An evaluation of the NSW Compulsory Drug Treatment Program (CDTP)

  • Date published: 02 Jun 2026
  • Author: Adam Teperski and Bonnie Ross
  • Publication series: Crime and justice bulletin
  • Subject: Drugs and drug courts, Prison and prisoners, Reoffending
Supporting documents

Summary

Background

The 2006 Compulsory Drug Treatment Program (CDTP) was established to allow the NSW Drug Court to order male sentenced, repeat drug offenders to the Compulsory Drug Treatment Correctional Centre (CDTCC) for comprehensive drug treatment and rehabilitation. 

Participation in CDTP is conditional on meeting eligibility criteria relating to drug dependence, sentence length, prior offending, mental health and location of residence, and being assessed as suitable for participation by corrective services staff. 

To estimate the association between CDTP participation and reoffending, we identified 360 CDTP participants and 1,352 control prisoners who met observable eligibility criteria for CDTP but did not participate in CDTP or any other drug or alcohol program while in custody. We used entropy balancing to match CDTP participants to control prisoners with similar observed characteristics.

We then compared the following reoffending outcomes between CDTP participants and matched control prisoners within 12 months of release from custody: 1) any reoffence; 2) a violent reoffence; 3) a drug reoffence; 4) a property reoffence; 5) a return to custody; and 6) a new sentenced custodial episode.

Key findings

Figure 1 shows the proportions of and differences in 12-month reoffending outcomes between CDTP participants and matched control prisoners after controlling for relevant sociodemographic characteristics and prior offending histories. 

We estimate that CDTP participants were 8p.p. less likely to reoffend with a drug offence relative to matched control prisoners (20% vs 12%). This represents a marginal decrease in drug reoffending of roughly 40%. CDTP participants were also slightly more likely to reoffend with any proven offence, a proven violent offence, a proven property offence, a new custodial episode and a new sentenced custodial episode by 3p.p., 3p.p., 5p.p., 6p.p. and 2p.p, respectively. However, these differences were not statistically significant at the 5% level.

Figure 1. Association between CDTP participation and reoffending outcomes, relative to matched control group 


Note. Error bars refer to the 95% confidence interval associated with the estimates. Stars indicate statistical significance at a variety of conventional thresholds of statistical significance: * – 10%, ** – 5%, *** – 1%.

Conclusion

CDTP participation is associated with reduced drug reoffending within 12 months of release but does not affect broader reoffending outcomes. Program refinements to better address a wider range of criminogenic factors may be needed to maximise the benefits of the CDTP.

Last updated: