随着现代生活方式的快速变革，公众对重大慢病主动健康干预管理的需求日益增长。传统健康管理干预方式在应对多样化健康问题时显现出局限性，而基于多模态大语言模型的主动健康管理数字疗法为创新重大慢病管理干预技术提供了新的可能性。数字疗法利用大数据、人工智能等现代信息技术，为重大慢病干预管理提供个性化、远程、数据驱动的方案，成为提升管理效果、优化医疗资源配置、提高患者生活质量的有效工具。多模态大语言模型利用图文并茂的智能模型，成为“数字健康助手”，可帮助医生和患者解读医疗影像、监测日常健康数据，并进行对话式决策支持。该文旨在探讨基于多模态大语言模型的主动健康管理数字疗法融合，分析其应用、国内外研究状况以及面临的挑战及未来发展趋势，为主动健康管理提供理论支持和实践指导。

The rapid evolution of modern lifestyles has led to increasing public demand for proactive health interventions and the effective management of major chronic diseases. Traditional health management approaches have demonstrated limitations in addressing diverse and complex health needs， whereas digital therapeutics powered by multimodal large language models present transformative opportunities for advancing intervention strategies in chronic disease care. By integrating big data， artificial intelligence， and other cutting-edge information technologies， digital therapeutics deliver personalized， remote， and data-driven solutions that enhance the prevention， monitoring， and management of major chronic conditions. These innovations not only improve clinical outcomes but also optimize the allocation of healthcare resources and elevate patients′ quality of life. Multimodal large language models， acting as "digital health assistants，" enable intelligent integration of text and visual data to support medical image interpretation， real-time health monitoring， and conversational clinical decision-making for both clinicians and patients. This paper aims to investigate the integration of multimodal large language models into digital therapeutics for proactive health management， examining their applications， current domestic and international research advancements， existing challenges， and future development trends， thereby offering theoretical insights and practical guidance for the advancement of proactive healthcare systems.