سامانه همایش های دانشگاه آزاد اسلامی واحد نجف آباد

صفحه اصلی / اولین همایش بین المللی تحول خدمات سلامت: تلفیق مراقبت، درمان، بازتوانی و امیدبخشی

Advancing Nursing and Clinical Care Quality With Artificial Intelligence in Acute Hospital Settings

نویسندگان :

Pedram Riahi¹ Zeinab Shakerin²

1- دانشگاه آزاد اسلامی واحد نجف آباد 2- دانشگاه آزاد اسلامی واحد نجف آباد

کلمات کلیدی :

artificial intelligence،nursing informatics،clinical decision support،predictive analytics،patient safety،acute care،machine learning،quality of care

چکیده :

Introduction: Acute hospital care places substantial cognitive and operational demands on nurses and clinicians, who must detect deterioration early, communicate across disciplines, sustain effective working relationships, and make rapid decisions under pressure. AI has been proposed as a tool to support these functions, yet its value in nursing practice ultimately depends on whether it improves timeliness of action, reduces avoidable harm, and fits the realities of bedside care. This review synthesizes recent evidence on AI in acute hospital settings, focusing on clinical effectiveness, implications for nursing, implementation barriers, and future priorities.(1,2,3,4) Methods: We conducted a narrative review of literature indexed in PubMed/MEDLINE, Scopus, CINAHL, and Web of Science between 2015 and 2026, supplemented by selected influential earlier studies. Priority was given to systematic reviews, umbrella reviews, scoping reviews, meta-analyses, and multicenter clinical studies involving hospitalized patients. The evidence was structured across five domains: decision support, patient safety and early warning, nursing workflow and documentation, ethical and human factors considerations, and implementation challenges. Results: Current hospital AI use includes EHR-integrated predictive modeling, natural language processing for documentation, and tools for planning and scheduling.(2,4) In some settings, machine learning systems identified deterioration earlier than rule-based scores and supported earlier recognition and escalation; however, benefits in routine practice were inconsistent and depended on workflow integration, data quality, and the burden of false-positive alerts.(2,4) Sepsis prediction models frequently demonstrated good discrimination, but their clinical impact varied with alert threshold, care pathway, and local case mix.(4,5) Documentation and information-retrieval tools showed promise for reducing administrative workload and improving access to information, although evidence of durable time savings or burnout reduction was limited.(1,3) Trust, interpretability, nurse engagement, fairness, and governance emerged as central determinants of adoption and safety.(1,3,5) Discussion: The evidence suggests that AI can strengthen acute hospital nursing and clinical practice when it is used to augment, rather than replace, professional judgment.(1,3,5) At present, the literature remains limited, inconsistent, and largely retrospective, with the strongest signals in early warning, risk stratification, and selected documentation functions.(2,4,5) Sustainable progress will require nurse-centred design, prospective evaluation, AI literacy, and robust governance within a broader patient safety framework.(1,3,5)