Gamified and artificial intelligence–enabled mobile health applications for eating behaviour and weight-related outcomes in children and adolescents: a systematic review

Abstract

Pediatric overweight and obesity are rising globally, including in Indonesia. Mobile health (mHealth) applications using gamification, artificial intelligence (AI)-based personalization, or both have been proposed for this group, but the evidence is unsynthesised and unclear. This study aimed to synthesize their effects on eating behavior and weight-related outcomes and describe engagement. Methods:  The following PRISMA 2020, four databases were searched (January 2016–July 21, 2026) with citation searching. Eligible studies enrolled participants aged 5–19 years, evaluated an mHealth application with gamification and/or AI-based personalization, and reported eating behavior or weight-related outcomes. Screening, extraction, and risk of bias assessment (RoB 2; NIH pre-post tool) were duplicated. The synthesis was narrative. Results: Five studies (four RCTs and one single-arm study; 486 participants) were included. Two interventions used conversational agents only, two used gamification only, and one used both; none was fully integrated. The dietary game increased simulated healthy food choices immediately after play (2.48 vs. 1.10; p<.001; d=1.25). Diet-tracking and gamified lifestyle applications produced no significant dietary changes, and none improved the BMI-SDS or z-score. Engagement metrics were non-comparable and showed a decline. Three RCTs had a high risk of bias. In conclusion, these applications are promising but not yet proven; current evidence is insufficient rather than negative. Adequately powered, registered trials with standardized outcomes and longer follow-ups are needed, including in Indonesia.