U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Police-Initiated Diversion for Youth to Prevent Future Delinquent Behavior: A Systematic Review

NCJ Number
David B. Wilson; Iain Brennan; Ajima Olaghere
Date Published
88 pages

The authors of this paper discuss the evidence on the effectiveness of pre-court interventions involving police warning or counseling and release; they provide a brief summary before discussing their review process, data collection and analysis, and results.


This Campbell systematic review examines the effects police-initiated diversion programs on delinquent behavior, compared to traditional system processing. The review summarizes evidence from 13 randomized controlled trials and six quasi-experimental studies, and includes studies that evaluated the effects of police-led diversionary practices compared to traditional processing for youth under 18 years of age. Many of the study designs included two or more diversionary conditions compared to a common control (traditional processing) producing 31 treatment-comparison contrasts for analysis. These studies were conducted between 1973 and 2011, inclusively. Eleven were conducted in the U.S. with the remaining conducted in Canada, Australia, and the U.K. The general pattern of evidence is positive, suggesting that police-led diversion reduces the future delinquent behavior of low-risk youth relative to traditional processing. Assuming a 50 percent reoffending rate for the traditional processing condition, the results suggest a reoffending rate of roughly 44 percent for the diverted youth. This overall benefit of diversion holds for the random assignment studies judged to be free from any obvious risks of bias. No meaningful differences were found across types of diversionary programs. Furthermore, the authors found no evidence to suggest these findings suffer from publication selection bias. Publisher Abstract Provided