Rollnick and colleagues developed a model of brief motivational interviewing for use by nonspecialists. The method was developed through a study of a 30-40 minute brief intervention with male heavy drinkers in a hospital setting. Following a preliminary assessment of readiness to change, the interviewer selected from a menu of strategies in discussing topics or areas of conversation. Each item on the menu involved 5-15 minutes of discussion. Topics were selected and prioritized according to the interviewee's readiness to change. Later work by Rollnick and colleagues further developed brief interventions for other health behaviors. Opportunistic applications of MI in primary care and other health-care settings have been used with problematic drinkers and cigarette smokers. This paper focuses on a randomized trial of MI with young "early" drug users. It describes the intervention, with its effectiveness discussed elsewhere. The MI model was used with youth aged 16-20 who were using illegal drugs on more than an occasional basis. The intervention, which was classified as a pilot endeavor, adapted much of the intervention material from Rollnick et al. The conversation opened with a discussion of lifestyle, with attention to college studies, friends, family, and the use of leisure time; the opening strategy concluded with a brief inquiry about drug use. After the opening segment, the conversation turned to a discussion of all drugs being used, followed by a broader discussion of values and goals. College studies and career and other long-term aspirations were featured in the conversation. "Hypothetical" exercises were used for those who were unable to identify or readily articulate risks or problems associated with their drug use. A discussion of controlled drug use focused on self-monitoring and the use of self-help methods and materials. The conversation concluded with a discussion of life plans and making changes to facilitate the achievement of personal goals. A total of 105 interventions were delivered over a 9-week period during the year 2000. The intended core components were discussed with almost all intervention recipients. Efficacy data from this trial were evaluated and have been reported separately, as indicated in the references cited in this paper. 54 references