Methods: This study utilizes a unique data sharing agreement in Washington that includes all child welfare, juvenile justice, and adult correctional records. The sample includes youth in substitute care settings between 2000 and 2003. The sample is limited to youth that were at least 7 years of age in 2000 (n=8,080), so that they would be eligible (old enough) for contact with the juvenile justice system by the end of the observation period (2009). We analyze administrative and standardized risk assessment data. We use propensity score procedures, logistic regression and event history analysis to understand the relative risk of placement in congregate care facilities, the likelihood of running from placement, and rate of juvenile and adult offending.
Results: The descriptive analyses indicate that 26% of youth in placement are associated with no formal allegations of maltreatment, but rather enter placement for reasons of “child behavioral problems.” An additional 22% of youth enter placement with an allegation of maltreatment, but also enter placement with a formal history of juvenile offending and stated “child behavioral problems.” Only 52% of youth in placement are referred solely for reasons of abuse/neglect. The regression models indicate that behavioral/delinquency referrals experience significantly worse outcomes as compared with abuse/neglect referrals. Specifically, behavioral/delinquency referrals are more likely to enter congregate/residential settings (15% v 4%), more likely to experience placement instability (39% v. 18%), more likely to run from placement (12% v. 4%), more likely to experience a juvenile arrest (Expβ = 4.13, p <.01), and more likely to experience an adult arrest (Expβ = 3.78, p <.01). The risk assessment analyses identify four predictors of juvenile and adult offending (1) positive peer relationships (2) drug/alcohol use, (3) school attendance, and (4) beliefs about aggression and anti-social conduct.
Conclusions and Implications: A relatively large proportion of youth in child welfare placements are referred for behavioral reasons, rather than for abuse/neglect. The findings indicate that these adolescents are not well served in the current child welfare system. As states such as Illinois, Michigan and New York seek to merge child welfare and juvenile justice systems, careful consideration must be given to the unique needs of these distinct populations. The findings from the risk assessment data can help guide the development and targeting of specific interventions for youth in substitute care settings.