Aim: To determine whether the inclusion of child protection data improves the ability to predict re-offending in young persons with few prior formal contacts with the criminal justice system.

Method: A cohort of young persons who had their first, second or third formal contact with the criminal justice system (i.e., police caution, youth justice conference, or court appearance) in 2011 or 2012 were identified in the Re-offending Database maintained by the NSW Bureau of Crime Statistics and Research. For these young persons, NSW Department of Family and Community Services provided data relating to risk of significant harm (ROSH) reports and out-of-home-care (OOHC). Models predicting re-offending using demographic and criminal history variables were developed and compared with models that also included child protection information. Separate models were developed for males and females, as well as for the subgroup of young persons having their first contact with the criminal justice system.

Results: While some variables relating to ROSH reports and OOHC data were significant predictors of re-offending, the inclusion of child protection data did little to improve the ability to predict re-offending within 12 months in young persons with few prior formal contacts. Models with and without child protection data were similarly poor at predicting those who would go on to re-offend.

Conclusion: Early identification of young persons at risk of re-offending using administrative data sources remains a challenge.