This study prospectively evaluated the influence of current electrocardiograms obtained at the time of emergency department presentation, as well as that of previous comparison electrocardiograms, on decision-making regarding coronary care unit admission of patients presenting with a chief complaint of chest pain or chest pain equivalent. Emergency department physicians were asked to commit themselves to recommending either coronary care unit admission or some other disposition, both before and after evaluating current comparison electrocardiographic findings. They were also asked, prior to reviewing these results, whether they thought information gained from the electrocardiograms would have any affect on their decision. Despite wide expectation that electrocardiographic findings would in fact affect decision-making, neither current nor comparison electrocardiograms virtually ever altered the ultimate decision of whether or not to admit. Faculty and house officers performed similarly in all regards, except insofar as attending physicians were less likely to expect electrocardiographic findings to help them in patients who were ultimately discharged. Emergency department nurses, who were asked whether they believed these patients needed admission to a coronary care unit on the basis of only a brief initial triage history, performed very similarly to the physicians. Thus, electrocardiographic findings are rarely if ever helpful in determining the need for admission to a coronary care unit in patients presenting to the emergency department with chest pain, and seem to have particularly little value in patients in whom myocardial infarction is considered clinically unlikely. Although physicians at all levels of training often feel a need to rely on electrocardiograms in patients they ultimately admit, greater experience allows more senior physicians to be comfortable in correctly discharging patients with no clinical evidence of disease without obtaining an electrocardiogram. Routine ordering of electrocardiograms in patients with chest pain in whom likelihood of significant acute ischemic pain is clinically low should be reconsidered.