Feasibility of the Social Media-Based Prevention Program "Leduin" for German Adolescents on Instagram: Mixed Methods Pilot Study.

Background: Digital platforms, particularly social media, including Instagram, present unique opportunities for health promotion among adolescents due to their widespread use with interactive features supporting high user engagement. However, the feasibility of effectively utilizing platforms like Instagram for health interventions requires careful consideration of adolescent engagement patterns. Objective: This pilot study evaluated the leduin program-designed to foster essential life skills and functional social media use among adolescents-while also exploring the broader feasibility of using Instagram to deliver complex social and psychological interventions in this population. Methods: The study adapted the feasibility framework by Bowen et al and used a mixed methods approach. Quantitatively, Instagram interaction metrics of 99 participants (women: 62/99, 63%; men: 37/99, 37%; aged 14-18 years; mean age 15.2, SD 0.74 years) were analyzed descriptively (means, medians, SDs) and inferentially (Welch ANOVA, Kruskal-Wallis, Pearson and Spearman correlations, linear and segmented regression analyses) using RStudio. Metrics included story views, retention rates, feature engagement (eg, polls, question stickers, quizzes), and drop-off rates. Recruitment efforts were also analyzed descriptively. Qualitatively, 13 postprogram semistructured interviews were conducted with 11 women (11/13, 65%) and 6 men (6/13, 35%; mean age 15.29, SD 0.99 years). Participants were sampled to reflect varying engagement levels (6 high, 5 medium, 6 low). The mean interview duration was 25 minutes 11 seconds (SD 6 minutes 34 seconds). Content analysis, with high intercoder reliability (κ=0.90), comprehensively explored participants' experiences and the program's impact. Results: Quantitative results indicated that the recruitment process was challenging, with 101 schools and 10 youth centers contacted, resulting in a participation rate of 12.8% (99/775 students). On Instagram, story views ranged from 34 to 81 per post, with an average daily retention rate of 87.7% (SD 7.8%). By week 4, 76% of the total drop in views had occurred (mean views declined from 66.1 to 53.4); by week 6, 97.3% of the drop had been reached (declined from 66.1 to 49.9 views), indicating sustained viewer interest over the 14-week program. Features requiring minimal user effort, including polls (mean 56.8%-54.4%), quizzes (mean 56.6%), and sliders (mean 51.2%), showed significantly higher interaction rates than more demanding features such as challenges (mean 21.7%) and question stickers (mean 20.6%; P<.001). Qualitative findings revealed that adolescents valued the program, its design and methods for its relevance to their daily lives, and its support in developing essential life skills. Suggestions for improvements were made. Conclusions: The study underlines the potential of various Instagram features and content posting schedules for health interventions to meet adolescent preferences and interests. Challenges with reaching the target group effectively emphasize the need for targeted recruitment strategies and optimizing initial content to boost engagement, underscoring the critical implications for prevention research and policy in leveraging digital platforms to enhance adolescent health.