Promoting Emotional Well-Being In Schools: Evaluating the Everybody Matters Intervention Using Multi-Year Administrative Data

Background/Purpose: Rates of child and adolescent mental health concerns continue to rise, while access to effective and scalable interventions remains limited. School-based services offer a promising avenue for addressing unmet need, yet evidence on effectiveness across diverse, real-world contexts remains mixed. This study evaluates the Everybody Matters (EM) intervention, a trauma-informed, relationship-centered program delivered through an intern-based service model designed to increase access and scalability. The study examines whether participation in EM is associated with improvements in student coping and emotional functioning and explores variation in outcomes across student and contextual characteristics.

Methods: This retrospective, multi-cohort study used de-identified administrative data collected between the 2018–2019 and 2023–2024 academic years. The analytic sample included 4,195–4,991 K–12 students across 112 schools in 16 districts. Multi-informant pre- and post-intervention assessments (caregiver and teacher reports) measured five domains of coping and emotional functioning using a 5-item Likert scale (α = .83). Linear mixed models (LMM) were used to estimate change over time while accounting for repeated measures within students. Models controlled for rater type, gender, caregiver language, grade level, district, number of sessions, and intervention timing relative to COVID-19.

Results: Results indicated a significant main effect of time, with post-intervention scores significantly higher than pre-intervention scores (β = –2.07, p < .001), reflecting meaningful improvements in coping and emotional functioning. Caregivers reported greater improvements than teachers (β = 1.51, p < .001). Male students demonstrated slightly higher scores than females (β = 0.36, p < .001), while caregiver language was not a significant predictor. District-level variation was observed, suggesting contextual influences on outcomes. Grade level was positively associated with scores (β = 0.10, p < .001). Session count was negatively associated with improvement (β = –0.02, p = .027), and students served pre-COVID had lower scores than those served during or after the pandemic (β = –0.72, p < .001).

Conclusions and Implications: Findings provide initial empirical support for EM as a scalable, school-based intervention associated with improved student coping and emotional functioning across diverse settings. Results underscore the importance of multi-informant assessment and highlight meaningful contextual variability in implementation. Future research should incorporate comparison groups, fidelity measurement, and longitudinal follow-up to strengthen causal inference and inform optimal implementation strategies.