Abstract
Patient care transitions across the fragmented healthcare continuum are one of the major challenges facing healthcare systems worldwide. Previous research on care transitions has identified anxiety levels as an important determinant of behaviour in patients with chronic conditions. Following hospital discharge, patients (and families) are often on their own, and the sudden transition can increase anxiety and stress, which has been shown to be associated with delayed recovery. On the one hand, emotional anxiety may be responsible for exacerbating patients' poor lifestyle habits, such as unhealthy diets, heavy alcohol consumption, and sedentary lifestyles, which in turn increase the risk of chronic diseases. On the other hand, people diagnosed with chronic illnesses may be prone to mood disorders and lack the ability to regulate their emotions on their own, requiring support from external sources. Prolonged anxiety in people with chronic illnesses can lead to a vicious cycle of "increased psychological anxiety - slow recovery from illness", highlighting the need for more comprehensive mental health support in the management of chronic illnesses. This study will develop an advanced large model for the psychological health management of chronic disease patients based on generative artificial intelligence models, intervene in the psychological anxiety of chronic disease patients, evaluate the effectiveness of alleviating their psychological anxiety, reveal the potential application of advanced large models in the field of healthcare, and lay a theoretical and practical foundation for constructing a transitional service mechanism for patient care.