Design Features Associated With Engagement in Mobile Health Physical Activity Interventions Among Youth: Systematic Review of Qualitative and Quantitative Studies

**Background:** Globally, 81% of youth do not meet physical activity (PA) guidelines, with lower adherence among youth from low socioeconomic position (SEP) families. Mobile health (mHealth) interventions are preferred by youth over in-person approaches, but long-term engagement remains a challenge. While design features associated with engagement have been studied in adults, evidence for youth is limited. This systematic review aimed to identify which design features are associated with engagement in mHealth PA interventions among youth aged 4–18 years.

**Methods:** A systematic search was conducted on June 2, 2021 (updated June 24, 2022) in EBSCOhost (MEDLINE, APA PsycINFO, Psychology & Behavioral Sciences Collection) and Scopus. Eligible studies included quantitative, qualitative, and mixed-methods research that reported an association between design features and engagement in mHealth PA interventions for healthy youth aged 4–18 years. Studies published between 2010 and 2021 in English were included. Screening and data extraction were performed by two reviewers (one-third double-coded; interrater reliability: titles/abstracts κ=0.73, full texts κ=0.54). Quality was assessed using the Mixed Method Assessment Tool (κ=0.63). Design features were coded using a framework developed by three digital behavior scientists, covering interface aesthetics, challenges, narrative, levels, feedback, monitoring, customization, reinforcement, navigation, goals, social, progress, and credibility.

**Key Results:** Twenty-one studies were included (total n=1443 youth; age range 3–18 years; published 2012–2021; from 15 countries). Interventions included mobile games (48%), mobile apps (38%), and text messaging systems (14%). Most studies (71%) measured only interest; 67% measured engagement as a primary outcome. Features positively associated with engagement included: clear interface with short, positive messages; tangible and intangible rewards; multiplayer gameplay and social interaction; variety of challenges with personalized difficulty; self-monitoring (especially via devices rather than manual input); self-set goals; customization options (avatars, personal goals, music); personalized real-time feedback; narrative elements; and gradually increasing levels. Features negatively associated with engagement included: poor technical functionality (a prerequisite); unclear or excessive text; robotic or click-associated sounds; leaderboards causing excessive competition; unclear instructions; poorly timed or excessive notifications; difficulty with virtual maps; manual input for self-monitoring; and repetitive challenges without progression. Only 29% of studies (6/21) included a substantial low-SEP population, and SEP definitions varied widely, preventing subgroup analysis. Behavior change techniques (BCTs) were referenced in 81% of studies, most commonly feedback (52%), social comparison (52%), goal-setting (48%), rewards (48%), monitoring (43%), and encouragement (29%).

**Clinical Implications:** Designers of mHealth PA interventions for youth should prioritize a clear interface, fair reward systems, social interaction features, and varied, customizable content. Features like competition, leaderboards, notifications, and manual-input tracking require careful implementation to avoid disengagement. Technical functionality is a prerequisite. The lack of research on low-SEP youth is a critical gap, as these populations stand to benefit most from effective interventions. Future studies should use experimental factorial designs, standardized engagement measures, and recruit low-SEP participants through schools to address health inequalities.