Delinquency Trajectories of Crossover Youth in Early- and Mid-Adolescence

Background and Purpose:

The dual involvement of youth in the child welfare and juvenile justice systems is a persistent social problem. “Crossover youth” are broadly defined as maltreated youth who have engaged in delinquency. Maltreated youth tend to enter into the juvenile justice system at younger ages than their delinquent counterparts who are not maltreated. Consistent with the literature in criminology, the early onset of delinquent behavior among crossover youth is a marker of increased risk for continued delinquent/criminal behaviors. Yet relatively little data considers the age of onset of delinquent behavior for crossover youth. Such data would provide insight into any distinct risk and protective processes across development. The purpose of this study is to examine delinquency trajectories of youth crossing over during early- and mid-adolescence and to generate developmentally-sensitive interventions to prevent the involvement of maltreated youth in the juvenile justice system.

Methods:

State-level administrative data were integrated from education, child welfare, and juvenile justice systems. Using a prospective, longitudinal design, we followed two cohorts of maltreated youth for 6 years. The study sample consisted of 5,002 youth in 3rd grade for the early adolescence cohort (EA) and 4,108 youth in 7th grade for the mid-adolescence cohort (MA) who had official reports of child maltreatment but no records of juvenile court involvement prior to September 1^st, 2008. Cox proportional hazard model was employed to model time to maltreated youth’s first involvement in the juvenile justice system and to examine the trajectories involving the different factors associated with the risk of crossing over between early- and mid-adolescence cohorts.

Results:

Over the 6-year study period, approximately 7% of youth in the early adolescence and 16% of youth in the mid-adolescence cohort crossed over to the juvenile justice system. The results of the Cox PH model indicate that males (EA: HR = 1.87, p< .001; MA: HR = 1.55, p< .001) and youth with out-of-school suspension (EA: HR = 1.53, p< .05; MA: HR = 1.97, p< .001) were more likely to engage in delinquency in both cohorts. For the early adolescence cohort, race (Black: HR = 1.8, p<.001; Hispanic: HR = 1.73, p<.05; Native American: HR = 2.34, p<.001), emotional/behavioral disorders (HR = 1.96, p<.01) and more than three previous maltreatment incidents (HR = 2.02, p<.001) were significant predictors of crossing over. For the mid-adolescence cohort, low socioeconomic status (HR = 1.31, p<.01), academic achievement below grade level (HR = 1.01, p<.05), and a history of out-of-home placements (HR = 1.23, p<.05) were significant predictors of crossing over.

Conclusions and Implications:

Results from this study suggest that risk and protective processes for maltreated youth’s involvement in the juvenile justice system vary with development. Additionally, the findings suggest that the risk and protective processes for maltreated youths’ involvement in delinquency occur within multiple embedded and interacting social structures. The design of effective prevention interventions requires an understanding of developmental processes as embedded within social contexts to identify maltreated youth at highest risk for delinquency.