Crime and Justice Bulletin, 2018, no. 213 (Open access)

The NSW Bureau of Crime, Statistics and Researdch (BOCSAR) evaluated the DVSAT screening tool (which is a key component of the Safer Pathway program), to see how reliable it is in identifying victims of domestic violence at high risk of repeat victimisation. This evaluation is separate but related to the evaluation of the NSW Safer Pathway Program (see #5835)..

The DVSAT consists of 30 questions and is made up of two parts: Part A and Part B. Part A involves 25 questions dealing with the background/current environment of the offender, the threat of violence, the dynamics of the specific relationship, the presence of children and any experience of sexual behaviours/assault. This part is only used for victims in intimate partner relationships.

Part B is completed for victims in both intimate and non-intimate relationships. The investigating officer provides responses to five questions relating to the level of fear felt by the victim, the reasons for those fears, whether there are children at risk of harm, and whether there are any additional factors that cause the officer to believe there is a threat or serious threat to the safety of the victim and/or children. Victims are assessed as being 'at serious threat' if:

They give 12 or more 'yes' responses to the 25 questions in Part A or
A police officer concludes on the basis of responses to Part B that the victim is at serious threat.
To evaluate the DVSAT BOCSAR tracked 24,462 victims of intimate partner violence who were administered the DVSAT between 1 January 2016 and 30 June 2016 to see which of them (a) experienced any repeat incident of domestic violence and/or (b) experienced a physical incident of domestic violence (including homicide, assault, sexual assault or robbery).

The DVSAT turned out to be a very poor instrument for measuring the risk of repeat domestic violence victimisation, often performing little better than chance. For example, while the rate of repeat victimisation overall in females who responded 'yes' to 12 or more items was higher than the rate in those who responded 'yes' to fewer than 12 items (45.0% vs. 34.2%), over one-third of those who responded 'yes' to fewer than 12 items actually went on to experience repeat victimisation.

Again, of the females who experienced repeat victimisation involving any type of intimate partner violence, only 10.0 per cent had responded 'yes' to 12 or more items. Put another way, 90.0 per cent of those who experienced repeat victimisation had responded 'yes' to fewer than 12 items in the DVSAT.

Commenting on the results of the research, the executive director of BOCSAR, Dr Don Weatherburn, said that the unreliability of the DVSAT may be one reason the Safer Pathway program has had only limited success in reducing domestic violence. "Many victims of domestic violence who are at serious risk of repeat victimisation are not getting the benefit of a safety action plan. BOCSAR is working on developing an improved screening tool to overcome this problem", he said. (From the website). Record #5836