Effectiveness of AI-Enhanced Nursing Assessment in Critical Care: A Randomized Controlled Trial

Abstract

Background: Integrating Artificial Intelligence (AI) into critical care nursing holds the promise of enhancing clinical decision-making through real-time feedback and predictive notifications from physiological and behavioural signals. The clinical effectiveness and nurse-led uptake of AI in ICU environments are yet to be fully explored. This RCT will examine whether AI-enhanced nursing assessments improve early detection of patient deterioration, response times among nurses, and general patient outcomes in critical care. A parallelgroup, two-arm RCT was implemented at a tertiary ICU, recruiting 200 critically ill adults (100 in the control group and 100 in the AI-assisted intervention group). The intervention was the incorporation of an AI-powered decision support system that processed continuous physiological data to alert for clinical deterioration. Primary outcomes were time to detection, length of stay in the ICU, and 30-day mortality. Secondary measures included nurse response time, usability and acceptance of the system, measured with structured tools and qualitative interviews. The AI-supported group exhibited notable improvements on a number of important outcomes. The mean length of stay in the ICU was decreased by 1.8 days (p < 0.05), detection of deterioration time was shortened by 3.4 hours (p < 0.05), and 30-day mortality was reduced by 33% (p < 0.05). Intervention group nurses also registered quicker response times and had high system usability and acceptance ratings (mean System Usability Scale score of 86.5/100, perceived usefulness score of 4.4/5). Open-ended feedback noted that nurses perceived the AI system as an auxiliary tool that supplemented their clinical judgment instead of diminishing it. This research clearly demonstrates that AI-enabled nursing assessment tools can lead to meaningful improvements in clinical outcomes and response times, along with being easily adopted into current ICU nursing workflow. The positive sense of usability for nursing personnel and belief in the AI tool's ability to enhance clinical decision-making suggests that involving nurses when designing AI into health care systems has value. These contributions point to the possibility of optimization of critical care utilization of resources by adopting AI in big-way approaches to integrate AI in health care systems - especially considering the significant shortage of resources (including human ones) in care settings around the world. Further studies which are multi-center and longer-term are needed to confirm the findings of this study and the long-term influence of an AI integration to regulate health care systems.

Ähnliche Arbeiten