Abstract:

Background:
Artificial intelligence (AI)-supported clinical simulation technologies are increasingly integrated into nursing and midwifery education. These technologies enable dynamic scenario design, real-time performance analytics, and personalized feedback, extending the pedagogical potential of conventional and virtual simulation. Evidence from nursing education indicates that digitally enhanced simulation can strengthen clinical reasoning, communication, and procedural skills; however, evidence specific to midwifery remains limited. Existing reviews have emphasized learning outcomes and technological features, providing limited insight into the contextual and pedagogical conditions that explain effectiveness. The mechanisms through which AI-supported simulation produces variable outcomes across settings therefore remain insufficiently synthesized. This paper presents a systematic review protocol that aims to identify predictors, moderators, and mediating mechanisms influencing educational outcomes associated with AI-supported clinical simulations in nursing and midwifery education. Unlike prior outcome-focused reviews, this study adopts a theory-informed explanatory approach to determine how and under what conditions AI-enhanced simulation contributes to clinical competency development.

Methods:
The review will follow PRISMA 2020 guidelines and will be prospectively registered in PROSPERO. The protocol is developed in accordance with PRISMA-P and Joanna Briggs Institute guidance. A comprehensive, reproducible search will be conducted in PubMed, Web of Science, Scopus, ERIC, CINAHL, and PsycINFO using controlled vocabulary (e.g., MeSH) and free-text terms combined with Boolean operators and truncation. Eligible studies will be peer-reviewed empirical research published in English between 2015 and 2026 examining AI-supported clinical simulations in nursing or midwifery education. Quantitative, qualitative, and mixed-methods designs will be included. Editorials, dissertations, conference abstracts, grey literature, and studies without direct AI-supported simulation components will be excluded. Two reviewers will independently perform screening, full-text assessment, and data extraction using a piloted standardized form, with third-reviewer arbitration when required. Inter-reviewer reliability will be calculated using Cohen’s kappa. Methodological quality and risk of bias will be appraised using JBI Critical Appraisal Tools. Primary outcomes include validated measures of clinical competency, including clinical reasoning, communication, procedural performance, and self-efficacy. Secondary outcomes include learner engagement and satisfaction. Determinants, moderators, and mediators will be coded using a predefined theory-informed framework covering instructional design, faculty capability, institutional and technological readiness, learner characteristics. Where appropriate, random-effects meta-analysis with heterogeneity assessment will be conducted; otherwise, structured narrative synthesis and evidence mapping will be applied.

Expected Contribution:
This review will generate the first explanatory synthesis of mechanisms and implementation conditions shaping the effectiveness of AI-supported clinical simulation in nursing and midwifery education. Findings will inform evidence-based curriculum design, faculty development, institutional decision-making, and future research priorities in technology-enhanced health professions education.

Keywords:

Artificial intelligence, clinical simulation, nursing education, midwifery education, clinical competency, systematic review protocol, determinants, mediators, moderators.