Australian Domestic and Family Violence Death Review Network Data Report: Intimate partner violence homicides 2010–2018

This report addresses work covered in the ANROWS research project 4AP.9 “Australian Domestic and Family Violence Death Review Network national data update”. Please consult the ANROWS website for more information on this project.

ANROWS research contributes to the six National Outcomes of the National Plan to Reduce Violence against Women and their Children 2010–2022. This research addresses National Plan Outcome 6 – Perpetrators stop their violence and are held to account.

Suggested citation

Australian Domestic and Family Violence Death Review Network, & Australia’s National Research Organisation for Women’s Safety. (2022). Australian Domestic and Family Violence Death Review Network Data Report: Intimate partner violence homicides 2010–2018 (2nd ed.; Research report 03/2022). ANROWS.

This report is the result of a partnership between Australia’s National Research Organisation for Women’s Safety (ANROWS) and the Australian Domestic and Family Violence Death Review Network, comprising membership
from the:

• Victorian Systemic Review of Family Violence Deaths, Coroners Court of Victoria;
• Domestic Violence Death Review Team, NSW Department of Communities and Justice;
• Domestic and Family Violence Death Review Unit, Coroner’s Court of Queensland;
• South Australian Coroner’s Court and Office for Women;
• Ombudsman Western Australia;
• Northern Territory Coroner’s Office;
• Magistrates Court of Tasmania, Coronial Division; and
• Office of the Coordinator General for Family Safety, Australian Capital Territory.

The Australian Domestic and Family Violence Death Review Network would like to acknowledge Celeste Koens, Project Officer at ANROWS for her work on this report.

The authors extend their thanks to the anonymous peer reviewers for their valuable feedback, and in particular the insights provided by our First Nations reviewers who helped us strengthen this report.

Most importantly the authors acknowledge all those who have tragically lost their lives to domestic and family violence and extend our sincere condolences to their surviving families and friends.

Acknowledgement of lived experiences of violence

ANROWS acknowledges the lives and experiences of women and children affected by domestic, family and sexual violence across Australia. We recognise the individual stories of courage, hope and resilience that form the basis of ANROWS research.

Caution: Some people may find parts of this content confronting or distressing. Recommended support services include 1800 RESPECT (1800 737 732) and Lifeline (13 11 14).

Contents

Acronyms

Definitions and concepts

Executive summary

Key data findings

Introduction

Domestic and family violence death review mechanisms

The Australian Domestic and Family Violence Death Review Network

National data on intimate partner violence homicide

Fourth Action Plan

Methods

Study design and setting

Data sources

Case identification and inclusion

Data collection

Data extraction and coding

Data analysis

Limitations

Results

Overview

Male IPV homicide offenders, 2010–2018

Female IPV homicide victims, 2010–2018

Female IPV homicide offenders, 2010–2018

Male IPV homicide victims, 2010–2018

IPV homicide and children

Domestic violence characteristics preceding IPV homicides

Discussion

Dataset strengths and limitations

Key findings

Conclusion

References

Appendix A: Australian Domestic and Family Violence Death Review Network: Terms of reference

Background and position summary

Domestic and Family Violence Death Review context

Australian Domestic and Family Violence Death Review mechanisms

Common elements of review teams

ADFVDR Network overview

Partner project with Australia’s National Research Organisation for Women’s Safety (ANROWS)

Appendix B: Australian Domestic and Family Violence Death Review Network: Domestic and Family Violence Homicide Consensus Statement

Background and purpose

Definitions

Surveillance

Categorisation

Consideration 1: Case type

Consideration 2: Intent

Consideration 3: Relationship

Consideration 4: Domestic and family violence context

Case inclusion criteria for the National Minimum Dataset on intimate partner homicides

Appendix C: Australian Domestic and Family Violence Death Review Network: Data sharing protocols

Purpose

Background

Jurisdictional governance

Partnership project with ANROWS

Issues

Shared specifications

Data dissemination

Review

Acronyms

Acronym	Definition
ANROWS	Australia’s National Research Organisation for Women’s Safety
IPV	Intimate partner violence
LGBTQ	Lesbian, gay, bisexual, trans and gender diverse, queer and questioning
NCIS	National Coronial Information System
NGMI	Not guilty by reason of mental illness
NMDS	National Minimum Dataset

Definitions and concepts

Concept	Definition
Abuser	A person who uses domestic and family violence behaviours against a victim.
Assault	The Australian Domestic and Family Violence Death Review Network (the Network) defines assault as an injury from an act of violence where physical force by one or more persons is used with the intent of causing harm, injury or death to another person; or an intentional poisoning by another person.
Coercive control	A concept that reflects the “multidimensionality of oppression” in the lives of women experiencing domestic and family violence (Stark, 2007, p. 10). Coercive control recognises the “various means to hurt, humiliate, intimidate, exploit, isolate, and dominate their victims over time”, which include non-physical and/or physical tactics (Stark, 2007, p. 5).
Cross-domestic violence order	A domestic violence order where both parties are named as protected persons and respondents in the order (see “Domestic violence order”).
Domestic and family violence	The National Plan to Reduce Violence against Women and their Children 2010–2022 (the National Plan) defines domestic violence as acts of violence that occur between people who have, or have had, an intimate relationship. While there is no single definition, the central element of domestic violence is an ongoing pattern of behaviour aimed at controlling a partner through fear, for example by using behaviour which is violent and threatening. In most cases, the violent behaviour is part of a range of tactics to exercise power and control over women and their children, and can be both criminal and non-criminal. Domestic violence includes physical, sexual, emotional and psychological abuse. (Council of Australian Governments, 2011) In addition to these abusive behaviours, this study also considers social abuse and economic or financial abuse. Domestic and family violence can also occur outside of intimate partner relationships, including between other family members, such as between parents and children, between siblings, or between other extended family members or kinship ties. The Network’s definition of domestic and family violence (as per the Homicide Consensus Statement in Appendix B) generally aligns with the definition in the National Plan. That is, domestic and family violence includes a spectrum of physical and non-physical abuse within an intimate or family relationship. Domestic and family violence behaviours include physical assault, sexual assault, threats, intimidation, psychological and emotional abuse, social isolation, and economic deprivation. Primarily, domestic and family violence is predicated upon inequitable relationship dynamics in which one person exerts power and coercive control over another.[1]
Domestic violence order	A civil order the object of which is to protect victims – or persons at risk – of domestic and family violence from another person with whom they are, or have been, in an intimate or familial relationship. The term includes provisional, interim and final orders. This is also referred to as an apprehended domestic violence order, family violence order, family violence intervention order, family violence restraining order, protection order or intervention order.
Domestic violence victim	A person who has domestic and family violence behaviours used against them.
Economic or financial abuse	The Fourth Action Plan of the National Plan defines financial abuse to occur “when another person manipulates decisions or controls access to money or property without consent. Financial abuse can include someone taking control of household finances, limiting access to funds or forcing someone to spend money or sell property” (Council of Australian Governments, 2019). For the purpose of this research, economic or financial abuse includes a spectrum of abusive behaviours intended by an abuser to diminish a victim’s ability to support themselves and that forces them to depend on the abuser financially.
Emotional or psychological abuse	Plan defines psychological and emotional abuse as “a range of controlling behaviours such as control of finances, isolation from family and friends, continual humiliation, threats against children or being threatened with injury or death” (Council of Australian Governments, 2011). The Fourth Action Plan further defines psychological abuse as actions that are used to threaten, intimidate, harass, belittle and humiliate someone else. It can include threats of violence or death toward a woman or to her children, family, friends, work colleagues or pets. It can also include isolating women from family and friends, yelling, damaging property, driving at excessive speed, making unfounded accusations of infidelity, interrogating someone and making threats of self-harm or suicide if the woman attempts to leave. (Council of Australian Governments, 2019) The Network includes the majority of these behaviours as emotional or psychological abuse, with the addition of verbally denigrating the victim, making threats regarding custody of children as a means to control the victim, blaming the victim for all adverse events, fabricating or exploiting a victim’s mental illness, and deliberately creating dependence. Isolating the victim from family and friends is excluded from the definition of emotional and psychological abuse as it is captured separately under “Social abuse”.
Family law proceedings	Proceedings commenced in the Family Court of Australia or in the Federal Magistrates’ Court (all states except for Western Australia). In Western Australia, this refers to proceedings commenced in the Family Court of Western Australia.
Gender	The term “gender” is used in this report to indicate people’s gender identity notwithstanding their biological sex classification. The Fourth Action Plan defines gender identity as “a person’s deeply felt sense of being male, female, both, in between, or something other. Everyone has a gender identity” (Council of Australian Governments, 2019). This report acknowledges that people’s biological sex may differ from their gender identity. The term also more comprehensively reflects the gendered nature of domestic and family violence related to the socially constructed classifications and characteristics attributed in particular to male and female sex categorisations.
Homicide	Includes all circumstances in which an individual’s intentional act, or failure to act, resulted in the death of another person, regardless of whether the circumstances were such as to contravene provisions of the criminal law.
Homicide offender	The person whose actions inflicted the injuries to the homicide victim that caused their death/homicide.
Homicide victim	The person who died because of the injuries inflicted by the homicide offender.
Intimate partner violence	A pattern of behaviour whereby one person intentionally and systematically uses violence and abuse to gain and maintain power over another person with whom they share, or have previously shared, an intimate relationship (see “Domestic and family violence”).
Intimate partner violence homicide	A homicide that occurs between individuals who are or have been in an intimate relationship following an identifiable history of domestic violence. See also “Homicide” above.
Mechanism of homicide	The manner by which a person perpetrates the homicide against another person, or the way in which one person kills another person. Can include methods such as assault with a sharp weapon, assault with a blunt weapon, assault with no weapon, suffocation/strangulation, or homicide by firearm.
Physical violence	The National Plan defines “slaps, shoves, hits, punches, pushes, being thrown down stairs or across the room, kicking, twisting of arms, choking, and being burnt or stabbed” as forms of physical violence (Council of Australian Governments, 2011). The Fourth Action Plan further includes whipping, hitting with objects, stomping and damaging property and specifies that “physical violence can be fatal due to physical injury intentionally caused by the perpetrator, or unintended consequences of physical abuse inflicted by the perpetrator” (Council of Australian Governments, 2019).
Primary DV abuser	The person who primarily initiated domestic violence in the life of the relationship and/or was the main aggressor of domestic violence after the relationship had ended. This term is designed to highlight that a person may have been the primary user of domestic violence prior to the homicide, and the homicide may have been perpetrated by a person who was typically a victim of domestic violence (for instance, a victim who kills an abuser in self-defence).
Primary DV victim	The person who primarily had domestic violence used against them (was victimised) during the relationship with an abuser, or after that relationship had ended. The term designates a person who experienced, but did not initiate, domestic violence. This term is designed to highlight that a person may be the primary victim of domestic violence prior to the homicide, but may ultimately perpetrate the homicide (for instance, a domestic violence victim who kills an abuser in self-defence).
Protected person	The person who is named as the protected person under an existing domestic violence order (see “Domestic violence order”).
Respondent	The person against whom an existing domestic violence order is made (see “Domestic violence order”).
Sexual abuse	Unwanted or non-consensual sexual behaviours used by an abuser against a victim. According to the National Plan, “sexual assault or sexual violence can include rape, sexual assault with implements, being forced to watch or engage in pornography, enforced prostitution, and being made to have sex with friends of the perpetrator” (Council of Australian Governments, 2011).
Social abuse	A range of abusive behaviours designed to prevent a person from spending time with family and friends and participating in social activities. Socially abusive behaviours often isolate victims, allowing abusers to maintain control over them.
Stalking	A range of tactics whereby an abuser intentionally and persistently pursues a victim in order to control or intimidate that victim or seek to make the victim fearful. Stalking behaviours can include the abuser following the victim, loitering near the victim’s home or work, and breaking into the victim’s residence. Stalking also includes acts of technology-facilitated abuse such as persistent text messaging, maintaining surveillance over the victim’s phone or email, covertly recording the victim’s activities, and engaging with the victim on social media/dating sites under a false identity. Stalking can occur both during an intimate relationship and after a relationship has ended.
Systems abuse	The manipulation of legal systems by an abuser to exert power or control over a victim. This may include the misapplication of domestic violence orders against a victim.
Transgender	Commonly shortened to “trans”, this umbrella term describes a person who does not identify with their gender assigned at birth or upbringing (Ussher et al., 2020).
Verbal abuse	A range of verbally abusive behaviours used by an abuser to belittle or denigrate a victim (see “Emotional or psychological abuse”).

