The electronic medical record: A challenge for computer science to develop clinically and socially relevant computer systems to coordinate information for patient care and analysis

Abstract

Creating a Patient Centered Information System (PCIS) or Electronic Medical Record System (EMRS)—the former terminology emphasizing the clinical purpose, the latter the missing product—is now viewed by the health care community as necessary to coordinate modern patient care in a manner that can control costs. However, despite 25 years of significant effort, the distance between the information systems available today in health care and what will be needed to fulfill this promise in the future remains very large. Success depends on achieving three goals that are generally in conflict: (1) giving the users the full scope of features and detail needed to create and use electronic clinical records for decision making, (2) providing the speed and reliability necessary for their online use, and (3) preserving system security and patient confidentiality. The vendors of the current generation of Hospital Information Systems (HIS) are being asked to supply these new capabilities before most have fully succeeded in more modest objectives. Their present software platforms have serious technological shortcomings. If a true EMRS is a priority, then more than a minor rearrangement of tradeoffs using present technology will be required to achieve it. Rather, a shift in perspective and a quantum leap in functionality are needed that markets, with their conservative momentum and incentives toward incremental change, do not encourage and that are beyond financial means of the vendors. To counter the drag of the status quo, it is in the U.S. government's (or any government's) best interest to facilitate needed research on identifiable new technologies. This will require a serious investment in medical informatics research, which must be supported under national or international, rather than corporate, auspices. An agenda for such research is outlined.