KAT: A Flexible XML-based Knowledge Authoring Environment

j J Am Med Inform Assoc. 2005;12:418–430. DOI 10.1197/jamia.M1701. In a recent report, the Institute of Medicine listed decision support as one of eight core capabilities that a successful electronic health record should incorporate to promote greater safety, quality, and efficiency in health care. At the heart of most modern decision support systems are two key components: an inference engine and a knowledge base. The success of any contemporary clinical decision support system is a function of the quality of the medical knowledge upon which it is built. The importance of a sound knowledge base in health care has been well detailed in the literature. At Intermountain Health Care (IHC), we are currently developing applications to systematically capture clinical knowledge for an enterprise-wide knowledge base. IHC is a nonprofit integrated delivery network with 21 hospitals, several outpatient clinics, more than 450 employed physicians, and a comprehensive insurance plan. The breadth and depth of the knowledge base content must meet the needs of the various clinician groups across the entire corporation. We are maintaining a strategy to create a knowledge management infrastructure that empowers clinical end users and fosters the alignment of their efforts with those of existing authoritative clinical groups. Given this strategy, the need for two fundamental applications emerged. One would be used primarily for the creation of knowledge content, while the other would serve as a gateway application that facilitates open review and collaboration amongst clinicians. This paper focuses on the development of the former, the ‘‘knowledge authoring tool’’ (KAT). Details about the initial prototype of the review and collaboration application can be found elsewhere.