Program Profile: Crossover Youth Practice Model

This is a model that uses a conceptual plan and organizational framework to strengthen collaborations between child welfare and juvenile justice system professionals and partners to prevent or reduce youths’ involvement in the juvenile justice system or related systems of care. The program is rated Promising. Treatment group youth showed a statistically significant reduction in recidivism rates, compared with youth in the comparison groups.

A Promising rating implies that implementing the program may result in the intended outcome(s).

Program Goals/Target Population

Crossover youth are dually involved in child welfare and the juvenile justice systems; they have experienced neglect or maltreatment and have also engaged in delinquent activity (Huang et al. 2017). The Crossover Youth Practice Model (CYPM) is a model that uses a conceptual plan and organizational framework to minimize the involvement of crossover youth in the juvenile justice system by improving communication and coordination between professionals in both the child welfare and juvenile justice systems, providing more individualized interventions to youth, and increasing family engagement in the process (Haight et al. 2016; Huang et al. 2017). A primary goal of the CYPM is to reduce delinquency and system involvement of crossover youth by providing early, coordinated, and individualized services.

Program Components

The CYPM is implemented in three phases. Phase I focuses on arrest, identification of crossover youth, and decisions regarding detention and charges. The goal is early identification of crossover youth, to divert them from the juvenile justice system (if appropriate). This phase includes 1) development of memoranda of understanding and information-sharing protocols that establish how client information databases are shared between child welfare and juvenile justice systems for identification purposes as soon as youth become involved with the juvenile justice system; 2) diversion meetings with youth’s family members and juvenile justice and child welfare professionals; and 3) cross-system case management at the onset of the case to support multi-system collaboration. This phase also addresses development of prevention strategies to reduce the risk of youth crossing over between child welfare and juvenile justice.

Phase II focuses on dual-system case assessments and planning once a youth is dually involved in the child welfare and juvenile justice systems. This phase includes 1) joint case assessment and planning by a multidisciplinary team, which includes, at minimum, a child welfare social worker, juvenile justice professional, the youth, and family members; and 2) consolidated court processing to handle delinquency and dependency hearings together, joint referrals to community service providers, and placement decisions that reduce the use of out-of-home placement, particularly that of group care.

Phase III focuses on case management and planning for case closure. This phase includes 1) regular information sharing between child welfare and juvenile justice professionals, and 2) partnering with child welfare and juvenile justice professionals to plan for permanency, which includes securing any necessary continual mental health, employment, housing, health care, and education support.

Youth and family engagement is fundamental to the CYPM. To facilitate family engagement, professionals from the child welfare and juvenile justice systems meet collectively with family members to explore how to work cohesively and how to support family participation in all three phases of CYPM (“family members” may include adult friends, foster parents, members of community groups, and other collateral family partners or key supporters in the youth’s life).

Program Theory

The theoretical structure of the CYPM is based on an integration of systems change and sociocultural perspectives. Both concepts are broadly concerned with understanding the process through which systems are maintained and changed (Stewart et al. 2010; Rogoff 2003). Through this framework, the CYPM interprets child welfare and juvenile justice systems as cultural systems with structural processes involving change and stability of official hierarchies, administrative structures, and formal policies. This concept acknowledges that change may not occur smoothly or simultaneously across all parts of any individual system, especially across multiple evolving systems. Thus, change efforts may require coordination of complex patterns of change and response to resistance across interacting systems (Haight et al. 2016).

Study

Haight and colleagues (2016) conducted a quasi-experiment to examine the impact of the Crossover Youth Practice Model (CYPM). The study’s participants were from Oak County (a pseudonym), which is a large urban county in a midwestern state. Youth in the treatment group who received services through the CYPM were compared with youth in three comparison groups: 1) an Oak County historical comparison group, 2) neighboring counties’ historical comparison group, and 3) neighboring counties’ contemporaneous comparison group. All youth were between the ages of 10 and 17 and had an open child protection case at the time they were also accused of a delinquent offense (other than a juvenile status or traffic offense). All youth in the treatment group and comparison groups were tracked for 12 months following their target offense dates.

Treatment group youth received CYPM services for more than 15 days between October 2012 and August 2013. Youth were identified for inclusion in the comparison groups if they were born between 1994 and 2004, and were present in all three databases that were used in the study, including 1) State Court Information System (MNCIS), 2) Social Service Information System, and 3) Automated Report Student System (MARSS). Youth were chosen for inclusion in the comparison groups based on a set of 10 dimensions related to recidivism (such as age, gender, race/ethnicity, and severity of target offense). The sampling frame was limited to youth whose offenses occurred between June 2008 and December 2010, for the historical comparison groups; and January 2011 through August 2013, for the contemporaneous comparison group. The comparison groups were created using propensity score matching (PSM). This matching method minimized selection bias and the distance within each pair, using a logistic regression analysis to control for various background covariates (i.e., time and locale) simultaneously by matching participants on a single scalar score variable. There were adjustments made to improve the distance between pairs when matching covariates after the initial PSM process.

