Preparing for the Unpredictable: Integrating Simulation-Based and Digital Twin-Driven Learning Systems for Healthcare Professionals in Emergency and High-Risk Clinical Care Scenarios

Abstract

Healthcare professionals working in high-risk and emergency care settings are expected to make quick, high-stakes decisions, yet many existing training approaches struggle to fully prepare them for the complexity, uncertainty, and dynamic nature of real clinical environments. Traditional simulation-based training and digital twin technologies remain largely disconnected from healthcare training systems. This disconnect represents a critical gap in the preparation of a workforce expected to operate in increasingly complex, data-driven healthcare systems. This study examines how simulation-based and digital twin-driven learning systems can be combined to enhance the training of healthcare professionals in high-risk and emergency care scenarios, using a critical integrative review approach to synthesise diverse evidence. The findings reveal that simulation-based learning is widely effective in developing technical and non-technical skills, confidence, and experiential knowledge, but is limited by its lack of adaptability and real-time responsiveness. In contrast, digital twin technologies offer dynamic, data-driven, and predictive capabilities that support personalised and adaptive learning, yet remain underdeveloped in training applications. From this synthesis, four key themes emerged: including the need for dynamic and adaptive learning environments; the role of real-time feedback and predictive insights; the importance of continuous and personalised learning; and the influence of contextual factors such as infrastructure, data governance, and digital competence. Building on these insights, the study develops a novel conceptual framework that positions simulation and digital twins as complementary components of an integrated learning system. In this model, simulation provides experiential grounding, while digital twins enhance adaptability, intelligence, and real-time responsiveness. Together, they enable a shift from static training models to continuous, data-driven, and context-aware learning environments. Therefore, this study concludes that integrating simulation-based and digital twin-driven learning systems has the potential to transform healthcare training by making it more realistic, adaptive, and aligned with clinical practice. However, realising this potential will require empirical validation, investment in infrastructure and workforce capabilities, and careful attention to ethical and data governance considerations.