Summary
Aim
To determine whether the Bail Assessment Officer (BAO) intervention which was piloted in Central and Parramatta local courts in late 2016 had an impact on the proportion of defendants being granted bail at first court appearance and/ or reducing time spent in custody.
Method
The impact of the BAO intervention on bail refusal and time spent in custody was assessed using a two-by-two design; with group (BAO intervention vs. control) and time (pre-intervention period vs. post-intervention period) as the two factors. The treatment group consisted of custody-based defendants from two local courts (Central and Parramatta) where the BAO intervention was operating. The control group consisted of defendants appearing at three local courts (Burwood, Campbelltown and Fairfield) where the BAO intervention was not operating. The pre-intervention period was SeptemberDecember 2015 and the post-intervention period was September-December 2016. Outcomes compared in the analysis were percentage of defendants granted bail at first court appearance, mean bail refused days and mean time on remand.
Results
There was no significant change from pre-intervention to post-intervention in the percentage of defendants granted bail at first court appearance or the mean number of bail refused days or days on remand for defendants appearing in the BAO intervention courts. Similar results were also evident for defendants appearing in the control group courts and held even after controlling for a wide range of covariates.
Conclusion
There is no evidence from this aggregate-level analysis that the BAO intervention had an impact on the
granting of bail at first appearance, bail refused days or time on remand. However, a rigorous assessment of program effectiveness was precluded because of the lack of clear, objective eligibility criteria for program entry and limited recording of essential program data.