We validated diagnoses of acute myocardial infarction (AMI) and death from coronary heart disease (CHD) found in the Finnish National Hospital Discharge Register and the Register of Causes of Death from a sample of the 29,133 men participating in the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. The cases were traced to hospitals and institutes performing medico-legal death cause examinations and all relevant information was collected. The cardiac events were re-evaluated according to the diagnostic criteria of the Finnish contribution to the WHO MONICA project, i.e. the FINMONICA criteria. Altogether 408 cases of non-fatal AMI (n = 217) and death from CHD (n = 191) were reviewed. In the re-evaluation 94% of them (95% confidence interval 92-96%) were diagnosed as either definite (57%) or possible (37%) AMI. Non-fatal cases were more often classified definite AMI in the review, whereas fatal cases were more often classified possible AMI. Age or trial supplementation group did not affect classification, and no secular trend was observed. In conclusion, the diagnoses of AMI and death from CHD in the registers were highly predictive of a true major coronary event defined by strict criteria, thus their use in endpoint assessment in epidemiological studies and clinical trials is justified.