Large Language Models in Nursing Simulation Training: Educational Impact, Practical Uses and Future Research Needs

In nursing and other health professions, umbrella and systematic reviews report that simulation can enhance knowledge, clinical performance, and confidence, especially when blended with other teaching methods (Cant & Cooper, 2010, 2017). Yet this evidence base is heterogeneous. Many studies are single-centre, use small samples, and apply diverse outcomes and designs that limit generalisability (Cant & Cooper, 2017; Issenberg et al., 2005; Zhang, 2023). Despite its pedagogical value, nursing simulation faces persistent challenges. Scenario development is time-intensive, faculty capacity for facilitation and debriefing is limited, and maintaining up-to-date, contextually relevant cases is difficult, particularly in high-acuity areas such as intensive care (Hravnak et al., 2007; Mould et al., 2011; Nagle et al., 2009). Access to high-quality simulation remains uneven across institutions, with additional constraints in regional and resource-limited settings. These pressures have driven interest in technology-supported approaches that can extend simulation capacity while preserving educational quality. the objectives of this scoping review are to map how ChatGPT and related conversational LLM tools are used to support simulation-based learning in nursing education, synthesise reported educational outcomes and the perceptions of nursing students, educators, and simulation instructors, describe the methodological characteristics and limitations of existing studies, and identify gaps, risks, and priorities for future research and practice in nursing simulation training. The primary review question guiding this scoping review is: How are ChatGPT-based tools used to support simulation-based learning in nursing education? Secondary questions include: What educational outcomes, benefits, challenges, or potential harms are reported when ChatGPT is used in nursing simulation training? What methodological approaches have been employed to design, implement, and evaluate LLM-supported nursing simulations? What gaps and priorities for future research are identified in relation to nursing simulation practice and education?