We assessed the impact of three conditions on one-year smoking cessation rates. Physicians in 70 community general practices were randomly allocated by practice to one of three groups: In the usual care group, smoking patients were to receive the care they normally would receive. In the gum only group, physicians were asked to speak to patients about smoking cessation and offer nicotine gum. In the gum plus group, physicians were trained in the experimental intervention. This intervention involved advice to stop smoking, the setting of a quit date, the offer of nicotine gum, and four follow-up visits. Smoking cessation was measured by self-report after one year and validated using saliva cotinine measures. Using a criterion of at least three months of abstinence, 8.8% of the patients of the trained physicians had stopped smoking at the one-year follow-up compared with 4.4% and 6.1% of the patients in the usual care and gum only groups, respectively.