Nikolai Kiselev

Background: Digital interventions are an emerging and promising avenue for addiction prevention and mental health promotion, but their reach and use are often limited, and little is known about the factors associated with youth accessibility. SmartCoach is a life skills training program for addiction prevention where adolescents are proactively invited for program participation in secondary school classes. The mobile phone-based program provides individualized coaching for a period of 4 months and addresses self-management skills, social skills, and substance use resistance skills. This study examined sociodemographic and other predictors of program participation and program use. Methods: A total of 476 adolescents in 28 secondary and upper secondary school classes in the German-speaking part of Switzerland were proactively invited for participation in the SmartCoach program. Using generalized linear mixed models (GLMMs), we examined predictors of both program participation and program use at the individual and school class levels. Results: In total, 315 (66.2%) of the present 476 adolescents gave their active consent and provided the necessary information to be included in the program. None of the individual sociodemographic characteristics significantly predicted program participation, however, the participation rate was significantly higher in upper secondary school classes (84%) than secondary school classes (59%). The mean number of interactions with the program was 15.9, i.e., participants took part in almost half of the 34 possible interactions with the SmartCoach program. None of the baseline characteristics on the level of the school class significantly predicted program use. On the level of the individual, the univariate models showed that, compared to the reference category of 14-year-old students, program use was significantly lower for students who were 16 or older. Furthermore, participants with a migration background or an origin from a non-German-speaking country showed significantly lower program use. Finally, students with a medium level of perceived stress showed higher program use compared to those with a low level of stress. Within the final multivariate model for program use, only the variable “origin from a non-German-speaking country” remained significant. Conclusions: SmartCoach is an attractive offer for young people, in which two out of three young people who are invited in the classroom to participate do so. Among the program participants, the use of the program is acceptable, with an average of almost half of the content being worked on. There is potential for improvement in terms of recruitment, especially in school classes with a lower level of education. The most important starting point for improving program use lies in taking greater account of needs and wishes of students with non-German-speaking countries of origin.