Executive summary

Domestic and family violence is a complex phenomenon characterised by ongoing, systematic patterns of behaviour used by abusers against their victims that may include physical, sexual, emotional, psychological, verbal, social or financial abuse (Council of Australian Governments, 2011). Domestic and family violence may manifest in a range of different familial or kinship relationships, for example, between intimate partners, siblings, or children and parents. However, in the overwhelming majority of cases domestic and family violence is perpetrated by a man against his current or former female intimate partner (Australian Bureau of Statistics, 2017; Council of Australian Governments, 2019; Cox, 2015). Recently, the term coercive control has been adopted to reflect the “multidimensionality of oppression” in the lives of women experiencing domestic and family violence (Stark, 2007, p. 10). Conceptualised in this way, coercive control recognises that domestic and family violence perpetrators use “various means to hurt, humiliate, intimidate, exploit, isolate, and dominate their victims over time” (Stark, 2007, p. 5).

Domestic and family violence can also be fatal. In Australia in 2018–19, intimate partner homicides accounted for 21 per cent of all homicides and for 62 per cent of all domestic homicides (Bricknell & Doherty, 2021).^[2] A significant proportion of domestic homicides occurs in a context of domestic and family violence, meaning there is an identifiable history of abuse between the parties that precedes the fatal episode (Australian Domestic and Family Violence Death Review Network, 2018).

The Australian Domestic and Family Violence Death Review Network (the Network) was established in 2011 to analyse and improve knowledge about deaths that occur in a context of domestic and family violence, and to share findings and recommendations across jurisdictions in order to improve the response system and thereby prevent future deaths. Under this mandate, the Network developed a first-stage National Minimum Dataset (NMDS) to examine national trends and patterns with respect to intimate partner homicides preceded by a reported or anecdotal history of domestic and family violence (IPV homicides).^[3] In 2018, the Network published the inaugural Australian Domestic and Family Violence Death Review Network data report which presented NMDS data for IPV homicides occurring between July 2010 and June 2014.

In 2020, Australia’s National Research Organisation for Women’s Safety (ANROWS) was funded to work in partnership with the Network to produce the second edition of the report. This report updates and builds on the data presented in the 2018 report, providing data findings from the NMDS on IPV homicides from July 2010 to June 2018. Intimate partner homicides where there was no identifiable history of domestic and family violence do not form part of this dataset.

The NMDS uses a retrospective population-based case series analysis to capture key trends that present in IPV homicide cases in Australia. Data is sourced from jurisdictional domestic and family violence death review teams in New South Wales, the Northern Territory, Queensland, South Australia, Victoria and Western Australia; and from the National Coronial Information System for the Australian Capital Territory and Tasmania. The dataset is informed by case reviews, which are drafted by the jurisdiction’s death review team, as well as source material such as coronial files, briefs of evidence, police reports, media reporting, sentencing remarks and agency records.

Improving our understanding of the characteristics and dynamics that precede an IPV homicide can help to guide reform of the domestic and family violence response system. This work seeks to contribute to the formation of evidence-based policy and decision-making in relation to domestic and family violence, enhancing opportunities for intervention and prevention so as to improve the safety and supports for victims of violence and hold abusers to account.

Key data findings

Complete IPV homicide dataset (n=311)

• Between 1 July 2010 and 30 June 2018, there were 311 IPV homicides across Australia.
• More than three quarters of all cases involved a male IPV homicide offender killing a current or former female partner (n=240, 77.2%).
  The vast majority of those male offenders had been the primary user
  of domestic violence behaviours against the woman they killed
  (n=227, 94.6%).
• Less than one quarter of all cases involved a female IPV homicide offender killing a current or former male partner (n=65, 20.9%). Even though the female partner was the homicide offender, in the majority of these cases she was also the primary domestic violence victim, who killed her male abuser (n=46, 70.8%).
• In six cases, a male IPV homicide offender killed a male partner. Of these, three homicide offenders were the primary abuser against the partner they killed; two offenders were the primary victim of abuse; and in one case both parties mutually used domestic violence against each other.
• There were no cases identified in this dataset where a female IPV homicide offender killed a female partner.
• IPV homicide occurs across a broad age range. There was an age range of 18 to 82 years for male offenders and 18 to 75 years for female homicide offenders. Homicide victims’ ages ranged from 16 to 78 years for female victims and from 18 to 76 years old for male homicide victims.
• The duration of relationship between homicide offenders and victims in this dataset ranged from less than a year to 45 years. This demonstrates that IPV homicides can occur at any stage during a relationship.
• The majority of IPV homicide offenders (n=183, 60%) engaged in problematic drug and/or alcohol use. Importantly, this finding does not purport to identify problematic substance use as a causative factor for IPV homicide, but rather represents a pattern of behaviour and identifies possible sites of intervention.
• Only approximately one third of all IPV homicide offenders and victims were engaged in paid employment at the time of the homicide (n=225 of 622, 36.2%). This is significant because workplaces can offer an additional site of intervention for domestic and family violence.
• The data demonstrates an overrepresentation of Aboriginal and Torres Strait Islander people as both IPV homicide victims and offenders. While acknowledging these high rates, it is important to recognise that domestic and family violence is not a part of Aboriginal and Torres Strait Islander cultures, and there is a complex range of interrelated factors associated with the disproportionate incidence and severity of family violence in Aboriginal and Torres Strait Islander communities (discussed in the report below).

Male IPV homicide offenders who killed a female intimate partner (n=240)

• Most male IPV homicide offenders killed a current female partner (n=154, 64.2%). Fewer killed a former female partner (n=86, 35.8%).
• The majority of male IPV homicide offenders killed their current or former female partner in her home (n=151, 62.9%). In 97 of these cases (64.2%) this was a home the woman shared with the offender and in 54 cases (35.8%) this was the home the woman lived in but did not share with the offender.
• The most common criminal justice outcome for male IPV homicide offenders who killed a female victim was a murder conviction (n=121, 63.0%). Forty-four male homicide offenders suicided after the homicide (18.3%) and in the majority of these cases, they suicided within 24 hours of the homicide (n=32, 72.7% of male offenders who suicided).

