PURPOSE To assess the difference in immediate (30 and 60 days after admission) and late (2-year) mortality between those who received inpatient post-myocardial infarction smoking cessation counseling and those who did not receive counseling. METHODS We conducted an observational study of a national random sample of inpatients from 2971 U.S. acute care hospitals participating in the Cooperative Cardiovascular Project in 1994-95. Medicare beneficiaries who were current smokers over age 65, admitted with a documented acute myocardial infarction, and who were discharged to home were included (n=16743). Our main outcome measures were early (30-, 60-day) and late (1-, 2-year) mortality. RESULTS Smoking cessation counseling was documented during their index hospitalization for 41% of patients. Compared with those not counseled, those who received inpatient counseling had lower 30-day (2.0% vs. 3.0%), 60-day (3.7% vs. 5.6%), and 2-year mortality (25.0% vs. 30%) (logrank P <0.0001). After adjustment for demographic characteristics, comorbid conditions, APACHE score, and receipt of treatments including aspirin, reperfusion, beta-blockers, and angiotensin-converting enzyme inhibitors, those receiving counseling were less likely to die within 1 year, but the effect was lost between 1 and 2 years [hazard ratio (HR) = 0.99 (0.91-1.10)]. The greatest reduction in relative hazard (19%) was seen within 30 days [HR = 0.81 (95% confidence interval 0.65-0.99)]. CONCLUSION Immediate and long-term mortality rates were lower among those receiving inpatient smoking cessation counseling.