Longitudinal Effectiveness of Communities That Care: Ten Years of Implementation Supported By Public Systems in the State of Pennsylvania

Introduction: The Communities That Care (CTC) prevention system has been supported by the State of Pennsylvania and implemented across the state since 1998. The Communities That Care model engages community coalitions to conduct a coordinated process of youth risk assessment, prevention program selection, and implementation of evidence-based practices responsive to their community’s specific needs. This community prevention system has been found to significantly improve adolescent outcomes in a quasi-experimental trial involving 24 communities. Additional research on the natural dissemination in the State of Pennsylvania in a short-term longitudinal quasi-experimental design with data from ten years ago also showed favorable effects. The goal of this study is to assess the continuing impact of the Communities That Care operating system on preventing youth delinquency and substance abuse across 10 years of implementation.

Methods: Utilizing extant longitudinal statewide youth risk surveillance data, this study is using propensity-score weighting techniques to form a matched sample of CTC and non-CTC communities across six waves of data collected bi-annually from 2001 to 2011. Multilevel regression models will be utilized to compare CTC to non-CTC communities on change in individual, school, and peer risk factors, as well as substance use, among students in grades 6, 8, 10, and 12. We are also exploring the relative impact of intervention quality on intervention outcome estimates obtained.

Results: Analyses are in progress. Descriptive analyses are showing that locations of CTC sponsored programs changed a fair amount from year to year. Early cross-sectional models are showing that the quality of programming (i.e., degree of evidence) and reach (universal vs. indicated) are affecting the sensitivity of CTC-sponsored programming to have an impact on youth outcomes. Next steps are to combine the cross-sectional datasets into one cohort-sequential longitudinal design that specifies Level 1 as time (within kids), level 2 equals kids, and level 3 equals schools and is the level at which the CTC programs were delivered.

Discussion: The long-term effectiveness of CTC under standard policy and practice conditions has not been investigated. Long-term effectiveness studies are necessary to make a case for sustained public health impact.