Female IPV offenders who killed a male intimate partner (n=65)

• Most female IPV homicide offenders killed a current male partner (n=50, 76.9%). Fewer killed a former male partner (n=15, 23.1%).
• Over two fifths of female IPV homicide offenders killed their partner in their shared residence (n=28, 43.1%). In nine cases the homicide occurred in the male partner’s home and in a further nine cases the homicide occurred in the female homicide offender’s home (13.8% respectively).
• The most common criminal justice outcome for female IPV homicide offenders was a manslaughter conviction (n=40, 62.5%). One female IPV homicide offender suicided after the homicide.

IPV homicide and children

• There were four cases in which children were killed together with their mother, resulting in the deaths of eight children.
• Of the 311 IPV homicides examined in this dataset, there were at least 172 children under the age of 18 who survived the homicide involving one, or both, of their parents.

Focused IPV homicide dataset (n=292)

Domestic violence death review teams are uniquely positioned to conduct in-depth analysis and reviews so as to identify discrete characteristics present within a relationship prior to an IPV homicide. Drawing on data from those jurisdictions with a formalised death review mechanism in place, this report presents focused data findings around IPV homicide characteristics relating to separation or intention to separate, family law proceedings, domestic violence orders, and the nature of domestic violence and abusive behaviours used by the abuser prior to the homicide.

This focused subset of cases includes 224 cases where a male IPV homicide offender killed a female victim; 62 cases where a female IPV homicide offender killed a male partner; and six cases where a male IPV homicide offender killed a male partner.

Separation as a characteristic of IPV homicide

Male IPV homicide offenders who killed a female partner (n=224)

• In about a third of the cases where a male IPV homicide offender killed a female victim (n=77, 34.4%) the relationship had ended prior to the homicide. In more than half of these cases, the relationship had ended within three months of the fatal episode of violence (n=44, 57.1% of separated couples).
• Of the 147 cases where the relationship was ongoing, one or both parties had expressed an intention to separate in 53 cases (36.1%). The overwhelming majority of these cases involved the female homicide victim indicating an intention to separate from the male offender who killed her (n=50, 94.3%).
• Accordingly, actual or intended separation was a feature in more than half of the cases where a male IPV homicide offender killed a female partner (n=130, 58.0%).

Female IPV homicide offenders who killed a male partner (n=62)

• In just under a quarter of cases where a female IPV homicide offender killed a male partner the relationship had ended prior to the homicide (n=14, 22.6%). Five of these separations occurred less than three months prior to the homicide (35.7% of separated relationships).
• Of the 48 cases where the relationship was ongoing, one or both parties had indicated an intention to leave the relationship in 14 cases (29.2%). In the majority of these 14 cases, it was the female homicide offender who had indicated an intention to separate from the male partner she then killed (n=8, 57.1%).
• Accordingly, actual or intended separation was a feature in just under half of the cases where a female IPV homicide offender killed a male intimate partner (n=28, 45.2%).

Domestic violence orders

Male IPV homicide offenders who killed a female partner (n=224)

• Current or historical domestic violence orders were evident in 96 cases where a male IPV homicide offender killed a female partner (42.9%). Accordingly, in 128 cases there was no evidence of a current or historical domestic violence order (57.1%).
• In 49 cases, a current domestic violence order was in place between the male IPV offender and female victim at the time of the homicide (21.9%). The vast majority of these orders named the female homicide victim as the person in need of protection from the male homicide offender (n=44, 89.8% of cases with a current order). In two cases (4.1%) there were cross-orders in place at the time of the homicide where both the male IPV homicide offender and the female victim were named as needing protection from the other. In three cases (6.1%) the male IPV homicide offender was named as the person in need of protection from the female partner they killed.
• Historical domestic violence orders between the male homicide offender and female victim were a feature in 67 cases (29.9%). In 53 of these cases, the female homicide victim was named as the protected person from the male offender (79.1%) and in 11 cases both parties were named as needing protection from the other (16.4%).

Female IPV homicide offenders who killed a male partner (n=62)

• Current or historical domestic violence orders were evident in 41 cases where a female homicide offender killed a male partner (66.1% of female-perpetrated IPV homicides).
• In 21 cases, a current domestic violence order was in place between the female homicide offender and the male homicide victim at the time of the homicide (33.9%). Twelve of these orders named the female homicide offender as the person in need of protection from the male homicide victim (57.1% of current orders); eight named the male homicide victim as the person in need of protection from the female homicide offender (38.1%); and one was a cross-order naming both parties as needing protection from each other (4.8%).
• Historical domestic violence orders between the female homicide offender and male homicide victim were a feature in 29 cases (46.8%). In 17 of these cases, the female homicide offender was named as the protected person from the male homicide victim (58.6%) and in 11 cases both parties were named as in need of protection from each other (37.9%).

Domestic violence abusive behaviours

• In the focused dataset (n=292), the vast majority of the 224 male IPV homicide offenders who killed a female victim were identified as the primary domestic violence abuser in the relationship (n=212, 94.6%). Of the 62 cases where a female IPV homicide offender killed a male partner, only a small proportion of women were identified as the primary abuser against the male partner they killed (n=5, 8.1%). In half of the six cases where a male IPV homicide offender killed a male partner, the homicide offender was identified as the primary abuser in the relationship
  (n=3, 50%).
• Of the 212 cases in which a male primary domestic violence abuser killed a female victim, the majority used emotional and psychological abuse (n=173, 81.6%) and physical abuse (n=169, 79.7%) against the female partner they killed. Over half had been socially abusive (n=134, 63.2%), just over a quarter were financially abusive (n=58, 27.4%) and far fewer were known to be sexually abusive (n=34, 16.0%).
• Stalking occurred in two fifths of the 212 cases in which a male primary domestic violence abuser killed a female victim (n=88, 41.5%). In 71 cases, the domestic violence abuser stalked the victim during the relationship (33.5%) and in 44 cases the abuser stalked the victim after the relationship ended (20.8%).

Conclusion

This research demonstrates the highly gendered nature of intimate partner violence and IPV homicides, with the male party being identified as the primary domestic violence abuser in the majority of cases where a male homicide offender killed a female victim and where a female homicide offender killed a male partner. In many cases, there had been a domestic violence order naming one or both parties as in need of protection from the other at the time of, or prior to, the homicide. The vast majority of these named the female party as needing protection from her male partner. This demonstrates that in many cases the domestic violence had been reported and there had been some level of police or court intervention prior to the homicide.

With separation being a prominent feature in over half of the male-perpetrated IPV homicides against women and almost half of the female-perpetrated IPV homicides, this research also demonstrates that the period leading up to and immediately following separation involves a heightened level of risk.

Analysis of the domestic violence behaviours used by the primary domestic violence abuser demonstrates the range of physical and non-physical violence used by abusive men to dominate and control their partner. The high prevalence of emotional and psychological abuse (such as verbally denigrating, threatening, blaming or gaslighting the victim) and social abuse (such as isolating the victim from support networks and controlling her movements) demonstrates the need for services and first responders to recognise, beyond the use of physical violence, the pattern of abusive and controlling behaviours that presents in a domestic violence relationship. Further, the diverse range of abusive tactics present in this dataset, including physical, emotional, social, financial and sexual violence, and stalking, suggests that any relationship that exhibits domestic violence, whether physical or non-physical, is embedded with a risk of lethality.

Introduction

Domestic and family violence encompasses a range of behaviours employed by perpetrators to maintain power and control over another person with whom they are in an intimate or familial relationship. Domestic and family violence is a complex phenomenon characterised by ongoing, systematic patterns of behaviour that may include physical, sexual, emotional, psychological, verbal, social or financial abuse (Council of Australian Governments, 2011). In the overwhelming majority of cases domestic and family violence is perpetrated by a man against his female current or former intimate partner (Australian Bureau of Statistics, 2017; Council of Australian Governments, 2019; Cox, 2015). However, it may also manifest as violence between same-sex partners in an intimate relationship; violence perpetrated by a woman against a male intimate partner; or violence within the family unit, including child abuse, elder abuse, violence between siblings or that from adolescent children towards their parents.

Recently, the term coercive control has been adopted to reflect the “multidimensionality of oppression” in the lives of women experiencing domestic and family violence (Stark, 2007, p. 10). Conceptualised in this way, coercive control recognises that domestic and family violence perpetrators use “various means to hurt, humiliate, intimidate, exploit, isolate, and dominate their victims over time” (Stark, 2007, p. 5). Coercive control is, therefore, not a discrete type of violence, nor does it relate only to non-physical manifestations of domestic violence. Rather it describes the context for, and intent of, such abusive behaviours. Viewed in isolation, a perpetrator’s coercive behaviours may not be recognised as abuse, particularly when such behaviour is non-physical in nature. However, when examined holistically, and recognised as pattern of intentional and systematic behaviour designed to assert and maintain control and dominance over another person, its insidious nature emerges.

Domestic and family violence occurs across all stratums of society and the experiences of victims can be influenced by other forms of structural inequality and violence, such as patriarchy, racism, sexism, homophobia and ableism (Council of Australian Governments, 2019; Koleth et al., 2020).

