Computerized clinical information systems clearly have a role in this era of managed care when outcomes research and cost/benefit analyses are becoming crucial. Despite anesthesiologists' leadership in developing physician-entry systems, automated recordkeeping systems have been underused. This report reviews the problems and possible solutions associated with establishing more effective and user-friendly systems in the anesthesia specialty. A key feature of any cost/benefit analysis or outcome study is the precise definition of data to be collected. Redesign of anesthesia information systems will be required to assist users to enter events according to standardized definitions. To filter the vast amount of data collected by electronic medical-records systems in the ordinary course of care that are not applicable to a specific study, some form of filtering or data reduction on transfer to research or administrative databases will be needed. To allow careful analysis of possible correlations of outcome to care choices requires both the capture of the clinical context-a detailed description of all relevant conditions extending well beyond merely the objective vital signs-throughout a specific medical episode and the establishment of postoperative evaluation systems to allow outcomes capture. Connections to new as well as existing outcome data will provide vast new opportunities for outcomes research.