Shared residential placement for child welfare and juvenile justice youth: current treatment needs and risk of adult criminal conviction

Background Although child welfare youth and juvenile offenders in residential care have different judicial placement reasons, there seems to be overlap in their demographic and psychosocial backgrounds. This could raise the question whether these adolescents should be placed in strictly separated institutions based on their judicial title (civil or criminal law) or together based on their needs. As systematic knowledge on the effects of shared placement of these groups is limited, the aim of the current paper is to examine the demographic, crime-related and psychosocial characteristics of child welfare and juvenile justice youths in shared residential care and subsequently examine its relationship with offending behavior in adulthood. Methods The sample was drawn from the Swiss study for clarification and goal-attainment in youth welfare and juvenile justice institutions (MAZ.) and consisted 354 juveniles (252 child welfare, 102 juvenile justice; 223 boys, 131 girls) between 10 and 18 years. Mental health problems were assessed with the Massachusetts Youth Screening Instrument-Version 2 (MAYSI-2), official adult criminal conviction data up to 10 years later was obtained from the Swiss Federal Office of Statistics. Three sets of logistic regressions were conducted investigating any, violent and non-violent convictions. Results Univariate results showed that that the child welfare sample included more females, more juveniles with the Swiss nationality, and was younger at the time of assessment and at first placement compared to the juvenile justice sample. Furthermore, child welfare youths showed less alcohol/drug use problems and offending behavior than their juvenile justice counterparts. Unadjusted models demonstrated that committing authority predicted adult criminal convictions, but that this distinction disappeared when it was controlled for demographic, crime-related and psychosocial factors. Gender and time at risk were found to be related to adult conviction in all three models. In addition, alcohol/drug use problems were risk factors for general, previous convictions for violent, and traumatic experiences for non-violent convictions in adulthood. Conclusions Our results support the approach of placement in residential care institutions based on treatment needs instead of on judicial title. Special attention should be devoted to trauma informed care and substance use coping. However, more research is needed.