OBJECTIVE: To compare the effects of three brief methods of reducing alcohol consumption among family practice patients. DESIGN: Patients randomly assigned to one of three interventions were assessed initially and at 3-, 6-, and 12-month follow-up appointments. SETTING: Family practice clinic composed of 12 primary care physicians seeing approximately 6000 adults monthly in a small urban community, population 40,000. PARTICIPANTS: Through a screening questionnaire, 134 men and 131 women were identified as hazardous drinkers (five or more drinks at least once monthly) during an 11-month screening of 1420 patients. Of 265 patients approached, 180 agreed to participate and 159 (83 men and 76 women) actually participated in the study. INTERVENTIONS: Three interventions were studied: brief physician advice (5 minutes), two 30-minute sessions with a physician using cognitive behavioural strategies or two 30-minute sessions with a nurse practitioner using identical strategies. MAIN OUTCOME MEASURES: Quantity and frequency (QF) of drinking were used to assess reduction in hazardous drinking and problems related to drinking over 12 months of follow up. RESULTS: No statistical difference between groups was found. The QF of monthly drinking was reduced overall by 66% (among men) and 74% (among women) for those reporting at least one hazardous drinking day weekly at assessment (N = 96). Men reported drinking significantly more than women. CONCLUSIONS: These results indicated that offering brief, specific advice can motivate patients to reduce their alcohol intake. There was no difference in effect between brief advice from their own physician or brief intervention by a physician or a nurse.