The disproportionate incidence and severity of family violence for Aboriginal and Torres Strait Islander peoples is evident in Australian data on domestic and family violence, including the data presented in this report (Australian Bureau of Statistics, 2019a; Australian Institute of Health and Welfare, 2019; Bricknell & Doherty, 2021; Voce & Bricknell, 2020). Accordingly, responding to and preventing family violence and supporting Aboriginal and Torres Strait Islander women has been identified as a national priority (Council of Australian Governments, 2019). Understanding domestic and family violence within Aboriginal and Torres Strait Islander communities in Australia requires a complex understanding of the ways in which colonial oppression and violence are reproduced through modern structures and institutions, impacting experiences of family violence and access to support services (Our Watch, 2018).^[4]

Domestic and family violence can also be fatal. According to the National Homicide Monitoring Program, just over one third of homicides in 2018–19 were domestic homicides (Bricknell & Doherty, 2021). Intimate partner homicide was the most prevalent form of domestic homicide, accounting for 21 per cent of all homicide incidents for that time period and 62 per cent of domestic homicides.^[5] The National Homicide Monitoring Program also demonstrates the gendered nature of intimate partner homicide, with almost three quarters of victims in 2018–19 being women (Bricknell & Doherty, 2021).

A significant proportion of domestic homicides occurs in a context of domestic and family violence, meaning there is an identifiable history of abuse between the parties that precedes the fatal episode (Australian Domestic and Family Violence Death Review Network [ADFVDRN], 2018). Domestic and family violence context deaths are, therefore, regarded as preventable, with there being opportunities for individuals, services or agencies to intervene and interrupt patterns of abuse prior to the death.

Domestic and family violence death review mechanisms

Analysing domestic violence context deaths and the patterns of behaviour that occur in the months and sometimes years prior to a death can help to inform the development of evidence-based strategies to improve the response to domestic and family violence and thereby prevent future deaths (Bugeja et al., 2013; Butler et al., 2017). Across multiple international and Australian jurisdictions such analysis is undertaken by domestic and family violence death reviews.

Domestic and family violence death reviews undertake quantitative and qualitative analyses of domestic and family violence context deaths, identifying common characteristics across cases and patterns of behaviours that precede a fatality, and examining limitations and areas for improvement in systemic responses to domestic and family violence (ADFVDRN, 2018).

Within Australia, domestic and family violence death review mechanisms are currently in place in New South Wales, the Northern Territory, Queensland, South Australia, Victoria and Western Australia. These mechanisms are located within coroners courts, ombudsman’s offices or government agencies and include members with specialist expertise in domestic and family violence analysis. They have access to extensive information not only regarding the circumstances of the death itself, but across the victim’s and perpetrator’s entire life course including their family networks, trauma histories, community interactions and service engagement. This far-reaching analysis provides insights into how victims and perpetrators are interacting with systems and services, and the nature and quality of those engagements. When examined cumulatively, these histories and interactions highlight the ways in which the system is working to hold abusers accountable and to support victims, thereby achieving a holistic system-wide view of domestic violence.

The Australian Capital Territory is in the process of establishing a formal death review mechanism and there is currently no formal death review mechanism in place in Tasmania.

The Australian Domestic and Family Violence Death Review Network

In 2011 the Australian Domestic and Family Violence Death Review Network (the Network) was established as a way of formalising and coordinating collaboration between the death review mechanisms across Australia. The Network developed goals which align with Strategy 5.2 of the National Plan to Reduce Violence against Women and their Children 2010–2022 (the National Plan) – that is, to strengthen leadership across justice systems. More specifically, the Network contributes to the strategy of “driv[ing] continuous improvement through sharing outcomes of reviews into deaths and homicides related to domestic violence” (Council of Australian Governments, 2011, p. 27). The overarching goals of the Network include:

• improving knowledge regarding the frequency, nature and determinants of domestic and family violence deaths
• identifying practice and system changes that may improve outcomes for people affected by domestic and family violence and reduce these types of deaths
• analysing and comparing themes and issues arising in domestic and family violence-related deaths
• analysing and comparing domestic and family violence death review findings and recommendations.

The Network comprises representatives from each of the established Australian domestic and family violence death review teams, namely:

• Victorian Systemic Review of Family Violence Deaths (Vic)
• Domestic Violence Death Review Team (NSW)
• Domestic and Family Violence Death Review Unit (Qld)
• Domestic Violence Unit (SA)
• Reviews Team (WA)
• Family Violence Death Review Unit (NT).

The Network recognises that Tasmania and the Australian Capital Territory are exploring the implementation of death review mechanisms within their jurisdictions.^[6] Representatives of these jurisdictions are considered standing members of the Network. See the terms of reference in Appendix A for more information about the Network members and structure.

National data on intimate partner violence homicide

Through the Network, the individual death review teams have collaborated to report on national data on intimate partner homicide. In 2018 the Network published the inaugural Australian Domestic and Family Violence Death Review Network Data Report, which provided national data with respect to all intimate partner homicides that occurred following an identifiable history of domestic violence between July 2010 and June 2014 (IPV homicides). The report was informed by the Network’s National Minimum Dataset on Intimate Partner Homicides (NMDS), which collects information relating to:

• the history of domestic and family violence between homicide offenders and victims and the types of abusive behaviours adopted by the domestic violence abusers
• details of the fatal episode
• socio-demographic characteristics of the homicide victim and offender
• relationship characteristics
• domestic violence order histories
• the number of surviving children.

Fourth Action Plan

In 2019, the Council of Australian Governments launched the Fourth Action Plan of the National Plan to Reduce Violence against Women and their Children 2010–2022 (the Fourth Action Plan). Australia’s National Research Organisation for Women’s Safety (ANROWS) was funded by the Department of Social Services to lead a program of research under this plan that continues to produce, disseminate and assist in providing evidence for policy and practice to address violence against women and their children. Under this program of research, ANROWS and the Network have worked in collaboration to produce this report – the next iteration of the Australian Domestic and Family Violence Death Review Network Data Report.

This report builds on the Network’s inaugural Data Report published in 2018 and includes IPV homicide data from 1 July 2010 to 30 June 2018. The report responds to Priority 5 of the Fourth Action Plan – improve support and service system responses – and contributes to Action 19:

Build the evidence base to inform responses to domestic, family and sexual violence by strengthening the focus on what works to reduce violence, improving data and supporting the Fourth Action Plan priorities. (Council of Australian Governments, 2011, p. 27)

The second iteration of the Australian Domestic and Family Violence Death Review Network Data Report aims to further enhance our understanding of IPV homicide in Australia. Improving our understanding of the characteristics and dynamics that precede an IPV homicide can help to guide reform of the domestic and family violence response system. This work seeks to contribute to the formation of evidence-based policy and decision-making in relation to domestic and family violence, enhancing opportunities for intervention and prevention so as to improve the safety of and supports for victims of violence and hold abusers to account.

Methods

Study design and setting

The data outlined in this report was captured through a retrospective population-based case series analysis. This study examined the deaths of people who were killed by their current or former intimate partner following an identifiable history of domestic violence in Australia between 1 July 2010 and 30 June 2018. The study includes and builds on data presented in the 2018 Australian Domestic and Family Violence Death Review Network Data Report (ADFVDRN, 2018).

Data sources

This report has been developed using data provided by jurisdictional domestic and family violence death review teams in New South Wales, the Northern Territory, Queensland, South Australia, Victoria and Western Australia. Data sources include case reviews, which are drafted by the jurisdiction’s death review team, as well as source material such as coronial files, briefs of evidence, police reports of death, media reporting, sentencing remarks and agency records.

Data for the Australian Capital Territory and Tasmania was sourced from the National Coronial Information System (NCIS) with approvals from those jurisdictions. Western Australia’s death review mechanism captured data from July 2012 onwards and the data preceding this has also been drawn from NCIS.^[7]

Case identification and inclusion

For this report, cases were identified and included in accordance with the Network’s Homicide Consensus Statement. This statement was developed and adopted by the Network in 2014 to establish a consistent definition of domestic and family violence homicide across all jurisdictions. This statement sets out the process for case identification and inclusion criteria for the NMDS and provides a framework for case categorisation having regard to the case type, the intent, the relationship between the deceased and the homicide offender, and the domestic and family violence context.

Since the Network’s publication of the 2018 Data Report, the Homicide Consensus Statement has been updated to define the following inclusion criteria for cases in this report (see Appendix B):

• the death was a result of a homicide that occurred in Australia between 1 July 2010 and 30 June 2018
• the homicide victim and homicide offender were either in a current or former intimate partner relationship
• there was an identifiable history of violence between the homicide victim and homicide offender
• the coronial or criminal proceedings in that homicide were complete on or before 31 December 2020.

Data collection

The data captured for this report is in accordance with the Network’s NMDS on IPV homicides.

The NMDS was originally developed, trialled and finalised in 2015. The NMDS identifies a set of data variables common to all jurisdictions and from this a comprehensive data dictionary was developed. Data variables captured by the NMDS include:

• socio-demographic characteristics of the homicide offender and victim, including age, gender, country of birth, visa status, employment, disability, convictions, and alcohol or other drug use
• characteristics of the relationship between homicide offender and victim, including history of domestic violence, domestic violence orders, separation or intention to separate, children in the relationship and the nature of abusive behaviours
• details of the homicide event, including location, mechanism of fatal assault and outcomes of criminal or coronial proceedings.

In 2016 the NMDS and data dictionary were disseminated to Network members to enter individual jurisdictional data for cases occurring between July 2010 and June 2014, and this data informed the first iteration of the Data Report, published in 2018. Since then, many jurisdictions have continued coding cases into the database, using the existing data dictionary. Accordingly, to preserve the integrity and reliability of the NMDS, the existing data dictionary has been utilised for this report.

