Design and evaluation of a knowledge-based clinical decision support system for the psychiatric nursing process

Abstract

Background and objectivesThe nursing assessment in the psychiatric department differ from those used in other departments considerably. We developed a psychiatric knowledge-based clinical decision support system (Psy-KBCDSS), which may aid nurses in solving patients’ problems in the psychiatric department. In addition, we compared the sensitivity and specificity for the nursing diagnoses between the psychiatric nursing process system (Psy-NPS) and Psy-KBCDSS to determine that the Psy-KBCDSS can assist nurses in performing the nursing assessment and diagnosis. MethodsVisual Studio 2019 was adopted as the primary software development tool, and C# as the main development language. The concept of the nursing process was applied to develop the Psy-KBCDSS user interface. We developed a clinical diagnostic validity inference engine to calculate the frequencies of the nursing assessment items and nursing diagnoses in clinical tasks in the Psy-NPS for generating a knowledge-based database of the Psy-KBCDSS. The sensitivity and specificity for nursing diagnoses formulated by senior and junior nurses were used to determining the effectiveness of adopting Psy-NPS and Psy-KBCDSS. ResultsThis study include 22 nursing diagnoses commonly encountered in psychiatric wards. The top eight most common diagnoses in the Psy-NPS and Psy-KBCDSS were altered thought processes, ineffective coping, sensory and perceptual alterations, insomnia, risk for other-directed violence, anxiety, impaired social interaction, and risk for suicide. Compared with the Psy-NPS, the Psy-KBCDSS had significantly higher sensitivity for sensory and perceptual alterations, ineffective coping, and insomnia and significantly higher specificity for ineffective coping. ConclusionsConsidering its high sensitivity and specificity for various nursing diagnoses, the Psy-KBCDSS, as an empirical patient-oriented nursing clinical decision-making support system, can assist nurses in clinical nursing tasks including nursing process–based patient assessment and nursing diagnosis.