GlycoQuest: An AI-Adaptive CGM-Integrated Serious Game for Pediatric Type 1 Diabetes

Type 1 diabetes diagnosed in childhood requires rapid acquisition of complex self-management skills while children and caregivers are also adapting emotionally and socially to a chronic condition. Standard diabetes education is often delivered early after diagnosis, is information intensive, and is not usually personalized to real-time glucose patterns, emotional state, child developmental stage, or family communication needs. This study will evaluate GlycoQuest, an AI-adaptive, continuous glucose monitoring-integrated serious game digital therapeutic for children and adolescents newly diagnosed with type 1 diabetes and their primary caregivers. GlycoQuest uses CGM-linked gameplay, a glucose-responsive digital avatar, AI-personalized learning pathways, parent-child collaborative modules, adaptive hypoglycemia and insulin decision simulations, and psychosocial adaptation content addressing diabetes distress, fear of hypoglycemia, stigma, illness acceptance, and autonomy transition. This randomized controlled trial will enroll 110 parent-child dyads. Children aged 6 to 17 years with type 1 diabetes diagnosed within the previous 3 months will be randomized 1:1 to GlycoQuest plus standard care or enhanced digital psychoeducation plus standard care. The intervention period will be 6 months, with follow-up to 8 months. The co-primary outcomes are change in CGM-derived time in range, defined as 70 to 180 mg/dL, and child diabetes distress at 8 months. Secondary outcomes include time below range, glycemic variability, HbA1c, severe hypoglycemia, diabetic ketoacidosis readmissions, insulin omission, quality of life, fear of hypoglycemia, parental stress, diabetes family conflict, self-efficacy, engagement, and implementation outcomes.