MiCARE: Designing an AI-Augmented Web App for Preventive Health Engagement Among Young Adults

Chronic diseases like diabetes, affecting 537 million globally and projected to reach 643 million by 2030, are preventable (IDF, 2021). Pre-diabetes is increasingly prevalent among young adults (18-34), digital natives are at risk due to lifestyle factors. Existing wellness apps targeting this group face 70% dropout rates within 100 days due to poor user experience (UX), lack of motivation, personalization, accessibility, and time as identified via user-centered design (UCD) and inclusive design lens. This TREO talk presents MiCARE, an AI-powered, empathetic web app developed through PhD research to enhance preventive self- management, to address critical gaps in preventive health interventions. Grounded in Design Science Research Methodology (DSRM) (Hevner et al., 2004), a Systematic Literature Review (SLR) defined MiCARE’s objectives to address attrition, motivation, and time constraints, highlighting personalization, gamification, and accessibility. A prototype developed using Figma tool integrates SLR insights featuring education, goal setting, gamification, and reminders via UCD and Inclusive Design. MiCARE incorporates AI methods such as a retrieval-based AI approach and Machine Learning (ML)-driven recommendations tied to user schedules and locations. Its AI-powered 24/7 chatbot, delivering pre-scripted, clinician-reviewed responses aligned with the CARE (‘Compassion’, ‘Assistance’, ‘Respect’, and ‘Empathy’) framework, supports wellness and prevents conditions like pre-diabetes. Expert reviews from clinical and technical professionals confirmed MiCARE’s feasibility, and it is now ready for pilot testing. Next, we will recruit participants to test a stable version, co-designed with young Victorians from urban and rural areas. Scenario-based use cases will demonstrate its relevance to real-world challenges, such as balancing busy schedules with healthy habits. Key metrics including usability, usefulness, and satisfaction will be evaluated using surveys guided by the Task-Technology Fit (TTF) and Unified Theory of Acceptance and Use of Technology (UTAUT) frameworks (Goodhue & Thompson, 1995; Venkatesh et al., 2003). Interaction sequences will inform predictive engagement models. Findings will be shared via relevant journals, conferences, and participant UX summaries. This TREO talk seeks feedback on optimizing UX and AI-integration for enhancing digital wellness engagement in young adults.