Background High rates of mental health disorders and exposure to trauma among the juvenile justice population highlight the importance of understanding whether and how mental health services can help prevent further justice system involvement as well as provide treatment. Aims We had two principal questions: Is out‐of‐home mental health treatment after arrest associated with reduced recidivism among young people who have been exposed to trauma? Are particular types of out‐of‐home treatment associated with better outcomes? We hypothesised that type of residential setting would affect outcomes among those with histories of serious trauma. Methods Primary data sources included Florida Juvenile Justice, Child Welfare and Medicaid data from July 2002 through June 2008. We identified all Florida Juvenile Justice cases with a record of ‘severe emotional disturbance’. Two groups were identified – one for whom an arrest was followed within 90 days by out‐of‐home placement for mental health treatment and the other for whom there was some other out‐of‐home placement. Generalised estimating equations (GEE) were used to calculate associations with likelihood of re‐arrest during a 12 month at risk period. Results Young people who had experienced severe trauma and were sent to out‐of‐home treatment settings after conviction for a criminal offence had lower recidivism rates when receiving treatment in foster care than other out‐of‐home placements, while those with less severe or no trauma histories had lower recidivism rates with any out‐of‐home placement. Conclusions and implications for practice We believe that this is the first study to have considered how trauma histories may mediate outcomes for young people in out‐of‐home placements after arrest or conviction for a criminal offence. Findings suggest that case managers and clinicians should consider the trauma histories when making decisions about the appropriate treatment setting in these circumstances. Copyright © 2017 John Wiley & Sons, Ltd.