From Predictive Performance to Bedside Integration: Implementation Evidence and Human–AI Interaction in Nurse-Involved AI-Enabled Clinical Decision Support Systems in Adult Intensive Care Units: A Scoping Review

Artificial intelligence-enabled clinical decision support systems are increasingly proposed for intensive care because they can transform high-volume, rapidly changing patient data into predictions, alerts, risk classifications, and recommendations. Their clinical value, however, depends on more than algorithmic performance. Nurses continuously monitor patients, verify alerts, implement interventions, escalate deterioration, document care, and coordinate with multidisciplinary teams; therefore, the usability, trustworthiness, workflow fit, and organisational integration of these systems are central to safe bedside use. This project will conduct a scoping review and evidence map of empirical studies involving nurses in the design, evaluation, implementation, or clinical use of AI-enabled clinical decision support systems in adult intensive care units. It will identify system characteristics, nurse roles, implementation stages, implementation outcomes, human–AI interaction findings, work-system determinants, nursing-process effects, patient outcomes, and safety, ethics, and equity issues. Proctor implementation outcomes will be used to classify what implementation success has been assessed, and SEIPS 2.0 will be used to interpret how people, tasks, technologies, organisational conditions, and environments shape work processes and outcomes. The OSF project will provide dated, reproducible materials from protocol registration through final evidence mapping.