• Smoking status at 6 months after enrollment [ Time Frame: 6 months ] [ Designated as safety issue: No ]
  

• smoking status at end of treatment [ Time Frame: 7 weeks ] [ Designated as safety issue: No ]
  

Two hundred and seventy patients will be enrolled;approximately 135 will be randomized to each study arm over the course of 24 months. Patients will be considered smokers if they have used tobacco products during the week prior to admission and if they have smoked> or = 5 cigarettes /day during the previous year.Smokers will be identified by review of admissions and given a flyer to invite them to join the study. Interested smokers will be screened and asked to fill out a Beck Depression Inventory. Patients with very high Beck Depression Inventory Scores, i.e. > or = 30, will not be enrolled because such individuals may require other treatment. Each patient's physician will be contacted to verify his/her eligibility for the study. Once eligibility has been ascertained, a research associate will visit the patient to obtain informed consent and to enroll them in the study. Study questionnaires and a Fagerstrom test for Nicotine Dependence will be completed and reviewed by a research associate for completeness at the time of enrollment. The associate will put in the prescription for the study drug and request the order for dispensing of the drug. The research pharmacist will randomize the patient to Study Arm 1 or Study Arm 2.

Subjects randomized to Study Arm 1 will receive a standard 7-week course of sustained release bupropion (150 mg/day for the first 3 days, then 150mg BID), a self-help booklet, counseling on smoking cessation strategies, and follow-up phone counseling during the first 3 months after randomization. Study Arm 2 participants will receive the same intervention as Study Arm 1, but will receive placebo instead of active bupropion therapy. Both study groups will complete study questionnaires regarding their medical and smoking history and will be followed one week after their hospital discharge to record their blood pressure and to monitor possible side effects.

Since quit rates tend to decline over time, relapse prevention approaches will be included in both study arms. Marlatt's theory of relapse, an extension of Bandura's social learning model, will be used to augment the quitter's perception of self-efficacy. Participants will be trained to resist and cope with the temptations and stresses likely to be encountered after discharge from the hospital. Behavioral self-management techniques to counter known relapse-triggers such as stress, the presence of other smokers, alcohol use and depression will be discussed during follow-up counseling calls.

We will measure saliva cotinine in all participants who give a history of smoking abstinence at the end of treatment and at 6 months. Cotinine levels of > or + 15 ng/mL will be considered evidence of current tobacco use. Participants who are self-reported quitters, but have failed to provide a saliva sample will be considered as smokers unless verification from a spouse or significant other can be obtained. Participants who have died will be analyzed as smokers or quitters based on prior self-report, information in medical records, or interviews with next-of-kin. Others will be censored (i.e. excluded from the analysis). All cotinine samples will be assayed at the University of California, San Francisco,laboratory of Dr. Neal Benowitz.