Factors affecting risk


The following factors are commonly identified in academic literature, domestic violence death reviews and a range of risk assessment tools as key signifiers of risk for the escalation of domestic and family violence. It is not an exhaustive list and other resources may include additional factors. The distinguishing characteristic of domestic and family violence is that it can present in many forms and can occur within a variety of relationships and is most likely to involve a complex pattern of controlling behaviour and violence over a period of time, rather than a single incident. This is often referred to as coercive control. One study has described coercive control as the ‘golden thread’ running through risk identification and assessment for domestic violence. It is likely therefore that risk will heighten where a perpetrator increasingly engages in multiple forms of violence or abuse, or does so more frequently, intensely or severely. Judicial recognition of this is critical to understanding the ongoing and ever-changing risks of domestic and family violence unique to the circumstances of each case before the court, and the need to regularly reassess risk throughout the course of judicial proceedings. It is also important to recognise, more broadly, that certain groups within the community as identified in this bench book may be at greater risk of experiencing domestic and family violence, and may be more vulnerable to its impacts. Some people may belong to multiple groups that have been identified as being particularly at risk of domestic and family violence.

It has become increasingly common for agencies, including police, to employ risk assessment tools in an attempt to identify victims at risk of escalation of domestic and family violence victimisation, and in particular, intimate partner homicide. Those tools are only effective if they accurately identify those victims who are at significant risk. Recent research has observed that risk assessment tools which fail to take account of the gendered nature of domestic and family violence may be ineffective.

  • Victim fear
    According to an evaluation of risk assessment tools carried out in NSW the victim’s self-perception of risk of future intimate partner violence is one of the strongest predictors of future intimate partner violence. A victim’s intuitive sense of being in danger has been identified as a key lethality indicator present in 53.2 per cent of intimate partner violence related homicides recorded in Queensland between 2011 and 2018. It is vital that victims are believed, regardless of how agencies assess their presentation. The Queensland Death Review and Advisory Board observed that agencies had assessed risk with a focus on the victims’ presentation rather than victims’ expressed concerns for their safety, and resulted in failures to refer victims to High-Risk teams for support. The Queensland Death Review and Advisory Board further observed that victims who do not fit agency understandings of the “ideal victim” may not have their fears taken seriously if they present as ‘jovial’ or ‘happy’, resulting in miscalculation of actual level of risk or potential future harm. Regardless of presentation, victim fear and help-seeking behaviours were the highest level lethality indicator observed across the period examined in the Queensland Death Review and Advisory Board Annual Report 2021-22.
  • Past domestic and family violence and escalation

    The risk of life-threatening injury or death is reported to be higher where the past violence experienced by a victim occurred within the last year and included at least one incident where the perpetrator used or threatened to use a firearm or knife or strangled or choked the victim, or where the perpetrator made a death threat of any kind to the victim, or where the frequency or severity of incidents of threatened or actual physical violence increased in the lead up to the life-threatening injury or death. Some victims however may never experience any form of actual or threatened physical violence and yet may still be at risk of death; in some reported cases, the homicide is the first incident. However, in many of these cases the homicide was preceded by coercive and controlling behaviour. In these cases, there may be other important signifiers of risk evident in the perpetrator’s behaviour, such as: physical violence outside the intimate relationship; misuse of alcohol or drugs; intense jealousy towards the victim; or exercising a high level of prolonged control over the victim’s daily activities and life. Research demonstrates links between the experience of domestic and family violence and the development of mental health conditions in victims and this may place victims at risk of self-harm and suicide.

