We are proposing a four year study in which a randomized clinical trial will be conducted to assess the effect (3-. 6-, and 12-months) of physician-delivered advice versus physician-delivered patient-centered counseling on high-risk and problem drinkers in a primary care medical setting. A secondary objective of the study is the development of a structured educational program for training physicians to recognize and Intervene effectively with these patients. The study will enroll and assess 1350 adult male and female medical patients aged 21-70. High-risk drinking is defined for women as greater than or equal to 3 drinks per day on 3 or more days per week; for men as greater than or equal to 4 drinks per day on 3 or more days per week. Problem drinkers are defined by DSM-IIIR criteria for alcohol abuse or dependence. Physiologically dependent patients will be excluded using an alcohol dependence scale. The study uses a randomized design and will be conducted in 2 Internal medicine practice sites affiliated with the University of Massachusetts Medical Center. It will include 30 physicians who will be randomized to receive training (intervention conditions) or no-training (control condition). The training of the physicians to do alcohol assessment and Intervention will be accomplished through a structured program over a three month training cycle at the beginning of each of two years. Regularly scheduled patients will be screened by a telephone Interview-administered alcohol screening instrument embedded in a health habits survey. Patients screening positive for high-risk or problem drinking will complete a face-to-face interview just prior to their visit with their physician. Patients of intervention physicians who meet entrance criteria for the study will be randomized to the advice or counseling condition. Patients of control physicians will receive usual (control condition). All study patients will be contacted at 3-, 6-, and 12-months for a telephone interview intended to assess self-reported outcome variables. A collateral telephone contact with a spouse or other person named by the study subject will be made to corroborate the subject's self report at 6- and 12-months. Six outcomes were selected: change in mean alcohol consumption; negative consequences of drinking; number of high-risk drinking episodes; progression through stages of change; awareness of alcohol as a health risk factor; and perceive global health status.