Session: Middle School Effects of the Dating Matters® Comprehensive Prevention Model: A Cluster-Randomized Controlled Trial (Society for Prevention Research 26th Annual Meeting)

Teen dating violence (TDV) is a significant public health problem. Despite its prevalence and long-term negative consequences, our understanding of how to best prevent TDV is still in its early stages. We know from other areas of violence prevention that comprehensive approaches spanning multiple levels of the social ecology are generally more effective than single component interventions. Therefore, the Centers for Disease Control and Prevention (CDC) developed and tested Dating Matters: Strategies to Promote Healthy Teen Relationships (Dating Matters) in high-risk urban communities to see if a comprehensive approach to TDV prevention and healthy relationship promotion was more effective than an existing evidence-based TDV curriculum. The Dating Matters comprehensive prevention model includes: classroom-delivered curricula for 6^th-8^th grade students; training for parents of 6^th-8^th graders; training on TDV for teachers/school personnel; a youth communications program; and activities at the local health department to assess capacity and readiness and track TDV-related policy and data. We hypothesized that the Dating Matters comprehensive model would be significantly more effective at preventing TDV and reducing other risk behaviors in middle school than an existing, single component TDV prevention program.

Using a cluster-randomized controlled trial design, this study examined the comparative effectiveness of Dating Matters comprehensive prevention model (DM) compared to a standard-of-care (SC) teen dating violence program (Safe Dates) in 46 schools across 4 urban U.S. sites. Ten to twelve middle schools per site were randomized to implement either DM or SC. Data were/are being collected from 5 cohorts of students, starting in 6^th grade through high school. Students completed self-report surveys twice per year while in middle school and once per year in high school. Papers in this symposium will present the middle school trial results on various outcomes for Cohorts 3 and 4, who started 6^th grade in either Year 1 (2012-13) or Year 2 (2013-14) of implementation, as students in the DM condition in these cohorts had the potential for full exposure to all 3 years of DM components. High school data for this sample are still being collected. The total student sample for these two cohorts is 3,301; each paper abstract lists the particular sample used for that study. All analyses are intent-to-treat. We used a structural equation modeling framework with multiple imputation to account for missing data to assess the equivalence of outcomes on a variety of outcomes presented in the three papers in this symposium, including but not limited to TDV, other relationship behaviors, bullying, substance use, delinquency, sexual violence, and sexual risk behaviors.