OBJECTIVE To describe patient characteristics, pre-hospital delay, treatment, complications and outcome in patients with acute myocardial infarction admitted to hospitals in Germany. RESULTS The study was of prospective observational multicentre design. Those involved were consecutive patients with acute Q-wave myocardial infarction admitted within 96 h of onset of symptoms to 136 German hospitals between July 1992 and September 1994 (n = 14980, median age 66 (quartiles 57, 74) years, 68% male, 48% anterior wall infarction). Median pre-hospital delay was 170 (90, 475) min, with 17% arriving within the first hour and 61% within 4 h of onset of symptoms. The following patient groups had a short pre-hospital delay: males, those aged less than 65 years, those admitted at night or the weekend, those with a previous myocardial infarction, those in need of cardiopulmonary resuscitation, and those with a diagnostic first ECG. The first ECG was diagnostic in 67.6% of cases. Reperfusion therapy was used in 53%, with thrombolytic therapy in 51.6%. Median time from admission to initiation of treatment was 30 (20, 55) min. Respective rates of treatment with aspirin, nitrates, and beta-blockers were 81%, 83%, and 16%. Major complications were cerebral bleeding (0.4%), bleeding requiring transfusions (0.9%), left ventricular rupture (0.6%) and anaphylactic shock (0.1%). Median hospital stay was 20 (13, 26) days. In-hospital death rate was 17.2%. Increased hospital mortality was observed with female gender, an unknown or long pre-hospital delay, a diagnostic first ECG, anterior wall infarction, trauma or major operation within the last 14 days, renal insufficiency and malignoma. CONCLUSIONS 'Real-life' hospital mortality is much higher than previously reported in clinical trials. To reduce hospital mortality, the efficacy of thrombolysis should be increased by shortening the pre-hospital delay, and the use of concomitant therapy, especially beta-blockers, should be increased.