 subjects presenting for medical care were asked to complete the SMAST and AUDIT questionnaires. Overall, 52 (78%) of the questionnaires were returned with complete data. Of the 52 patients, 25 (48%) admitted to drinking. Using a "positive" score on either the SMAST or AUDIT as a positive test for alcohol use yielded a sensitivity of 40% and a predictive value positive of 100%. Using a "negative" score on both the SMAST and AUDIT as a negative test for alcohol use yielded a specificity of 100% and a predictive value negative of 64%. Although alcohol use was relatively common considering the age group, using the SMAST and AUDIT to screen for alcoholism is labor intensive and is not practical in this situation. Because patients appeared to misinterpret some questions and were often accompanied to the office by their parents, their answers may not be valid. History of alcoholism taken upon typical office examination and relevant advice appears to be a better alternative to the use of questionnaires in determining the prevalence of alcohol use in this age group.