256 - A digital health game to prevent opioid misuse and promote mental health in adolescents: A randomized controlled trial

Publication Number: 3246.256

Lynn Fiellin, Dartmouth College, NEW HAVEN, CT, United States; Tyra Boomer, Geisel School of Medicine at Dartmouth, Waxhaw, NC, United States; Katie Haile, Geisel School of Medicine at Dartmouth, Hanover, NH, United States; Caroline M.. Barry, Geisel School of Medicine at Dartmouth, Hanover, NH, United States; Kaitlin Maciejewski, Yale University, New Haven, CT, United States; Tassos C. Kyriakides, Yale School of Public Health, New Haven, CT, United States; Haiyi Xie, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States

Background: Opioid misuse and mental health disorders are pressing, interconnected issues among adolescents, yet prevention and treatment gaps persist. Digital games offer a scalable, engaging way to address both risks.

Objective: To evaluate the impact of PlaySmart, a digital health game, on adolescent opioid misuse and mental health over 12 months.

Design/Methods: A randomized controlled trial (RCT) assigned adolescents (aged 16–19, with recent symptoms of depression or anxiety, or non-opioid substance use; no prior opioid misuse) 1:1 to PlaySmart or control digital games. PlaySmart includes 6 interactive storylines and 5 minigames targeting opioid misuse prevention and mental health promotion. Assessments at baseline, 6 weeks, 3, 6, and 12 months measured opioid outcomes (perceived risk of harm, self-efficacy, intentions, knowledge, attitudes) and mental health outcomes (PHQ-8, GAD-7, BAPS, emotion regulation, self-stigma, help-seeking). Mixed-effects models included group, time, and group × time, adjusting for baseline, sex, and grade. Moderation (sex, grade, family history, food insecurity for opioids and sex and grade for mental health) and mediation (early changes in knowledge for opioids and BAPS for mental health) were tested.

Results: We enrolled 532 students (mean age 16.6; 47% female; racially/ethnically diverse) from 15 high schools. For opioid outcomes: At 6 weeks, more PlaySmart participants perceived “great risk of harm” from opioids (26% vs 18%; p = .047). Compared to control, PlaySmart yielded higher opioid knowledge at 6 weeks (26.6 vs 24.4; p = .003), 3 months (26.4 vs 24.9; p = .03), and 6 months (26.3 vs 24.7; p = .03). Negative expectancies increased more at 6 weeks (+2.2 vs +0.1; p = .001) and 6 months (+2.5 vs +0.7; p = .006). Family history moderated opioid risk perceptions, and mediation showed early gains in knowledge/perceived harm explained negative expectancies at 6 months. For mental health outcomes: At 12 months, PHQ-8 scores were 1.12 points lower for PlaySmart vs control (p = .002); anxiety was unchanged. BAPS were consistently more favorable (p = .003), and early BAPS gains partially mediated PHQ-8 at 6 months (p = .023).

Conclusion(s): PlaySmart improved short-term opioid risk perceptions and produced sustained improvements in knowledge, attitudes, depressive symptoms, and help-seeking beliefs. Though modest, these effects in a large, diverse sample support PlaySmart as a scalable tool for adolescent substance use prevention and mental health promotion.