Artificial Intelligence Applications for Chronic Condition Self-Management: A Narrative Review (Preprint)

BACKGROUND Artificial intelligence (AI) has significant potential in promoting and supporting self-management in patients with chronic conditions. However, there is a lack of understanding of how current AI technologies apply to address specific chronic condition self-management issues. Evidence is needed to guide the development and selection of appropriate AI solutions to support self-management in patients with chronic conditions. OBJECTIVE This study aims to provide a narrative review of the literature on AI applications for chronic condition self-management, categorizing self-management components supported by AI technologies based on three essential self-management tasks: medical, behavioral, and emotional management, and identify the current developmental stages of AI applications for chronic condition self-management. METHODS A literature review was conducted for studies published in English between January 2011 and March 2023. Four databases, including PubMed, Web of Science, CINAHL, and PsycINFO, were searched using combined terms related to self-management and artificial intelligence. The inclusion criteria included studies focused on the adult population with any type of chronic conditions and AI technologies supporting self-management. RESULTS Of the 1288 articles retrieved from the search, 49 (3.8%) were eligible and included in this review. The most commonly studied chronic condition was diabetes (15/49, 30.6%). Regarding self-management tasks, the majority of studies aim to support medical (31/49, 63.3%) and/or behavioral self-management (20/49, 40.8%), but there is a lack of focus on emotional self-management (9/49, 18.4%). Conversational AI (13/49, 26.5%) and multiple machine learning algorithms (13/49, 26.5%) were frequently used types of AI technologies. Categorization of the technology development stage identified that most studies remain in the algorithm development or early feasibility testing stages. CONCLUSIONS The application of AI technologies in chronic condition management promises to empower patients to effectively perform self-management. AI technology has been widely applied in varied chronic condition self-management; however, most studies remain in the early stages. More studies are needed to generate evidence for integrating AI in self-management to obtain optimal outcomes.