The baseline characteristics of the treatment and comparison groups were as follows:

• In the Oak County CYPM treatment group ( n = 57), 54 percent of the youth were male. Of this group, 75 percent identified as African American, 12 percent as white, 9 percent as Native American, and 4 percent as Hispanic. Child welfare allegations included neglect (61 percent), physical abuse (33 percent), and sexual abuse (18 percent). The target offenses that treatment group youth were adjudicated for included felony offenses (25 percent), gross misdemeanors (16 percent), misdemeanors (47 percent), and petty misdemeanors (12 percent). The mean average age of youth in the treatment group, at the time of their target offense, was 14.8 years.
• In the matched Oak County historical comparison group ( n = 57), 60 percent were male. Of this group, 63 percent identified as African American, 26 percent as white, 7 percent as Hispanic, and 4 percent as Native American. Child welfare allegations included neglect (51 percent), physical abuse (40 percent), and sexual abuse (18 percent). Thirty-two percent were adjudicated for a felony target offense, 16 percent for a gross misdemeanor, 42 percent for a misdemeanor, and 10 percent for a petty misdemeanor. The average age of these youth, at the time of the target offense, was 15.0 years.
• In the matched neighboring counties’ historical comparison group ( n = 57), 47 percent were male. Of this group, 72 percent identified as African American, 18 percent as white, 6 percent as Native American, and 4 percent as Hispanic. Child welfare allegations included neglect (51 percent), physical abuse (47 percent), and sexual abuse (19 percent). Thirty percent were adjudicated for a felony target offense, 21 percent for a gross misdemeanor, 42 percent for a misdemeanor, and 7 percent for a petty misdemeanor. The average age of these youth, at the time of the target offense, was 14.7 years.
• In the matched neighboring counties’ contemporaneous group ( n = 57), 44 percent were male. Of this group, 70 percent identified as African American, 23 percent as white, 5 percent as Native American, and 2 percent as Hispanic. Child welfare allegations included neglect (47 percent), physical abuse (39 percent), and sexual abuse (25 percent). Twenty-five percent were adjudicated for a felony target offense, 16 percent for a gross misdemeanor, 47 percent for a misdemeanor, and 12 percent for a petty misdemeanor. The average age, at the time of target offense, was 15.0 years.

The primary outcome of interest was recidivism, defined as being adjudicated for one or more offenses within 1 year following the date of the target offense. If a youth recidivated, a dichotomous variable was used to indicate one of two statuses [(0) did not recidivate; (1) recidivated] within the follow-up year. Recidivism rates were examined using a logistic regression analysis. Log odds ratios were used to quantify the differences between the treatment and comparison groups after considering the effects of time, locale, and other covariates. Analyses were conducted for the comparisons between 1) the CYPM treatment group and the Oak County historical comparison group; 2) the CYPM treatment group and the neighboring counties’ contemporaneous comparison group; 3) the neighboring counties’ historical comparison group and the neighboring counties’ contemporaneous comparison group; and 4) the CYPM treatment group with the three comparison groups combined. This CrimeSolutions review focused on the comparison between the CYPM treatment group and the three comparison groups. Subgroup analyses were not conducted.

A study by Wright and colleagues (2017) calculated costs for the Crossover Youth Practice Model (CYPM) in a jurisdiction in another midwestern state. The analysis showed that initial implementation costs were approximately $59,752, which included data system costs and staffing costs for the Nebraska Center for Justice Research, which conducted training for child welfare and juvenile justice system professionals. The annual administration cost was estimated to be approximately $212,264. The study also found a total annual benefit of $385,425 resulting from lowered court and probation costs, netting a total benefit of $173,161 per year.

To implement the Crossover Youth Practice Model (CYPM) thoroughly, child welfare and juvenile justice system professionals received training and technical assistance from representatives from the Center for Juvenile Justice Reform (CJJR) at Georgetown University. For more information on the CYPM, please see CJJR’s website: https://cjjr.georgetown.edu/our-work/crossover-youth-practice-model/ .

For the Oak County CYPM implementation, professionals received training over a 2-year period. There was a full year of training prior to implementation, where CJJR representatives met every 1 to 2 months to train professionals on the CYPM components and implementation. The remainder of the training coincided with program implementation and comprised of regular training sessions that included discussions of progress and challenges with implementation (Haight et al. 2016).

These sources were used in the development of the program profile:

These sources were used in the development of the program profile:

Bilchik, Shay, Denise C. Herz, and Anika M. Fontaine. 2012. Final Data Report for the Crossover Youth Practice Model in King County, Washington. 2010/2011 Cases. Washington, D.C.: Georgetown University, Center for Juvenile Justice Reform.

