Learning Objective #1: Understand how the knowledge engineering approach employed in this project may prove useful for handling knowledge representation problems

Learning Objective #2: Describe how the diagnostic nursing decision support system works and can help nurses who provide diabetes care

---

As Health initiatives in the United States accelerate adoption of electronic health records (EHR), nurse informaticists are challenged to create information systems and decision support tools that facilitate the nursing process. Historically, decision support systems for nursing have been limited by difficulties in defining and representing the nursing knowledge base. In this projects, knowledge representation issues of nursing diagnoses, specifically NANDA nursing diagnoses relevant to diabetes care, were addressed through knowledge engineering. A literature review established nursing diagnoses and clinical assessment criteria for patients with diabetes mellitus type II. Twenty-five NANDA diagnoses and 138 clinical assessment variables were structured in a criteria table. The criteria table was then used as the knowledge base for a prototype decision support system. The knowledge engineering approach employed in this project may prove useful for handling knowledge representation problems inherent in other nursing decision support systems. The executable knowledge module is appropriate for integration with existing HIS (health information system) or EHR (electronic health record) systems. Knowledge engineering manages the complex process of acquiring and managing knowledge to create executable (structured) knowledge models. Such structured models are necessary for data-driven, automated decision support systems, including diagnostic decision support systems for inpatient diabetes care. NANDA is the most highly developed standard taxonomy for representation of nursing problems, but practical issues of inaccuracy and inconsistency have arisen in clinical implementation. Workflow integration of nursing diagnosis via embedded decision support systems may ameliorate inaccuracy/ inconsistent clinical nursing diagnosis.