OBJECTIVE: To determine which method of intervention will best assist ED patients who smoke and are interested in quitting. METHODS: This was a prospective, controlled randomized survey of patients seen at a military ED with an annual census of approximately 30,000. All patients 18 years old or older who were medically stable and who admitted to smoking were eligible. Exclusion criteria included: predominantly psychiatric complaints, patients going directly to labor and delivery, those who left without being seen, those referred to acute care, and those who lived outside a 30 mile radius. Research assistants kept track of all patients who presented for care during four-hour time blocks representing peak census hours and asked all eligible patients if they smoked. Those who responded yes were asked if they were interested in quitting. Those who were interested were enrolled in the study and given a questionnaire. This instrument assessed demographic factors, attitudes towards quitting, and included the previously validated Fagerstrom Test for Nicotine Dependence. Once the questionnaire was complete, the subject was randomized into one of two groups. The first group was given no information or advice from the emergency department physician (EDP), but was referred to a formal smoking cessation program. The second group was not enrolled in any formal program, but was given brief counseling from the EDP. Those in the first group who completed the smoking cessation program and all patients in the second group were contacted by telephone for six-month follow-up to assess success in quitting. RESULTS: A total of 747 patients were seen during the study period, 423 of whom were eligible for the study. Of these, 86 (20%) admitted to smoking. Of these smokers, 42 (49%) were interested in quitting. These 42 patients comprised our study group and filled out questionnaires. The questionnaires showed that 35% of our study group stated they wanted to quit smoking within the next 30 days, 30% stated that they had never been advised to quit by a physician, and 43% were considered moderately to highly nicotine addicted. Twenty-one patients were randomized to each sub-group, but one was lost in each sub-group secondary to lost paper-work. This left 20 in each sub-group for data analysis. In the formal smoking cessation group, none completed the class. Of those assigned to the EDP intervention subgroup who were followed-up at six months, only one patient was able to quit smoking and stay tobacco free. However, this patient was not considered a success because she attributed her successful effort to the fact she was diagnosed with a CVA two days after enrolling in the study. Most patients who were contacted felt that other pharmacological interventions may have assisted their efforts. CONCLUSIONS: ED based smoking cessation programs in our population will likely require more than brief EDP intervention or referral to a smoking cessation program if they are going to be effective. Future studies could evaluate the possibility of initiating other medical interventions like nicotine replacement therapy, or coordination with the patient's primary care provider.