Culhane, Dennis P., Thomas Byrne, Stephen Metraux, Manuel Moreno, Halil Toros, and Max Stevens. 2017. Young Adult Outcomes of Youth Exiting Dependent or Delinquent Care in Los Angeles County . Los Angeles, Calif.: Conrad Hilton Foundation.

Culhane, Dennis P., Stephen Metraux, and Manuel Moreno. 2011. Young Adult Outcomes of Youth Exiting Dependent or Delinquent Care in Los Angeles County . Los Angeles, Calif.: Conrad Hilton Foundation.

Farn, Amber, and Michael Umpierre. 2017. Creating an Integrated Continuum of Care for Justice-Involved Youth: How Sacramento County Collaborates Across Systems . Washington, D.C.: Georgetown University, McCourt School of Public Policy, Center for Juvenile Justice Reform.

Haight, Wendy L., Laurel N. Bidwell, Jane Marie Marshall, and Paramananda Khatiwoda. 2014. “Implementing the Crossover Youth Practice Model in Diverse Contexts: Child Welfare and Juvenile Justice Professionals’ Experiences of Multisystem Collaborations.” Children and Youth Services Review 39:91–100.

Halemba, George, and Gene Siegel. 2011. Doorways to Delinquency: Multi-System Involvement of Delinquent Youth in King County (Seattle, WA). Pittsburgh, Pa.: National Center for Juvenile Justice.

Heldman, Jessica. 2016. Dual Status Youth Initiative Report: Early Gains and Lessons Learned . Boston, Mass.: Robert F. Kennedy Children’s Action Corps.

Herz, Denise C. 2015a. Research Summary: Crossover Youth Practice Model. Washington, D.C.: Georgetown University, McCourt School of Public Policy, Center for Juvenile Justice Reform.

Herz, Denise C. 2015b. Examination of the Los Angeles County 241.1 Multidisciplinary Team: A Summary of Findings form October 2013 to December 2014. Report to the Los Angeles County Board of Supervisors. Los Angeles, Calif.: California State University—Los Angeles, School of Criminal Justice and Criminalistics.

Herz, Denise C. 2016. A Summary of Findings for the Los Angeles County 241.1 Multidisciplinary Team. Report to the Los Angeles County Board of Supervisors. Los Angeles, Calif.: California State University—Los Angeles, School of Criminal Justice and Criminalistics.

Herz, Denise C., and Carly B. Dierkhising. 2019. OJJDP Dual System Youth Design Study: Summary of Findings and Recommendations for Pursuing a National Estimate of Dual System Youth. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Huang, Hui. 2017. The Effectiveness of Service Integration: Studying the Crossover Youth Practice Model. Tallahassee, Fla.: Florida State University, School of Social Work.

Huang, Hui, and Michelle-Ann Rhoden. 2017. The Effectiveness of Service Integration: Studying the Crossover Youth Practice Model. Tallahassee, Fla.: Florida State University, Florida Institute for Child Welfare. (This study was reviewed but did not meet CrimeSolutions criteria for inclusion in the overall program rating.)

McCroskey, Jacquelyn, Denise C. Herz, and Emily Putnam-Hornstein. 2017. Crossover Youth: Los Angeles County Probation Youth with Previous Referrals to Child Protective Services . Los Angeles, Calif.: Children’s Data Network.

Rogoff, Barbara. 2003. The Cultural Nature of Human Development. New York, N.Y.: Oxford University Press.

Stewart, Macon, Lorrie Lutz, and Denise Herz. 2010. Crossover Youth Practice Model. Washington, D.C.: Georgetown University, McCourt School of Public Policy, Center for Juvenile Justice Reform.

Wright, Emily M., Ryan Spohn, and Joselyn L. Chenane. 2017. Evaluation of the Crossover Youth Practice Model (Youth Impact!): Executive Summary. Omaha, Neb.: Nebraska Center for Justice Research, University of Nebraska Omaha. (This study was reviewed but did not meet CrimeSolutions criteria for inclusion in the overall program rating.)

Young, Douglas, Alex Bowley, Jeanne Bilanin, and Amy Ho. 2015. Traversing Two Systems: An Assessment of Crossover Youth in Maryland. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, National Institute of Justice.

Following are CrimeSolutions-rated programs that are related to this practice:

Wraparound Process for Children with Serious Emotional and Behavioral Disorders

This practice is a team-based, collaborative process for developing and implementing individualized care plans for youth with serious emotional and behavioral disorders and their families. The practice is rated Promising for improving mental health outcomes, but rated Ineffective for measures related to youths’ living situations, school functioning, and recidivism outcomes.

Evidence Ratings for Outcomes

	Mental Health & Behavioral Health - Multiple mental health/behavioral health outcomes
	Family Functioning - Living Situation
	Education - Multiple education outcomes
	Crime & Delinquency - Multiple crime/offense types

Why might a practice's outcome ratings differ from the ratings of specific programs encompassed by that practice?