Large language model-based diabetes continuation care intervention: A scoping review

Abstract

This review aims to synthesize current evidence on the application of large language models in continuity of care for patients with diabetes mellitus, to inform clinical health care professionals and support research in diabetes care continuity. Large language models were typically applied offline in conjunction with mobile intelligence platforms, enabling real-time responses to patient inquiries and effective health status management. Compared with traditional health education methods, AI platforms utilizing large language models demonstrated greater efficacy in enhancing patients’ self-management abilities, blood glucose control, and disease awareness. The application of artificial intelligence, large language models to support patient continuity of care after hospital discharge, remains in the exploratory stage. Within the domain of continuity of care, no studies have reported a specialized intelligent question-answering system designed specifically for postdischarge patient needs. Future research should focus on developing and validating more suitable AI tools for this purpose, leveraging artificial intelligence to address the limitations of current care models.