Data extraction and coding

In 2017 Network members agreed to the Network’s data sharing protocols to facilitate the sharing of domestic and family violence death review data across jurisdictions, in support of the establishment of the NMDS (see Appendix C). The protocols remain current and have facilitated the sharing of data for this report.

To develop the dataset for this report, Network members from the Northern Territory, Queensland, South Australia, Victoria and Western Australia^[8] coded from source material directly into the NMDS. The project’s research officer coded the New South Wales data predominantly from case reviews prepared by the New South Wales death review team. These case reviews provide a comprehensive summary of the source material for the relevant IPV homicides, describing in detail the life course of the offender and victims. Where no case review was available, the research officer coded directly from the New South Wales source material.

The project’s research officer coded cases from Tasmania and the Australian Capital Territory into the dataset from the limited source material available in NCIS.

Data analysis

Complete IPV homicide dataset

The complete dataset was analysed using univariate and bivariate descriptive statistics. Data was disaggregated by homicide offender and victim gender to distinguish between cases where a male IPV homicide offender killed a female partner, those where a male offender killed a male partner, and those where a female homicide offender killed a male partner. Data about demographic details, relationship details and homicide characteristics presented in this report are based on the total number of cases for each disaggregated group of offender and victim.

Focused IPV homicide dataset

Domestic violence death review teams have access to a range of data sources that provide detailed information about the life course and relationship histories of homicide offenders and victims and the nature of domestic violence in the relationship. Drawing on in-depth data provided by the death review mechanisms in New South Wales, the Northern Territory, Queensland, South Australia, Victoria and Western Australia,^[9] this report includes an analysis of a focused dataset presenting data findings around IPV homicide characteristics relating to separation or intention to separate, family law proceedings, domestic violence orders, and the nature of domestic violence and abusive behaviours used by the abuser prior to the homicide.

This focused dataset excludes cases that have been coded from NCIS, namely, all cases from Tasmania and the Australian Capital Territory, and Western Australia’s cases from July 2010 to June 2012. In addition, three New South Wales cases were excluded as the original source material was not able to be accessed and reviewed within the project timelines.

Findings from the focused dataset have been carefully identified in the relevant fields in the report.

Limitations

The datasets in this report draw on the expertise of the Network and the diverse range of primary source materials that the Network has access to. The combined expertise and diversity of primary source material offers a unique opportunity to undertake in-depth analysis of the domestic violence characteristics preceding an IPV homicide, setting this dataset apart from many other studies exploring homicide in Australia.

There are challenges in using this dataset to accurately reflect the rate of IPV homicides over the eight-year reporting period. In accordance with the Network’s Homicide Consensus Statement, the IPV homicide dataset only includes cases where the coronial or legal proceedings have been finalised. Accordingly, a significant proportion of the more recent (2017–18) IPV homicides, as well as a number of older cases, have not yet been counted in the dataset because they have ongoing coronial or legal proceedings. These cases will be added to the dataset as the cases are finalised and reported in future iterations of this report.

Other limitations for this dataset include:

• challenges relating to the accurate identification of homicides as an IPV homicide, due to undisclosed or unreported domestic violence in the relationship. This may result in an undercount of the true incidence of IPV homicides in Australia
• a potential under-reporting of Aboriginal and Torres Strait Islander peoples in this dataset, due to a reliance on service data, which may not consistently collect accurate administrative data of this nature
• a potential under-reporting of people with disability in this dataset as a result of inconsistencies in the identification and definition of disability in service data
• a potential under-reporting of LGBTQ couples in this dataset, due to the relationship not being disclosed or recognised by services, families or friends prior to the homicide
• challenges relating to the accurate data collection of surviving children, especially stepchildren or children living outside the home.

These limitations will be discussed in greater detail in the “Discussion” section of this report.

Results

This chapter sets out the national data on IPV homicides in Australia. The term “domestic violence” has been adopted in this chapter to describe the abusive behaviours that occurred within the relationship prior to the homicide. The term “intimate partner relationship” has been adopted to describe a current or former intimate relationship between the homicide victim and offender.

As noted earlier in this report, domestic and family violence constitutes a range of physical and non-physical forms of behaviours intended to exercise power and control over a partner through the use of fear, coercion and intimidation (Council of Australian Governments, 2019). These abusive behaviours can occur not only within intimate partner relationships but within a broader family unit, including between children and parents. Aboriginal and Torres Strait Islander communities often prefer the terminology of “family violence” as it acknowledges relationship ties beyond those of intimate partners and is more encompassing of kinship relationships (Langton et al., 2020a; Olsen & Lovett, 2016). However, as the dataset that informs the report is limited to intimate partner relationships, when we use the terminology “domestic violence” we refer to violent and abusive behaviours in current or former intimate partner relationships prior to the homicide.

This results chapter will first present an overview of the national data on IPV homicides in this dataset, first disaggregated by jurisdiction and then by gender of homicide offenders and victims.

It will then explore the characteristics relating to the relationship and the homicide and demographic characteristics of IPV homicides in Australia. This is disaggregated by offender and victim gender. We present the incidents of male-perpetrated IPV homicide first (perpetrated against both female and male victims), as these account for the largest proportion of cases in this dataset. Following that, we present the characteristics of the female IPV homicide victims as the second-largest cohort in the dataset. We then present incidents of female-perpetrated IPV homicide, followed by characteristics pertaining to male IPV homicide victims (including those victims killed by a male intimate partner). Following this, we present the data on surviving children.

The final section of this chapter provides a detailed analysis of separation trends, family law proceedings, the presence of domestic violence orders, and domestic violence behaviours identified in the relationship. These findings are drawn from the focused IPV homicide dataset.

Overview

Intimate partner violence homicides in Australia, 2010–2018

Between 1 July 2010 and 30 June 2018, there were 311 intimate partner homicides across Australia where an identifiable history of domestic violence preceded the fatal episode (311 IPV homicides). These include homicides of intimate partners by both male and female homicide offenders. An identifiable history of domestic violence includes both anecdotal histories of domestic violence (unreported) and instances of domestic violence that were reported to the police or other services. Table 1 sets out the number of IPV homicides that occurred in each Australian jurisdiction between 1 July 2010 and 30 June 2018 and provides a comparative analysis of the IPV homicide rate in each jurisdiction based on its relative population size.

Table 1: IPV homicides in Australia by jurisdiction, 2010–2018 (n=311)

State/territory	IPV homicide incidents, 2010–2018	Number of IPV homicides per 100,000 of jurisdiction’s population over the age of 15a
New South Wales	100	1.5
Queensland	61	1.5
Victoria	53	1.0
Western Australia b	42	2.0
Northern Territory	25	12.9
South Australia	19	1.3
Tasmania	8	1.8
Australian Capital Territory	3	0.9
National (total)	311	1.5

^a Population estimates for each jurisdiction are based off the Australian Bureau of Statistics’ National, state and territory population from December 2020 (Australian Bureau of Statistics, 2021b).
^b The increase in IPV homicides for Western Australia, from the 2018 Data Report (the first edition), does not reflect an increase in IPV homicides in Western Australia; rather, it relates to the inclusion of Ombudsman Western Australia data in this second edition which provides accuracy of data.

IPV homicides and gender

Of the 311 IPV homicides in the reporting period, approximately three quarters involved a male offender killing a female intimate partner (n=240, 77.2%). Sixty-five cases (20.9%) involved a female homicide offender killing a male intimate partner and six cases (1.9%) involved a man killing a male intimate partner. No cases involved a woman killing a female intimate partner.

One case involved a male offender killing his partner who was a transgender woman. The experiences of domestic violence and the access to support services for transgender and gender diverse people can be different from those of cisgender people (Campo & Tayton, 2015; Ussher et al., 2020). However, in order to maintain the privacy of the homicide victim and offender, this case has been aggregated with the male homicide offenders who killed female victims in the data presented in this report. Figure 1 sets out the gender breakdown of IPV homicides from July 2010 to June 2018.

Figure 1: Gender breakdown of IPV homicide offenders, July 2010–June 2018 (n=311)

In the next section of this chapter, we present the relationship, homicide and demographic details for the offenders and victims, disaggregated by male homicide offenders, female homicide victims, female homicide offenders and male homicide victims.

Male IPV homicide offenders, 2010–2018

The following section presents the details of the 246 male IPV homicide offenders, including their socio-demographic information, characteristics of their relationship with the homicide victim and characteristics of the homicide event. The following data disaggregates between the 240 men who killed a female intimate partner and the six men who killed a male intimate partner.

Relationship details

History of domestic violence victimisation/perpetration preceding the homicide

Of the 240 IPV homicides where a man killed a female intimate partner, 227 male offenders had been the primary domestic violence abuser against the female partner they killed (94.6%). In far fewer cases, both parties mutually used and experienced domestic violence behaviours against each other (n=6, 2.5%). In six cases, the male homicide offender killed a female partner who had been the primary user of domestic violence against him (see Table 2).

Table 2: Domestic violence (DV) perpetration and victimisation: Male homicide offender, female homicide victim (n=240)

DV perpetration/victimisation status	IPV homicide incidents	% of incidents where a male offender killed a female intimate partner
Primary DV abuser	227	94.6
Both a DV abuser and a DV victim	6	2.5
Primary DV victim	6	2.5
Unknown	1	0.4
Total	240	100

Of the six men who killed a male intimate partner, three offenders were the primary DV abuser against the partner they killed. Two male homicide offenders were the primary victim of abuse from the male partner they killed and in one case the offender and victim mutually used domestic violence against each other.

