ABSTRACT
Clinical decision support (CDS) represents a critical tool for improving the quality and safety of healthcare. CDS has been defined in many ways, but at its core, it is any computer-based system that presents information in a manner that helps clinicians, patients, or other interested parties make optimal clinical decisions. For the purposes of this paper, we will limit our attention to real-time, point-of-care, computer-based CDS systems, such as drug-drug interaction alerting, health maintenance reminders, condition-specific order sets, and clinical documentation tools. A substantial body of evidence suggests that, when well-designed and effectively implemented, CDS can have positive effects for healthcare quality, patient safety, and the provision of cost-effective care. [1 2]
Although the benefi ts of decision support are numerous, only a small number of sites in the USA have achieved signifi cant success with it. [3] A variety of challenges limit the wide adoption of CDS, but a critical one is the diffi culty of developing and maintaining the required knowledge bases of clinical content. [4 5] Effective CDS often requires extremely large knowledge bases of clinical facts (eg, drug-interaction tables). This
content must be engineered (or purchased) and must also be kept current as clinical knowledge, guidelines, and best practices evolve. [6] The knowledge bases that are developed must refl ect both universal best practices and local practices and needs. Knowledge management is an iterative process that involves both the creation of new content and the continuous review and revision of existing content.