    An escalation in victim experience of domestic and family violence was observed in 37 percent of intimate partner homicides recorded in Queensland between 2011 and 2018. The presentation of victims to agencies on multiple occasions may be an indication that violence is escalating in frequency and severity, heightening risk of death and serious injury. Indicators of escalation may include:

    • Victims taking further protective action, such as seeking to vary a protection order by strengthening conditions may indicate a change in circumstances or that current conditions are ineffective;
    • Victim action to secure safety of children through legal processes may increase potential risk of violence.
  • Non-fatal strangulation
    US research indicates that women who had experienced non-fatal strangulation by the perpetrator in the last year were twice as likely to be killed as women who had not. Women who had experienced non-fatal strangulation were also six times more likely to be a victim of attempted murder by their abusive partner. Strangulation is sometimes referred to as garrotting or choking. Strangulation has been identified as one of the behaviours that often forms part of a pattern of behaviours underpinning coercive control. A recent Western Australian study highlights the strong association between non-fatal strangulation and intimate partner sexual assault.
  • Weapons and threats to kill

    Limited Australian research in this area necessitates referencing Canadian and US sources. It is important to note that firearms are more prevalent and easier to obtain in the US than in Canada and Australia. A Canadian study found that where a firearm was present in the home, the risk of severe harm caused by weapons was heightened. This was the case even though the firearm was generally not used, and the harm was caused by another kind of weapon.

    American research indicates that the severity of abuse related harm is significantly heightened when weapons are involved. Studies found that women whose abusers used or threatened use of a weapon were 20 times more likely to be killed (with or without a weapon) than women whose abusers did not use or threaten weapon use. A Chicago study of women subjected to lethal and non-lethal harm found that 23 per cent of abuse incidents involving a firearm had a lethal outcome. In the same study, 35 per cent of incidents involving a knife had a lethal outcome. An analysis of the results from 17 studies that measured completed or attempted intimate partner homicide found that the perpetrator’s direct access to guns increased the likelihood of intimate partner homicide compared to intimate partner violence by 11 times.
  • Separation

    Victims, and their family and friends, may not always recognise domestic and family violence where there has been a prolonged history of controlling behaviours by the perpetrator and no acts of physical violence or harm. For some victims, leaving an intimate relationship will be the first time they identify an experience of domestic and family violence by their former partner, although research suggests that usually there are indications of controlling behaviours within the relationship prior to separation. Where violence has occurred during the relationship, it is common for perpetrators to continue or escalate the violence after separation in an attempt to gain or reassert control over the victim, or to punish the victim for leaving the relationship. Where women leave an intimate relationship and first experience or continue to experience violence after separation, their former partner may experience an intense sense of loss of control and the violent response may be severe, life threatening or lethal. The Queensland Domestic and Family Violence Review and Advisory Board in its 2022 Annual Report noted a strong correlation between separation and homicide. Between 1 July 2016 and 30 June 2022, 50.6 per cent of Queensland victims of intimate partner homicide were known to have separated (40.5 per cent) or intended (10.1) to separate from the perpetrator. In the Victorian Systemic Review of Family Violence Deaths 1 January 2011 to 31 December 2015 42.1 per cent of the 38 family violence intimate partner homicide incidents in Victoria between 2011 and 2015 involved individuals who had separated, 23.7 per cent in the three months preceding their death and 15.8 per cent involved individuals with intention to separate or separation pending.

    The NSW Domestic Violence Death Review Team Annual Report 2019-2021 recorded that in 61.3 percent of all intimate partner homicides where a female was killed by a former partner, the victim and perpetrator had separated within three months of the killing and where a female was killed by a current partner in 51.1 per cent of cases one or both parties had indicated an intent to end the relationship, most often within three months of the killing, concluding that in over two thirds of intimate partner homicides involving a female victim and male offender the relationship had ended or was breaking down. This indicates that the period during and directly after separation may be high risk for women in relationships involving domestic and family violence. The Australian Bureau of Statistics reported that in 2019 domestic and family violence-related homicide victims accounted for over a third of the total number, and females accounted for almost two-thirds of all victims.

