Emotion Regulation for Middle Schoolers (iTRAC+)

This study is a multi-phase implementation-optimization project guided by the ACCESS Center and the Consolidated Framework for Implementation Research (CFIR), unfolding across three sequential steps over approximately 36 months and culminating in a randomized pilot effectiveness trial. In the final step, classrooms and schools are randomly assigned, using a stratified allocation procedure, to one of three conditions: iTRAC+ (the digital emotion-regulation program paired with teacher training and classroom activities), iTRAC alone (the digital program only), or assessment-only control. Findings from this work are intended to inform a future, fully powered R01 implementation trial.

The intervention itself centers on iTRAC, a self-guided digital emotion-regulation (ER) skill-building program delivered via school-provided tablets or laptops. Its four modules are split into eight shorter segments, each 15-20 minutes long, delivered once weekly for up to 12 weeks. In the enhanced iTRAC+ arm, teachers receive additional training (kept under four hours, delivered in person and through asynchronous video) and lead classroom activities-such as small-group "emotion generator" games-after each digital module, reinforcing ER concepts through modeling, scaffolding, and praise. Both versions are embedded directly into the regular school day rather than delivered after school.

The rationale for the study rests on the idea that emotion regulation is a transdiagnostic mechanism tied to adolescent mental health, school functioning, and long-term adjustment, yet few evidence-based social-emotional learning programs specifically target ER, and teacher-focused ER training is often overlooked entirely. The original iTRAC program already demonstrated moderate effects (d = .44-.49) on ER outcomes along with strong feasibility and acceptability. Adding a teacher-training "coaching" layer is expected to strengthen uptake and durability of the digital content, addressing well-documented barriers to school-based intervention implementation such as cost, fidelity, and limited school resources.

Recruitment draws from public middle schools across school districts in Oregon. Step 1 recruits an eight-member Expert Panel of administrators, counselors, and teachers through district-level meetings to inform iterative qualitative interviews conducted across up to five rounds. Step 2 involves four to six classrooms from a single middle school, with teachers self-selecting into the project through faculty meetings and recruitment emails. Step 3 expands to six middle schools, enrolling 48 classroom teachers (16 per condition) and roughly 864 students, recruited with the help of school "champions," flyers with QR codes, and staff meetings. Consent follows a passive model for parents-opt-out letters and emails sent across two rounds over four weeks-combined with active student assent and active consent from teachers and administrators.

Assessment procedures scale up across the three steps. Step 1 relies solely on qualitative data: audiotaped and transcribed Expert Panel interviews of about an hour each, analyzed thematically through matrix displays and refined with input from the Community Partners and Advisors Committee. Step 2 adds baseline-only questionnaires for students (two 30-minute sessions) and teachers (about 60 minutes), administered via Qualtrics, along with post-intervention teacher feedback interviews. Step 3 introduces a full baseline and 4-month post-intervention assessment for both students and teachers across all three conditions, supplemented by brief acceptability and usability surveys for iTRAC+ participants, classroom observations (sampling 25% of sessions), and teacher-completed fidelity forms.

Evaluation is organized around the RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) and CFIR domains. Reach and adoption are captured through enrollment rates, demographic representativeness, and program completion; implementation is assessed via classroom observations and fidelity ratings covering activity occurrence, competence, and time allocation; effectiveness is measured through both proximal mechanisms (ER abilities) and distal outcomes (mental health, student-teacher relationships, school connectedness, and disciplinary actions); and maintenance is explicitly excluded from the current project's scope.

Analytically, Steps 1 and 2 rely on qualitative matrix and thematic analysis to guide iterative refinement of the program, while Step 2's feasibility is judged against benchmarks of at least 70% enrollment, greater than 75% completion, and at least 80% fidelity. The pilot trial in Step 3 uses a 3-group by 2-time (pre/post) design, analyzed with linear mixed models or analysis of covariance (ANCOVA) that account for student clustering within classrooms. The study is powered at .80-.83 to detect small-to-moderate effects-approximately d = .38-.40 at the student level and d = .557 at the teacher level-based on 48 classrooms and roughly 734 students expected to have complete follow-up data.