OBJECTIVE: To analyse the effects of smoking reduction and smoking cessation on incidence of myocardial infarction after adjustment for established cardiovascular risk factors. DESIGN: Prospective cohort study with record linkage to mortality and hospital registers. The association of individual change in smoking with myocardial infarction was examined in Cox proportional hazard analyses with continuous heavy smokers (> or =5 cigarettes/day) as reference. SETTING: Pooled data from three population studies conducted in Copenhagen, Denmark. PARTICIPANTS: 10 956 men and 8467 women with complete information on smoking habits at two examinations five to ten years apart were followed up from the second examination for a first hospital admission or death from myocardial infarction. Mean duration of follow up was 13.8 years. MAIN RESULTS: A total of 643 participants who were heavy smokers at baseline reduced their daily tobacco consumption by at least 50% without quitting between first and second examination, and 1379 participants stopped smoking. During follow up 1658 men and 521 women experienced a fatal or non-fatal myocardial infarction. After adjustment for cardiovascular risk factors, people who stopped smoking had a decreased risk of myocardial infarction, hazard ratio 0.71 (95% confidence intervals 0.59 to 0.85). Smoking reduction was not associated with reduced risk of myocardial infarction, hazard ratio 1.15 (95% confidence intervals 0.94 to 1.40). These associations remained unchanged after controlling for baseline illness in different ways. CONCLUSIONS: Smoking cessation in healthy people reduces the risk of a subsequent myocardial infarction, whereas this study provides no evidence of benefit from reduction in the amount smoked.