    Importantly, in the context of separating parents, there is also an increased risk of harm to children’s psychological and physical wellbeing due to exposure to domestic violence, history of maltreatment, parental stress, social isolation of the family, and inadequate resources and support.
  • Pregnancy of victim
    The NSW Legislative Council Standing Committee on Social Issues found that pregnant women are 230 per cent more likely than non-pregnant women to experience domestic and family violence. 47.5 per cent of the respondents in the 2016 Australian Bureau of Statistics Personal Safety Survey had experienced violence by a former partner while pregnant. For 24.1 per cent of these respondents, pregnancy was the first time they had experienced violence. Of the respondents who had experienced violence by a current partner, 18.per cent experienced the violence during pregnancy, 5.2 per cent for the first time. Notably in the United States homicide is a leading cause of death during pregnancy and the postpartum period – with the majority of these homicides occurring in the home-pregnancy and the postpartum period are recognised as times of elevated risk for homicide among all females of reproductive age.
  • Misuse of alcohol or drugs by perpetrator
    National death review data indicates that problematic substance abuse was a factor in over 60 per cent of homicides where a male offender killed a female victim. A quarter of male homicide offenders engaged in problematic drug and alcohol use in the lead up to or at the time of the homicide, over half engaged in problematic alcohol use at the time of the homicide (27.1 per cent used alcohol and no other drug) and one third engaged in problematic drug use in the lead up to or at the time of the homicide (8.3 per cent engaged in problematic drug use only). The Victorian Systemic Review of Family Violence Deaths 1 January 2011 to 31 December 2015 found that 35.3 per cent of family violence homicide offenders misused substances at the time of the fatal incident. The accessibility of alcohol also has a relationship to the risk of violence. Australian research indicates that rates of violence are heightened in areas where takeaway alcohol is available for purchase.
  • Stalking
    In assessing the relative significance of risk factors for predicting lethal or near lethal harm to victims of domestic and family violence, a large sample size study across ten US cities revealed that significantly more of the women killed or nearly killed than those who were not had histories of stalking in their abusive relationships. The same study indicated that being “followed or spied on” by the abuser in the 12 months before the lethal or near lethal incident resulted in a nearly 2.5-fold risk. A later study showed that psychological abuse and stalking contributed uniquely to the prediction of severe injuries. The New South Wales Domestic Violence Death Review Team’s 2021 report identified that almost a third of the 245 female intimate partner homicide victims killed by a male predominant abuser had been stalked during the relationship. In over half of the 109 cases where the relationship had ended the male predominant abuser stalked the female predominant victim after the relationship had ended.
  • Coercive and controlling, jealous, obsessive behaviours by the perpetrator

    Findings from the New South Wales Domestic Violence Death Review Team’s 2019-2021 Report indicate a strong correlation between accounts of coercive control in relationships prior to homicide; of male homicide perpetrators identified as the predominant abuser:

    • 96.7 per cent used emotional and/or psychological abuse against the female predominant victim before the homicide, including verbal denigration, threats regarding child custody, victim-blaming, gaslighting, exploiting a victim’s mental illness, unfounded allegations of infidelity and threats of self-harm or suicide to control a victim.
    • 15.1 per cent had an identifiable history of sexual abuse (this figure is suspected to not reflect the true prevalence of sexual abuse in these relationships);
    • 59.2 per cent used social violence against the female predominant victim, including controlling the victim’s contact with family and friends, abuse, threats or rudeness to the victim’s friends or family, intentional relocation away from support networks, friends and family, restricting access to transport, and controlling the victim’s personal appearance;
    • 31.4 per cent used economic and/or financial violence against the female predominant victim, including withholding and controlling use of and access to money, scrutinising spending and unrealistic expectations for expenditure on necessities, preventing the victim from working or controlling wages and coerced debt.
    • 30.6 per cent stalked the female predominant victim during the relationship and 51.4 per cent did so after the relationship had ended, including physical following, hiring someone to surveil the victim, loitering near or breaking into the victim’s home or work, reading a victim’s diary, using devices to track the victim’s location, persistently messaging a victim’s phone or social media, maintaining surveillance over the victim’s phone, email or other accounts, covertly recording the victim’s activities, and engaging with the victim on social media/dating sites under a false identity.

    In addition, 72.7 per cent of male predominant abusers used physical violence against the female predominant victim before her death.

    Analysis of the National Death Review Network’s National Minimum Dataset identified the following types of coercive and controlling behaviour exhibited by male primary domestic violence abusers who killed female domestic violence victims in its 2022 report:

    • 81.6 per cent emotionally and/or psychologically abused their victim;
    • 63.2 per cent socially abused their victim;
    • 41.5 per cent stalked their victim;
    • 27.4 per cent financially abused their victim;
    • 16 per cent were identified as having sexually abused their victim (this is recognised as likely an undercount of the true incidence of sexual violence).

