While you’re here… help us stay here.

As we confront the economic impacts of the pandemic, we need your support even more. So that we can continue to bring you the latest in policy and research, please donate to APO.


Assessing the impact of NSW’s Safer Pathway Program on recorded crime outcomes – an aggregate-level analysis

Family violence Violence against women Victims of family violence Intimate Partner Violence (IPV) New South Wales
Attachment Size
apo-nid143596.pdf 902.02 KB


Aim: To evaluate the effectiveness of the Safer Pathway program in reducing the incidence of domestic violence.

Method: Data was obtained from the NSW Police Force’s COPS and CIDS databases. Each of the nine LACs where the Safer Pathway program had been implemented were matched with statistically similar control LACs where the program was not available. Seven different domestic violence (DV) outcomes were examined; the number of incidents of DV offences, DV assaults, persons of interest (POIs) proceeded against for DV offences, POIs proceeded against for DV assault, victims of DV incidents, victims of DV assault and police call-outs for DV. Each outcome was analysed separately using a fixed-effect negative binomial or Poisson panel model. This analysis was performed both at the pair level (the pair analysis) and pooling LACs based on stage of implementation (the stage analysis). The main variable of interest was the difference in the trend change between the Safer Pathway and control LACs.

Results: The stage analysis showed no significant results for any stage 1 LACs (where the program commenced in September 2014) for any of the outcomes examined. However, the stage 2 LACs (where the program commenced in July 2015) showed a significant decrease in three outcomes relative to the matched control LACs (all by 0.8% per month). In the pair analysis, downward differences in trend changes were found in three outcomes in Canobolas (of 2.2%, 1.6% and 2.2% per month) and one in Rose Bay (2.9% per month). Botany Bay however showed a significant upward difference in trend change in four different outcomes (ranging between 1.3 and 1.9% per month).

Conclusion: In stage 2 LACs, there is evidence for a small improvement in several DV indicators after the program was implemented. The results for stage 1 LACs are mixed.

Related Information


Publication Details


License type:
All Rights Reserved
Access Rights Type:
Contemporary Issues in Crime and Justice no. 210