The Institute of Medicine report on the frequency and consequences of medical errors in clinical practice has stimulated physicians to evaluate current practice and means of improving medical care. In the treatment of patients with acute myocardial infarction, previous studies have found that dosing of fibrinolytic therapy is closely related to outcomes, with too low a dose associated with lower rates of infarct-related artery patency and higher doses associated with increased intracranial hemorrhage. Thus there is a narrow ″therapeutic window” for fibrinolytic-antithrombotic regimens, and the potential for adverse outcomes is high if incorrect doses are administered. The first demonstration of this concept came from the GUSTO-I trial, in which 13.5% of patients treated with streptokinase and 11.5% of patients treated with tissue plasminogen activator (t-PA) had a dosing regimen that deviated from the protocol, that is, an incorrect total dose or infusion length. In patients with protocol deviations, 24-hour and 30-day mortality rates were significantly higher compared with those of patients with per-protocol dosing: for t-PA, patients who received incorrect dosing had a 30-day mortality rate of 7.7% versus 5.5% for patients who received correct t-PA dosing (P < .001), with similar findings for streptokinase. More recent data from the InTIME-II trial have shown that the use of a bolus fibrinolytic agent significantly increases the percentage of patients who receive complete and optimally dosed fibrinolysis. Thus use of the simpler bolus fibrinolytic agents may reduce medication errors and thus may optimize clinical outcomes. (Am Heart J 2000;140:S154-60.)