Although admission and preoperative chest radiography has been done for many years in various settings, existing data do not support its utility in enhancing patient care. Calculations based on estimates of the accuracy of chest radiographs and the likelihood of disease suggest that routine chest radiography may result in many more misleading than helpful results. Patients in whom chest radiographs are likely to improve outcome are best identified by a careful history and physical examination. We recommend that the practice of doing routine chest radiographs on admission and preoperatively be stopped and that the procedure be reserved for patients with clinical evidence of chest disease and patients having intrathoracic surgery.