Effects of the communities that care model in Pennsylvania on change in adolescent risk and problem behaviors

Mark E. Feinberg, Damon Jones, Mark T. Greenberg, D. Wayne Osgood, Daniel Bontempo

Research output: Contribution to journalArticlepeer-review

88 Scopus citations

Abstract

Despite the public health burden of adolescent substance use, delinquency, and other problem behavior, few comprehensive models of disseminating evidence-based prevention programs to communities have demonstrated positive youth outcomes at a population level, capacity to maintain program fidelity, and sustainability. We examined whether the Communities That Care (CTC; Hawkins and Catalano 1992) model had a positive impact on risk/protective factors and academic and behavioral outcomes among adolescents in a quasi-experimental effectiveness study. We conducted a longitudinal study of CTC in Pennsylvania utilizing biannual surveillance data collected through anonymous in-school student surveys. We utilized multilevel models to examine CTC impact on change in risk/protective factors, grades, delinquency, and substance use over time. Youth in CTC communities demonstrated less growth in delinquency, but not substance use, than youth in non-CTC communities. Levels of risk factors increased more slowly, and protective factors and academic performance decreased more slowly, among CTC community grade-cohorts that were exposed to evidence-based, universal prevention programs than comparison grade cohorts. Community coalitions can affect adolescent risk and protective behaviors at a population level when evidence-based programs are utilized. CTC represents an effective model for disseminating such programs.

Original languageEnglish
Pages (from-to)163-171
Number of pages9
JournalPrevention Science
Volume11
Issue number2
DOIs
StatePublished - 2010

Keywords

  • Academic performance
  • Deliquency
  • Prevention

Fingerprint

Dive into the research topics of 'Effects of the communities that care model in Pennsylvania on change in adolescent risk and problem behaviors'. Together they form a unique fingerprint.

Cite this