Objective Our purpose was to describe the mortality rate and mode of death over 10 years and factors associated with death among patients admitted to the emergency department with acute chest pain or other symptoms consistent with acute myocardial infarction (AMI). Methods All patients who came to the emergency department at Sahlgrenska University Hospital in Göteborg, Sweden, with acute chest pain or other symptoms consistent with AMI during a 21-month period were studied. Results In all, 5362 patients were registered, for whom information on 10-year mortality was available in 5158 (96.2%). In all, there were 2126 deaths (41.2%). Fifty-two percent of patients were ≤65 years old. Independent predictors of death registered on admission to hospital during the subsequent 10 years were age (relative risk 1.08, 95% CI 1.07-1.09), male sex (1.38, 1.25-1.52), initial degree of suspicion of AMI (1.13, 1.06-1.19), a pathologic initial electrocardiogram (1.76, 1.56-1.98), symptoms of congestive heart failure (1.66, 1.39-1.98), “other” nonspecific symptoms (1.22, 1.07-1.39), a history of diabetes mellitus (1.65, 1.44-1.88), a history of congestive heart failure (1.42, 1.26-1.60), a history of previous myocardial infarction (1.26, 1.12-1.40), and a history of hypertension (1.14, 1.03-1.26). For all these predictors there was a strong interaction with age, thus a much more marked influence on outcome among patients ≤65 years old than among patients >65 years old. When the above risk indicators were simultaneously considered, development of AMI during the first 3 days after hospital admission was still an independent predictor of death (1.63, 1.43-1.86). Conclusion For patients admitted to the emergency department with acute chest pain or other symptoms consistent with AMI, several predictors based on clinical history and clinical presentation are related to the 10-year prognosis. They are more strongly associated with outcome among patients aged ≤65 years. However, whether the patients have an AMI during the subsequent days will independently influence the long-term prognosis from observations on admission.