Despite the compelling relationship between early treatment and outcome from reperfusion therapy in patients with acute myocardial infarction, significant delays in early treatment are imposed by the patient, prehospital systems, and hospital processes and protocols used in the identification and treatment of patients with myocardial infarction. A survey instrument designed to determine the prevalence of hospital policies and protocols that might delay or expedite treatment with thrombolytic therapy in patients with acute myocardial infarction was completed by 524 hospital participating in the National Registry for Myocardial Infarction (NRMI). Participating hospitals had treated 17,646 patients with tissue plasminogen activator. The door to drug time for the entire population of patients treated at each hospital was available. Door to drug times were compared between those hospitals that had a positive response to a policy and those that had a negative response to that policy. Among respondent hospitals, thrombolysis was excluded by protocol in 34.4% for age above 75 and in 55% for presentation after 6 hours of chest pain onset. Furthermore, 29.4% of hospitals required routine laboratory testing other than electrocardiography (ECG), including chest x-ray, prior to determination of eligibility for thrombolysis. Door to drug times were shorter in those hospitals with prehospital 12-lead ECG availability, assessment of the 12-lead ECG by the emergency department nurse and physician as soon as it was available, and initiation of thrombolysis by the emergency physician (in patients with clear-cut ST elevation myocardial infarction) without bedside cardiology consultation. Door to drug times were longer in those hospitals in which predecision laboratory results were required, written informed consent was mandated, and drug was initiated in the cardiac intensive care unit rather than in the emergency department itself. Door to drug times were not significantly different in those hospitals with a designated chest pain center compared with those operating under a focused patient care protocol. We conclude that the earliest possible hospital treatment of acute myocardial infarction patients may be precluded by multiple components of emergency department policies and process, many of them inappropriate for safe, efficient, and effective identification and management of these patients.