Reperfusion delay in patients with acute myocardial infarction leads to increased morbidity and mortality. We sought to measure the rates of reperfusion delay and to identify factors associated with reperfusion delay after arrival to hospitals. We included 360 patients who had acute myocardial infarction with ST-elevation or left bundle branch block on electrocardiogram and received reperfusion therapy from the three participating academic medical centers from 1997 to 2000. Through retrospective chart review, we collected data about time to reperfusion therapy, patient and hospital factors potentially associated with reperfusion delay. Factors independently associated with reperfusion delay were determined by logistic regression analysis. Median door­to­needle time was 60.0 minutes, and median door­to­balloon time was 102.5 minutes. According to recommendation of the American College of Cardiology/American Heart Association Guidelines, 226 out of 264(85.6％) of thrombolytic patients and 43 out of 96(44.8％) percutaneous transluminal coronary angioplasty(PTCA) patients experienced reperfusion delay. The significant factors associated with delay were type of reperfusion therapy, patient factors including hypertension and delayed symptom onset to presentation(>4 hours), and hospital factors including nocturnal presentation(6pm∼8am), weekend, and an individual hospital. A significant proportion of patients experienced reperfusion delay. The identified predictors of reperfusion delay may help design a hospital system to reduce the delay in reperfusion therapy