Relationship type

Of the 240 male offenders who killed a female intimate partner, 154 killed a partner he was in an ongoing relationship with (64.2%) and 86 killed a former intimate partner (35.8%; see Table 3).

Almost a third of male offenders were in a current de facto relationship with the victim (n=77, 32.1%); 51 men killed their current wife (21.3%); and 26 men killed a current girlfriend (10.8%). Thirty-three men killed their former de facto wife (13.8%); 30 killed a wife they had separated from (12.5%); and 23 men killed their former girlfriend (9.6%).

Of the men who killed male partners, three killed a current boyfriend, two killed a de facto husband and one killed a former de facto husband.

Table 3: Relationship type: Male homicide offender, female homicide victim (n=240)

Relationship type	IPV homicide incidents	% of incidents where a male offender killed a female intimate partner
De facto wife	77	32.1
Wife	51	21.3
Girlfriend	26	10.8
Former de facto wife	33	13.8
Former wife	30	12.5
Former girlfriend	23	9.6
Total	240	100

Length of relationship

Of the male-perpetrated IPV homicides against a female intimate partner, relationships ranged from less than 1 year to 45 years in duration. Figure 2 presents the distribution of IPV homicide by relationship length, from less than a year to 20 or more years. Notably, in 25 cases the offender had been in a relationship with the homicide victim for less than a year (10.4%) and in 28 cases the homicide occurred after a year-long relationship (11.7%). There were 33 cases in which the IPV homicide occurred after more than 20 years in a relationship (13.8%). This data finding demonstrates that IPV homicides can occur at any stage during a relationship, with homicides occurring during or after short relationships as well as after many years of protracted violence by abusers.

Figure 2: Relationship duration: Male homicide offender, female homicide victim (n=240)

Relationship length (years)	Percentage
< 1	10.4%
1	11.7%
2	10.4%
3	5.4%
4	5.4%
5	5.0%
6	4.2%
7	2.5%
8	2.9%
9	2.1%
10	6.7%
11	1.3%
12	1.3%
13	0.4%
14	1.7%
15	0.8%
16	2.1%
17	1.7%
18	2.5%
19	2.2%
20+	13.8%
Unknown	5.8%

Note: the data in the table is an estimate of what is visually represented in the figure.

For the six men who killed a male intimate partner each relationship varied in length: 1 year, 2 years, 5 years, 7 years, 16 years and 20+ years.

Homicide characteristics

Location

Of the 240 IPV homicides by a male offender against a female partner, the highest number of fatal episodes occurred within the residence shared by the homicide offender and victim (n=97, 40.4%). In 54 cases, the homicide occurred in the homicide victim’s residence (22.5%) and in 45 cases it occurred in a public location (18.8%). Twenty-two homicides occurred in the homicide offender’s residence (9.2%) and 20 occurred in a residence that neither the offender nor victim resided in, for example, a friend or family member’s residence (8.3%). Two male homicide offenders killed their female victims in their workplace (0.8%; see Table 4).

Of the homicides perpetrated by a male offender against a male partner, four occurred in a residence shared by both the homicide offender and victim and two occurred in the homicide victim’s residence.

Table 4: Homicide location: Male homicide offender, female homicide victim (n=240)

Homicide location	IPV homicide incidents	% of incidents where a male homicide offender killed a female intimate partner
Shared residence	97	40.4
Homicide victim residence	54	22.5
Public/open place	45	18.8
Homicide offender residence	22	9.2
Other residence	20	8.3
Workplace	2	0.8
Total	240	100

Mechanism of fatal assault

In almost one third of the cases where a male homicide offender killed a female intimate partner, the offender fatally assaulted the victim with a sharp weapon, such as a knife (n=77, 32.1%). In 39 cases the offender killed the victim by assaulting them without a weapon (16.3%) and in 31 cases the cause of death was suffocation or strangulation (12.9%). In 29 cases the offender shot the victim (12.1%) and in another 29 cases the death was due to multiple assaultive behaviours from the homicide offender, for example where the cause of death was due to multiple types of assault (e.g. suffocation or strangulation and assault with a blunt weapon; 12.1%).

Other manners of death are set out in Table 5 and include assault with a blunt weapon (such as a bat or hammer; n=21, 8.8%), vehicle-related assaults such as intentional vehicle-related collisions (n=3, 1.3%), fire or heat-related assault (n=3, 1.3%), drowning (n=2, 0.8%) and poisoning (n=1, 0.4%). In five cases the cause of death remains unknown because the victim’s body has never been found or the cause of death was unable to be ascertained (2.1%).

Of the male homicide offenders who killed their male intimate partners, three assaulted the victim with a sharp weapon, two assaulted the victim with a blunt weapon and one suffocated the victim.

Table 5: Mechanism of fatal assault: Male homicide offender, female homicide victim (n=240)

Mechanism of fatal assault	IPV homicide incidents	% of incidents where a male homicide offender killed a female intimate partner
Assault – sharp weapon	77	32.1
Assault – no weapon	39	16.3
Suffocation/strangulation	31	12.9
Multiple assaultive behaviours	29	12.1
Shooting	29	12.1
Assault – blunt weapon	21	8.8
Vehicle-related (e.g. collision)	3	1.3
Fire/heat-related	3	1.3
Drowning	2	0.8
Poisoning/noxious substance	1	0.4
Unknown	5	2.1
Total	240	100

Male homicide offender suicide post-homicide

Of the 240 cases where a male homicide offender killed a female intimate partner, 44 suicided after the homicide (18.3%). Almost three quarters of these offenders suicided within 24 hours of the homicide (n=32, 72.7% of offenders who suicided).

None of the men who killed a male intimate partner died by suicide after the homicide.

Coronial and criminal court outcomes

Of the 240 cases where a male homicide offender killed a female intimate partner, 48 cases were finalised by way of coronial finding because the offender was deceased. This includes the 44 cases where the offender died by suicide following the homicide and an additional four cases where the offender died due to other causes prior to the finalisation of criminal proceedings. The remaining 192 cases were subjected to a criminal investigation and proceedings.

Of the 192 IPV homicides that were finalised by way of criminal proceedings, the most common outcome for the male homicide offender was a murder conviction (n=121, 63.0%). Just over a quarter of offenders were convicted of manslaughter (n=50, 26.0%), eight were found not guilty by reason of mental illness (NGMI; 4.2%), and seven were convicted of less serious charges than murder or manslaughter (3.6%).^[10] In three cases charges were not laid or the charges were withdrawn (1.6%) and in a further three cases the offender was acquitted (1.6%; see Figure 3).^[11]

Figure 3: Criminal outcomes: Male homicide offender, female homicide victim (n=192)

Criminal outcome	Percentage
Guilty plea murder	35.9%
Guilty verdict murder	27.1%
Guilty plea manslaughter	16.1%
Guilty verdict manslaughter	9.9%
NGMI	4.2%
Guilty plea lesser charges	3.1%
Acquitted	1.6%
No charges laid / withdrawn	1.6%
Guilty verdict lesser charges	0.5%

Of the six cases where a male homicide offender killed a male intimate partner, four were acquitted and two submitted a guilty plea for murder.

Demographic details

Age

The age of the 240 male homicide offenders who killed a female intimate partner ranged from 18 to 82. The average age was 42 and there was a standard deviation of 12.84.

The ages of the six male homicide offenders who killed a male intimate partner ranged from 33 to 58. The average age of these offenders was 43 and the standard deviation was 9.48.

Aboriginal and/or Torres Strait Islander status

There were 53 male homicide offenders who identified as Aboriginal (22.1%), one offender who was Torres Strait Islander (0.4%) and two offenders who were both Aboriginal and Torres Strait Islander men (0.8%). All of these homicide offenders killed a female intimate partner. These findings demonstrate an overrepresentation of Aboriginal and Torres Strait Islander peoples per percentage of the population (approximately 3.3%; Australian Bureau of Statistics, 2019c).

While acknowledging these high rates, it is important to recognise that domestic and family violence is not a part of Aboriginal and Torres Strait Islander cultures (Adams et al., 2017; Cripps & Adams, 2014). Prominent literature explores the complex range of interrelated factors associated with domestic and family violence in Aboriginal and Torres Strait Islander communities. Much of this literature calls for greater recognition of the impacts of colonisation, dispossession, child removal, institutional and structural violence, overrepresentation in the criminal justice system, and limited access to services (Adams et al., 2017; Cripps & Adams, 2014; Langton et al., 2020a). This does not serve to absolve individuals of accountability for engaging in violent behaviour; rather, it provides an important framework to better understand, respond to and prevent family violence (Adams et al., 2017). Across Australia there are a range of initiatives led by Aboriginal and Torres Strait Islander communities that address some of these factors and seek to respond to and prevent domestic and family violence through holistic programs that focus on healing and wellbeing and promote reconnection to culture (Adams et al., 2017; Carlson et al., 2021). Academics and the community continue to advocate for agencies to work in partnership with Aboriginal and Torres Strait Islander communities to improve response measures, especially in the criminal justice system (Langton et al., 2020a).

