In July 1987 a prehospital emergency medical service (EMS) was activated in Verona (Italy) and a broad educational campaign was introduced. Prehospital care is delivered by emergency physicians and/or qualified nursing staff, who travel by ambulance or helicopter and have radio contact with the hospital alarm centre. During a 1-year period before the activation of the EMS, 476 patients with acute myocardial infarction (AMI) were admitted to the coronary care unit (CCU) of Verona, with a median delay time of 4 hours. In the period between July 1990 and June 1991, 412 patients were admitted, with a median delay time of 3 hours. Age, gender, previous AMI and infarct location were not related to delay time. In the second period, 34% patients used the EMS, while 66% used their own transport. In patients who used the EMS, median delay time was 2 hours (1 hour and 20 min shorter, p < 0.01, than in patients who did not). The time between symptoms onset and reaching the decision that medical care should be sought (1 hour and 18 min) was the longest component of the total delay time. The time from EMS call to hospital arrival was 25 min and the time which elapsed in the Emergency Department before reaching the CCU was 15 min. In these patients, decision time and Emergency Department time were significantly shorter (p < 0.01) than in patients who did not use the EMS. We conclude that the EMS is effective in reducing delay time in patients with AMI.