Co-Principal Investigator: Koschmann, Elizabeth

Purpose: The purpose of this project is to examine the efficacy of the TRAILS Early Intervention program (TRAILS-EI) by evaluating its impact on the mental health and academic performance of students in Detroit Public Schools Community District. TRAILS (Transforming Research into Action to Improve the Lives of Students) is a University of Michigan program that promotes utilization of Cognitive Behavioral Therapy (CBT) by providing school mental health professionals (SMHPs) curriculum resources, training, and post-training implementation support to sustain uptake and use of best practices. As in many high-poverty, high-density settings throughout the United States, students in Detroit experience high rates of environmental stress and untreated mental illness and consequently, poor school engagement. Rigorous evaluation of TRAILS-EI will provide critical direction for TRAILS as it begins a rapid expansion and for school systems and other stakeholders about ways to improve both educational and behavioral outcomes among students, particularly in urban, majority-minority communities with limited resources.

Project Activities: The study team will evaluate TRAILS-EI, the program's tier two model, as part of a multiyear partnership between TRAILS and the Detroit Public Schools Community District (DPSCD), the largest urban school district in Michigan. Students are predominantly Black/African American (82%), low socioeconomic status (85% qualify for free or reduced lunch), and fewer than 15% are college ready at graduation. TRAILS-EI will be implemented in a total of 75 DPSCD schools, serving 36,000 students and 300 SMHPs each year. The study team will randomly assign schools to three cohorts, with a new cohort of schools beginning program implementation each year in years 1–3. Comparison of cohorts over time will allow for evaluation of implementation outcomes and short-term and longer-term program efficacy. The team will draw on annual survey data and administrative records to measure social and behavioral as well as academic outcomes among students. Exploratory analyses will include implementation measures and components of primary outcomes.

Products: The team will create written reports and presentations for school administrators; fliers, mailings, brochures, and social media campaigns for students, families and community members; policy briefs and white papers for legislators, policymakers, and health and education funders; and conference presentations and peer-reviewed publications for academic researchers.

Structured Abstract

Setting: This project will take place in a total of 75 middle and high schools in Detroit Public Schools Community District (DPSCD), the largest urban school district in Michigan, with over 50,000 students spread across 105 buildings.

Sample: The primary study population will be DPSCD students in grades 6–12. DPSCD has a 97.5% minority student population, consisting primarily of students who identify as Black/African American (82%). Eighty-five percent of students are eligible for free or reduced cost lunch, more than half of all DPSCD students are chronically absent (missing more than 10% of school days a year), one-fifth drop out, and less than 15% are college ready at graduation. More than 4,000 teaching, support, and administrative personnel among DPSCD staff and faculty will provide implementation data.

Intervention: TRAILS-EI is designed to increase student access to effective mental health services by training school mental health professionals (SMHPs) in Cognitive Behavioral Therapy (CBT) – an evidence-based approach for the management of adolescent mental health difficulties. TRAILS-EI prepares SMHPs to address student stress, depression, anxiety, and related symptoms in small-group or individual formats, both in person or virtually, by providing didactic training and post-training implementation support.

Research Design and Methods: The study will involve a cluster randomized design in which 75 middle and high schools are randomly assigned to one of three groups at the start of the study. During the first project year (school year 2022–2023), TRAILS-EI will be implemented in the first cohort of 25 schools. During the second project year (2023–2024), a second cohort of 25 schools will begin implementing the TRAILS-EI program, and in the third project year the final cohort of 25 schools will begin implementation. Randomly selecting schools to implement the program over time will enable rigorous assessment of both the short term impact of TRAILS-EI as well as the longer-term impacts of the program. Phasing in the intervention over several years will also allow fidelity monitoring and necessary program adjustments. The team will administer surveys to students and staff in the spring of the first three study years. To answer questions about the timing and dosage of the TRAILS-EI model, the study team will assess outcomes at the end of each school year.

Control Condition: TRAILS-EI implementation will be phased in over three years across three randomly allocated cohorts to allow for efficacy evaluation. All schools will eventually receive the TRAILS-EI program. Schools not receiving TRAILS-EI in years 1 and 2 of the study will have access to the typical emotional and behavioral supports that are available to all DPSCD schools.

Key Measures: Academic outcomes will be measured by students' current year GPA, math and reading test scores, and on-time grade progression. Social and behavioral outcomes include rates of clinical depression and anxiety as measured by the Patient Health Questionnaire-9 Item, Modified for Adolescents([PHQ-9A) and the Generalized Anxiety Disorder Scale (GAD-7), absenteeism, school engagement, and involvement with school discipline. The researchers will also collect information about TRAILS-EI implementation through surveys of organizational readiness for change and implementation climate.

Data Analytic Strategy: The goal of the analyses is to estimate the impact of TRAILS-EI on student behavioral and academic outcomes. Given the random assignment, simple cohort comparisons will capture the causal impact of the intervention. Program staff will estimate these differences in a regression framework, which will allow increased statistical precision by controlling for baseline covariates. All of the analyses will be conducted at the student level, but standard errors will be appropriately adjusted to account for clustering of students within schools.

Cost Analysis: The study team will calculate the average costs and average outcomes of TRAILS-EI. They will calculate Incremental Cost Effectiveness Ratios (ICERs) by dividing the incremental average costs by the estimated impact for each outcome. The ICER reflects the incremental program cost (C) of TRAILS-EI per incremental change in average outcomes (E [for effect]) relative to business as usual. Outcomes will include changes in social, behavioral, and academic outcomes assessed at each data collection time point. The study team will also explore the human capital benefits of the intervention. On-time grade progression will serve as a proxy for high school graduation. High school graduation is a key predictor of lifetime earnings and they will calculate the average predicted increase in lifetime earnings as predicted by any increase in high school graduation rates.