    This was in addition to the 79.7 who used physical violence against their victim.

    A recent study which interviewed loved ones of female intimate partner homicide victims found that coercive control was present in all informant’s narratives of the relationship between the victims and homicide perpetrators. The Victorian Systemic Review of Family Violence Deaths 1 January 2011 to 31 December 2015 found that 78.9 per cent of domestic and family violence homicide victims were identified as having a prior history of domestic and family violence victimisation at the hands of the homicide perpetrator.

    The controlling behaviours referred to in the previous paragraphs (sometimes referred to as coercive control) may involve the perpetrator spatially confining or restraining the victim; asserting exclusive possession over the victim; monopolising the victim’s skills and resources; restricting the victim’s access to finances and employment; or preventing the victim from keeping in touch with social networks, escaping the abusive relationship, or seeking help and support. The effect may be to physically and socially isolate the victim, and, over time, undermine the victim’s sense of identity, independence and self worth, and place them at greater risk of further domestic and family violence.
  • Suicide threat by perpetrator
    There is evidence to suggest a correlation between severe domestic and family violence and threat of suicide by the perpetrator. The NSW Domestic Violence Death Review Team Annual Report 2017-2019 recorded that 23 per cent of all male intimate partner perpetrators suicided after murdering their intimate partners. In the period 2000-2010 in Victoria, of the 31 homicide-suicides, 17 involved intimate partners and the Victorian Systemic Review of Family Violence Deaths 1 January 2011 to 31 December 2015 found that 14 per cent of the 85 family violence homicide perpetrators suicided at the time of the incident or subsequently. The Ombudsman Western Australia reported that, in the period 2012-2015, of the 35 domestic and family violence related deaths, 8 were homicide-suicides where in all cases the male killed the female and subsequently suicided.
  • Step-child in the family
    A number of studies have investigated the prevalence and severity of domestic and family violence against women with children fathered by someone other than the perpetrator. A recent comparative study found that women with some or all children not fathered by the perpetrator were 30 per cent more likely to experience lethal harm and 20 per cent more likely to experience at least one life-threatening violent incident than women whose children were all fathered by the perpetrator; further, these women represented 65 per cent of the homicide victims in the study. An earlier study found that the presence of a step-child in the family more than doubled the risk of the mother being killed. A recent Swedish study observed that the presence of any children (not just step-children) was related to a higher risk rating for imminent intimate partner violence re-victimization and warranted recommendations of more than standard levels of risk management strategies.
  • Parenting proceedings and other court proceedings
    An Australian study found that domestic and family violence is a common experience among separated parents, with mothers reporting physical or emotional abuse in greater proportions than fathers. Another study reported high rates of ongoing fear and abuse associated with post-separation parenting arrangements and decision-making. A 2015 court outcomes project found that more than one third of all parenting matters filed in the then Family Court of Australia and then Federal Circuit Court of Australia raised allegations of family violence, while a study of the experiences of separated parents found that a significant proportion of mothers who had experience family violence may have difficulty corroborating it because they had not previously disclosed it to police or other agencies. In these circumstances, a parent may use their joint parenting role or related judicial options as a means of exercising ongoing control over their former partner and may be considered as systems abuse and part of a pattern of abusive behaviours aimed at controlling the victim, sometimes referred to as coercive control. The Australian Domestic and Family Violence Death Review Network Data Report found that protection orders were a feature in over 40 per cent of the 240 cases where a male intimate partner homicide offender killed a female intimate partner, demonstrating a history of police or court intervention due to domestic or family violence, and in 49 of those cases there was a current protection order in place at the time of the homicide, the vast majority naming the female victim as the protected person. Active family law proceedings were identified in only 3.8 per cent of the total 290 cases in the focused dataset, reflecting the fact that many disputes relating to separation, property division and child arrangements are settled informally. It notes that the Queensland Domestic and Family Violence Death Review Board has identified that there is no process whereby the family court system is notified of domestic violence deaths involving parties engaged with the court.