Citation
Wiebush, R., Wagner, D., McNulty, B., Wang, Y., & Le, T. (2005). Implementation and outcome evaluation of the Intensive Aftercare Program: Final report. National Council on Crime and Delinquency. Washington, DC: U.S. Department of Justice, Office of Justice Programs.
Highlights
-
- The study’s objective was to measure the impact of participation in an Intensive Aftercare Program (IAP) for high-risk youth from juvenile justice facility placements on recidivism.
- The authors used a random assignment design and multivariate regression analysis to estimate impacts on aggregate recidivism scores, the sum of all subsequent offenses, weighted by severity. Data for the study were collected from standardized forms, surveys, risk assessment tests, and state agency and police records.
- The study found no statistically significant effects of the program on participants’ aggregate recidivism scores.
- The quality of causal evidence presented in this report is moderate for some outcomes and low for other outcomes. This means we have little confidence that any estimated effects would be attributable to the IAP. However, the study found few statistically significant effects.
Intervention Examined
The Intensive Aftercare Program (IAP)
Features of the Intervention
The IAPs in the study were implemented by juvenile correctional facilities managed by county juvenile justice departments. The program components included an assessment for risks and needs; individualized case planning to determine and prioritize service needs and determine a timeline for addressing them; continuity in case management and service delivery; coordination of pre- and post- release staff and program administrators; formalized transitions to facilitate post-incarceration reentry into the community; intensive supervision; and use of appropriately fitted and immediately applied rewards and sanctions that were scaled to match the severity or positivity of actions.
Features of the Study
The study took place in three juvenile justice facilities that received grants to implement the IAP. The program in Colorado served high-risk youth from the Denver metropolitan area who were incarcerated at Lookout Mountain Youth Services Center in Golden, Colorado; the program in Nevada was operated by the Clark County Youth Parole Bureau for high-risk youth from Clark County; and the program in Virginia was operated by the Virginia Department of Juvenile Justice and served high-risk youth from the city of Norfolk assigned to either the Beaumont or Hanover Juvenile Correctional Centers.
The authors randomly assigned eligible youth to either a treatment group, which could participate in IAP, or a control group, which received the transitional and post-incarceration services typically available in the community. A total of 515 youth were randomized into either treatment or control groups across three programs under study, and data on 435 youth were available for outcomes and impact analysis. Youth were eligible for the study if they were male, had been committed to the custody of juvenile correction agency, had been placed within a participating correctional facility in a county participating in the study, and scored as being at high risk for reoffense using a locally validated risk-assessment tool. Youth who had committed sexual offenses, were in custody until after age 21, or had a mental health diagnosis requiring special attention or treatment were not eligible for the study.
The authors collected data on characteristics and outcomes of interest from a series of forms, surveys, standardized tests, and state agency and police records one year after program participation. The primary outcome of interest was a criminal offense recidivism score, which represented the mean seriousness and frequency of criminal arrests during the follow-up period.
Findings
-
- The study did not find a statistically significant effect of IAP on the criminal offense recidivism score.
Considerations for Interpreting the Findings
Although the study was based on a randomized controlled trial, all three of the participating sites had high study attrition. Randomized controlled trials with high attrition are evaluated against the CLEAR evidence guidelines for nonexperimental analyses. The analysis of the criminal offense recidivism score included controls for criminal history, race, previous mental health or school discipline problems, gang membership, and having a family member incarcerated, among other characteristics. Therefore, this analysis receives a moderate evidence rating. However, the analyses of other outcomes in the study did not include these control variables, and therefore they receive a low causal evidence rating.
The authors calculated that, given the small sample sizes and other features of their study and analysis, they would be able to detect statistically significant differences in the recidivism rate of only 22 percentage points or larger.
Causal Evidence Rating
The quality of causal evidence presented in this report is moderate for some outcomes and low for other outcomes. This means we have little confidence that any estimated effects would be attributable to the IAP. However, the study found few statistically significant effects.