Country of birth

Just under two thirds of the 240 male homicide offenders who killed a female intimate partner were born in Australia (n=157, 65.4%). Seventy-four (30.8%) were known to be born outside of Australia and in nine cases the country of birth is unknown. According to the Australian Bureau of Statistics (2021a), 29.8 per cent of Australia’s population was born outside of Australia, so this data finding would appear to accord with the general population statistics. Thirty-four different countries of birth were identified for the male homicide offenders born outside of Australia who killed a female intimate partner.

Of the six male homicide offenders who killed a male intimate partner, three were born in Australia and the remaining three were born outside of Australia, each in different countries.

Disability

Of the 240 male homicide offenders who killed a female intimate partner, 23 (9.3%) were identified as a person with disability, meaning they had a formal disability diagnosis and/or were in receipt of a disability pension. This figure excludes offenders who sustained injuries resulting in disability in the course of the homicide event. According to the Australian Bureau of Statistics, 17.7 per cent of the Australian population are living with disability, so this data finding would appear to be below the population statistics (Australian Bureau of Statistics, 2019b). Of the 23 male offenders with disability, 11 had physical disability, three had intellectual disability and 12 had psychiatric disability.^[12]

Of the six male homicide offenders who killed a male intimate partner, one was identified as a person with disability, however the type of disability is unknown.

This data finding should be interpreted with caution and may reflect systemic issues with the identification and definition of disability in service data. This is recognised as a limitation in the “Discussion” section of this report.

Employment status

Workplaces can be an important site for the primary prevention of and early intervention for domestic and family violence (Domestic Violence Death Review Team, 2020; Special Taskforce on Domestic and Family Violence in Queensland, 2015). Some workplaces actively promote this, through the delivery of primary prevention awareness-raising programs and active bystander programs and by improving the nature of support offered to victims and survivors of domestic and family violence. Formal employment may provide a victim and survivor an opportunity to access support through their peers or other services, while away from their abusive partners.^[13] Given the important role that workplaces play in the primary prevention of and early intervention for domestic and family violence, this dataset tracks the employment status of homicide offenders and victims at the time of the offence.

Of the 240 male homicide offenders who killed a female intimate partner, over a third of offenders were engaged in paid employment (n=90, 37.5%) and 105 (43.8%) were not engaged in paid employment at the time of the homicide.^[14] Twelve male homicide offenders were in receipt of a disability pension and 12 were retirees or pensioners (5.0% respectively).^[15] Two male homicide offenders were students (0.8%). In 19 cases, the employment status of the offender could not be ascertained. Figure 4 graphs the employment status of male homicide offenders who killed a female intimate partner.

Figure 4: Employment status: Male homicide offender, female homicide victim (n=240)

Employment status	Percentage
Not in formal employment	43.8%
Employed	37.5%
Disability pension	5.0%
Retired / Pensioner	5.0%
Student	0.8%
Unknown	7.9%

Of the six male homicide offenders who killed a male intimate partner, one was engaged in paid employment, four were not engaged in paid employment and one was on a disability pension.

Problematic alcohol and drug use

The information contained in this section is derived from toxicology analysis soon after the homicide, witness statements, and offender self-reports concerning their alcohol and other drug use in the period leading up to and at the time of the homicide.^[16] The information presented represents a pattern of behaviour and identifies possible opportunities for intervention. These findings do not purport to identify problematic substance use as a causative factor for the homicide.

Over 60 per cent of male homicide offenders who killed a female intimate partner engaged in problematic drug and/or alcohol use in the lead-up to or at the time of the homicide (n=145, 60.4%). A quarter of these male homicide offenders were engaging in both problematic drug and alcohol use in the lead-up to or at the time of the homicide (n=60, 25.0%).

Just over half of the male homicide offenders who killed a female intimate partner engaged in problematic alcohol use at the time of the homicide (n=125, 52.1%). Of these, 65 used alcohol only and no other drug (27.1%).

One third of male homicide offenders who killed a female intimate partner engaged in problematic drug use in the lead-up to or at the time of the homicide (n=80, 33.3%). Of these, 20 (8.3%) engaged in problematic drug use only (and not problematic alcohol use).

Figure 5 presents the breakdown of alcohol and substance use by male homicide offenders who killed a female intimate partner.

Figure 5: Problematic substance use: Male homicide offender, female homicide victim (n=240)

Of the male homicide offenders who killed a male intimate partner, two engaged in problematic alcohol and drug use and two engaged in problematic drug use only (i.e. not alcohol) in the lead-up to or at the time of the homicide. Two had no reported history of problematic substance use in the lead-up to the homicide.

Female IPV homicide victims, 2010–2018

A total of 240 women were killed by a male intimate partner in the reporting period. This section presents the demographic characteristics of the female victims of IPV homicides from July 2010 to June 2018.

Age

The age of 240 female IPV homicide victims ranged from 16 to 78. The average age was 38 with a standard deviation of 12.33.

Aboriginal and/or Torres Strait Islander status

Of the 240 women killed by a male intimate partner, 55 identified as Aboriginal (22.9%) and four were Aboriginal and Torres Strait Islander women (1.7%). This data finding demonstrates an overrepresentation of Aboriginal and Torres Strait Islander women as victims of IPV homicide (Australian Bureau of Statistics, 2019c).

It is important to recognise that Aboriginal and Torres Strait Islander women experience family violence from both Aboriginal and Torres Strait Islander men and non-Aboriginal and Torres Strait Islander men (Our Watch, 2018). This is borne out in the data, where one in 10 Aboriginal and/or Torres Strait Islander women were killed by a male intimate partner who did not identify as Aboriginal and/or Torres Strait Islander (n=8, 13.6%). Aboriginal and Torres Strait Islander women who experience intimate partner violence also experience complex barriers to accessing and receiving help. For example, the literature highlights that poor or discriminatory practices from frontline services, such as the police or child protection services, can result in poor outcomes for victims and survivors and a mistrust of these services (SNAICC National Voice for Our Children, National Family Violence Prevention Legal Services Forum, & National Aboriginal and Torres Strait Islander Legal Services, 2017). This experience can be compounded for women living in regional or remote locations, where there may be fewer service options, less access to information, and concerns pertaining to privacy and confidentiality in small communities (Langton et al., 2020b).

Country of birth

The majority of the 240 female IPV homicide victims were born in Australia (n=169, 70.4%). Sixty-eight were known to be born outside of Australia (28.3%), which accords with population statistics for people born outside of Australia (29.8% of the Australian population; Australian Bureau of Statistics, 2021a). For three cases the country of birth is unknown. The women born outside of Australia were born in 39 different countries.

Employment status

As noted previously, workplaces have been identified as an important site of intervention and prevention for domestic violence victims and abusers (Domestic Violence Death Review Team, 2020; Special Taskforce on Domestic and Family Violence in Queensland, 2015; Workplace Gender Equality Agency, n.d). Over a third of the 240 women who were killed by a male intimate partner were engaged in formal employment at the time of their death (n=92, 38.3%). Another two fifths of female IPV homicide victims were not engaged in formal, paid employment but may have undertaken unpaid home duties, care responsibilities or volunteering (n=104, 43.3%). Eleven women were receiving a disability pension and 11 were retired or pensioners (4.6% respectively). Six women were students at the time of their death (2.5%). In 16 cases, employment status could not be ascertained (see Table 6).

Table 6: Employment status: Female homicide victims (n=240)

Employment status	IPV homicide incidents	% of female IPV homicide victims
Not in formal employment	104	43.3
Employed	92	38.3
Disability pension	11	4.6
Retired/pensioner	11	4.6
Student	6	2.5
Unknown	16	6.7
Total	240	100

Disability

Sixteen women had a diagnosed disability or received a disability pension (6.7%). This data finding appears to be less than the general population statistics of people living with disability, which is approximately 17.7 per cent (Australian Bureau of Statistics, 2019b). Of these 16 women, 12 had physical disability, four had psychiatric disability, two had intellectual disability and two had sensory disability.

As noted previously, this data finding should be interpreted with caution. It is acknowledged that women with disability experience intimate partner violence at higher rates than women without disability (Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability, 2020) and, therefore, this finding may reflect limitations around the identification and definition of disability in service data (further explored in the “Discussion” section of this report).

Pregnancy

Five of the 240 female IPV homicide victims were pregnant at the time that they were killed (2.1%).

Female IPV homicide offenders, 2010–2018

The following section presents the details of the 65 female IPV homicide offenders, including their socio-demographic information, characteristics of their relationship with the homicide victim and characteristics of the homicide event. All 65 women killed a male intimate partner.

Relationship details

History of domestic violence victimisation/perpetration preceding the homicide

Of the 65 cases where a female homicide offender killed a male intimate partner, 70 per cent killed a man who was the primary domestic violence abuser against them in the relationship (n=46, 70.8%). This means that the vast majority of female IPV homicide offenders killed their abusive male intimate partners (see Table 7).

In just over a fifth of cases the female IPV homicide offender was both a domestic violence victim and abuser in the relationship – that is, no primary aggressor was identifiable because there was evidence of both parties using abusive behaviours towards each other (n=14, 21.5%). In five cases the female IPV homicide offender was identified as the primary domestic violence abuser against the male victim (7.7%).

Identification of the primary aggressor can be complicated in domestic violence relationships, especially in circumstances where physical violence is used as a measure for whether domestic violence is occurring or for the severity of this violence. A growing body of literature reflects on the misidentification of the primary aggressor in a domestic violence relationship in cases where a woman uses retaliatory or defensive violence in response to a pattern of physical or non-physical abuse perpetrated against her by her partner (Boxall et al., 2020; Nancarrow et al., 2020; Ulbrick & Jago, 2018).

