If physicians and other health-care providers were able to facilitate long-term cessation in 1 of every 10 smokers--a safe estimate based on the outcomes of randomized physician-delivered intervention studies--almost 4 million smokers would become former smokers. Evidence from randomized trials and observational studies demonstrates that physicians who intervene with smokers have an impact on their cigarette-smoking behavior. Greater smoking cessation occurs as physician interventions become more intensive and contacts more frequent. Adding modalities such as counseling, self-help booklets, and nicotine-containing chewing gum appears to augment the effectiveness of advice alone. The consistent demonstration of the need for follow-up and the possible use of several modalities is not unlike the demonstrated components of effective overall management approaches to such medical problems as hypertension. Surveys have indicated that physicians often do not intervene because they experience a low success rate with smokers. Educational programs have demonstrated substantial success in training physicians to counsel smokers; therefore, new goals for medical education, house-staff training, and continuing medical education need to include the development of skills and office management practices that can be used by providers to facilitate health-promoting behaviors among their patients. This article reviews the impact of physician-delivered smoking interventions on smokers, physician attitudes toward intervention, and physicians' reported intervention practices. It identifies those strategies that have been demonstrated to be effective in an out-patient office setting and emphasizes the need for every smoker to be considered a candidate for intervention.