Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Enhancing healthcare providers’ advance care planning competence with large language models: protocol for the development of an AI chatbot and its evaluation in a randomised controlled trial
2
Zitationen
8
Autoren
2025
Jahr
Abstract
INTRODUCTION: Advance care planning (ACP) can support individuals to express their autonomy in the decision-making process for future care. Traditional ACP training for healthcare providers faces significant challenges related to interactivity, accessibility, scalability and sustainability. Cutting-edge generative artificial intelligence (AI) holds promise in enabling intelligent and interactive chatbots for ACP. The present protocol outlines the development and evaluation of a large language model (LLM)-based ACP chatbot for healthcare providers. METHODS AND ANALYSIS: The development of the LLM-based ACP chatbot will follow four stages: construction of dialogue data sets, fine-tuning, multi-LLM orchestration and ablation studies. A randomised controlled trial will then evaluate the LLM-based ACP chatbot's effectiveness in enhancing ACP competence among healthcare providers. A total of 66 healthcare providers will be recruited from China and randomly assigned (1:1) to either: (1) The LLM-based ACP chatbot intervention or (2) An ACP knowledge manual. The primary outcome will be ACP competence, while secondary outcomes will include (1) ACP knowledge, (2) Attitudes/beliefs, (3) Practice willingness, (4) Readiness, (5) Self-efficacy, (6) Processes of change, and (7) Decisional balance. Both primary and secondary outcomes will be assessed to evaluate the immediate impact (postintervention) and short-term impact (3-month follow-up and 6-month follow-up) of the chatbot on ACP. ETHICS AND DISSEMINATION: The research was approved by the Ethical Review Board, Xiangya School of Nursing, Central South University (E202442). Study modifications will be discussed among the research team members until a consensus is reached. Amendments reflecting study modifications will be submitted for institutional review board approval at all sites, updated on the clinical trial registry, and fully detailed and explained in the manuscript reporting the results of the study. All participants will provide written informed consent. The study will be conducted according to the principles outlined in the Declaration of Helsinki. The results of this study will be submitted for publication in peer-reviewed journals and presented at (inter)national conferences. TRIAL REGISTRATION NUMBER: ChiCTR2400091022.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.700 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.605 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 8.133 Zit.
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
2019 · 6.873 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.