This study examined the sensitivity and specificity of the revised Self-Administered Alcohol Screening Test (SAAST-R)1 among adult outpatients evaluated at community addictions and psychiatry and psychology clinics. A total of 417 adults (mean +/- SD age of 41.4 +/- 15.0 years, 52% female, 91% Caucasian) completed the 36-item SAAST-R prior to their evaluation. Based on the clinician's documentation of the patient's alcohol use, two criterion groups were established: one composed of subjects classified as having no alcohol problems (n = 214), and one with alcohol problems (n = 203). The SAAST-R was found to have high internal consistency reliability (Cronbach's alpha = 0.96), sensitivity (90%), and specificity (82%) for differentiating outpatients with and without alcohol problems, respectively, using a cut point of >or=4 to indicate risk for alcohol-related problems. The SAAST-R was found to have good sensitivity and specificity among both males and females.