Table 7: Domestic violence (DV) perpetration and victimisation: Female homicide offender, male homicide victim (n=65)

DV perpetration/victimisation status (female homicide offender)	IPV homicide incidents	% of incidents where a female homicide offender killed a male partner
Primary DV victim	46	70.8
Both a DV abuser and a DV victim	14	21.5
Primary DV abuser	5	7.7
Total	65	100

Relationship type

Of the 65 female IPV homicide offenders, more than three quarters killed a male intimate partner they were still in a relationship with (n=50, 76.9%) and 15 killed a former male intimate partner (23.1%).

Over half of the women were in a current de facto relationship with the male intimate partner they killed (n=37, 56.9%); seven women killed their current husband (10.8%); and six killed a current boyfriend (9.2%). Eleven women killed their former de facto male partner (16.9%); two women killed their former husband (3.1%); and two killed their former boyfriend (3.1%). Table 8 presents the breakdown of relationship type for current and former intimate partners.

Table 8: Relationship type: Female homicide offender, male homicide victim (n=65)

Relationship type	IPV homicide incidents	% of incidents with a female IPV homicide offender
De facto husband	37	56.9
Husband	7	10.8
Boyfriend	6	9.2
Former de facto husband	11	16.9
Former husband	2	3.1
Former boyfriend	2	3.1
Total	65	100

Length of relationship

Of the 65 cases where a female homicide offender killed a male intimate partner, the relationship length ranged from less than one year to more than 35 years in duration. Figure 6 demonstrates the distribution of female-perpetrated IPV homicide by relationship length, from less than 1 year to over 20 years. Notably, almost a quarter of women who killed a male intimate partner had been in a relationship with that partner for less than 1 year (n=15, 23.1%).

Figure 6: Relationship duration: Female homicide offender, male homicide victim (n=65)

Relationship length (years)	Percentage
< 1	23.1%
1	7.6%
2	12.3%
3	4.6%
4	7.7%
5	7.7%
6	7.7%
7	4.6%
8	4.6%
9	1.5%
10	3.0%
11	0%
12	3.1%
13	1.5%
14	3.1%
15	1.5%
16	0%
17	0%
18	0%
19	0%
20+	4.6%
Unknown	1.5%

Note: the data in the table is an estimate of what is visually represented in the figure.

Homicide characteristics

Location

Of the 65 homicides where a female homicide offender killed a male intimate partner, more than two fifths occurred in the residence shared by the homicide offender and victim (n=28, 43.1%). In 10 cases, the homicide occurred in a public place (15.4%). Other locations where the homicide occurred included the homicide offender’s residence, the homicide victim’s residence and another residence (such as the home of a family member or friend), with each occurring in nine cases respectively (13.8%; see Table 9).

Table 9: Homicide location: Female homicide offender, male homicide victim (n=65)

Homicide location	IPV homicide incidents	% of incidents where a female homicide offender killed a male intimate partner
Shared residence	28	43.1
Public/open place	10	15.4
Homicide victim residence	9	13.8
Homicide offender residence	9	13.8
Other residence	9	13.8
Total	65	100

Mechanism of fatal assault

More than three quarters of the female IPV homicide offenders killed their homicide victim by assaulting them with a sharp weapon, such as a knife (n=51, 78.5%). Three female homicide offenders killed the homicide victim by assaulting them with a blunt weapon (4.6%) and another three used a firearm (4.6%). In two cases the manner of death was determined to be the result of multiple assaultive behaviours (3.1%) and in a further two cases the victim died as a result of a vehicular-related assault. In one case the cause of death was assault without a weapon and in another it was fire or heat-related (1.5%). In one case the cause of death was unable to be ascertained (see Table 10).

Table 10: Mechanism of fatal assault: Female homicide offender, male homicide victim (n=65)

Mechanism of fatal assault	IPV homicide incidents	% of incidents with a female IPV homicide offender
Assault – sharp weapon	51	78.5
Assault – blunt weapon	3	4.6
Shooting	3	4.6
Multiple assaultive behaviours	2	3.1
Vehicle-related	2	3.1
Assault – no weapon	1	1.5
Fire/heat-related	1	1.5
Other	1	1.5
Unknown	1	1.5
Total	65	100

Female homicide offender suicide post-homicide

Of the 65 cases where a female homicide offender killed a male partner, one woman suicided after the homicide. In this case, she died within 24 hours of the homicide.

Coronial and criminal court outcomes

A criminal investigation occurred in 64 of the 65 cases where a female homicide offender killed a male partner (see Figure 7).^[17] Of these, the majority resulted in the female homicide offender receiving a manslaughter conviction (n=40, 62.5%). The next most common outcomes were a murder conviction (n=7, 10.9%) and an acquittal (n=7, 10.9%). In four cases no charges were laid or the charges were withdrawn (6.3%) and in a further four cases the female homicide offender was found not guilty by reason of mental illness (NGMI; 6.3%). In two cases the female homicide offender was convicted of lesser charges (3.1%).

Figure 7: Criminal outcomes: Female homicide offender, male homicide victim (n=64)

Criminal outcome	Percentage
Guilty plea manslaughter	50.0%
Guilty verdict manslaughter	12.5%
Acquitted	10.9%
Guilty verdict murder	7.8%
NGMI	6.3%
No charges laid / withdrawn	6.3%
Guilty plea lesser charges	3.1%
Guilty plea murder	3.1%

Demographic details

Age

The ages of the 65 female IPV homicide offenders ranged from 18 to 75 years. The average age was 35 with a standard deviation of 12.08.

Aboriginal and/or Torres Strait Islander identity

Twenty-seven of the 65 female IPV homicide offenders identified as Aboriginal (41.5%). Of these women, almost two thirds had been the primary victim of violence in the relationship, that is, they experienced domestic violence perpetrated against them by the homicide victim (n=17, 63.0%). Just over a quarter of the women were identified as mutually experiencing and using violence with the partner they killed (n=7, 25.9%) and three had been identified as the primary domestic violence aggressor against the partner they killed (11.1%).

Having regard to comments made earlier in this report, caution must be adopted when interpreting this data as Aboriginal and Torres Strait Islander women’s experiences and use of domestic and family violence are shaped by complex factors. For example, research demonstrates that societal and systemic racism, coupled with stereotypes of the expected behaviour of a domestic violence victim, contribute to the misidentification of Aboriginal and Torres Strait Islander women as aggressors rather than victims and survivors in domestic violence cases (Douglas & Fitzgerald, 2018; Nancarrow et al., 2020).

Country of birth

Of the 65 female homicide offenders, 80 per cent were born in Australia (n=52). Eleven were born outside of Australia (16.9%). This is below the national population statistics of people born outside of Australia (29.8%; Australian Bureau of Statistics, 2021a). The country of birth is unknown for two offenders. There were nine different countries of birth identified, other than Australia.

Employment status

A fifth of the female IPV homicide offenders were employed at the time of the homicide (n=14, 21.5%). As mentioned previously, workplaces have been identified as important sites of domestic and family violence intervention and prevention and may provide people who experience violence with an opportunity to seek support away from their abuser (Domestic Violence Death Review Team, 2020; Special Taskforce on Domestic and Family Violence in Queensland, 2015).

Over half of the female IPV homicide offenders were not engaged in formal, paid employment at the time of the offence, although some may have engaged in unpaid work such as home-based duties or caring for family members (n=34, 52.3%). Four received a disability pension (6.2%) and two were retired or pensioners (3.1%). For 11 female IPV homicide offenders, employment status could not be ascertained (see Figure 8).

Figure 8: Employment status: Female homicide offender, male homicide victim (n=65)

Employment status	Percentage
Not in formal employment	52.3%
Employed	21.5%
Disability pension	6.2%
Retired / Pensioner	3.1%
Student	16.9%

Disability

Of the 65 female homicide offenders who killed a male intimate partner, six (9.2%) were identified as a person with disability, meaning they had a formal disability diagnosis and/or were in receipt of a disability pension. This figure excludes offenders who sustained injuries resulting in disability in the course of the homicide event. This figure is less than the national population statistics of 17.7 per cent of Australians living with disability (Australian Bureau of Statistics, 2019b) but should be interpreted with caution given the limitations around the identification and definition of disability in service data (see “Discussion”).

Of these, three had physical disability, three had psychiatric disability and one had intellectual disability.

Problematic alcohol and drug use

The information set out in this section is derived from toxicology analysis soon after the homicide, witness statements, and offender self-reports concerning their alcohol and other drug use in the period leading up to and at the time of the homicide. As noted previously, the information presented represents a pattern of behaviour and does not necessarily mean that the problematic substance use was a causative factor for the homicide.

The majority of female homicide offenders engaged in problematic drug and/or alcohol use in the lead-up to or at the time of the homicide (n=38, 58.5%).

In a fifth of the 65 cases where a female homicide offender killed a male partner, the woman engaged in both problematic alcohol and drug use in the lead-up to or at the time of the homicide (n=13, 20.0%).

Almost half of the female homicide offenders engaged in problematic alcohol use at the time of the homicide (n=32, 49.2%). Of these, 19 consumed alcohol only (see Figure 9).

Almost 30 per cent of female IPV homicide offenders engaged in problematic drug use in the lead-up to or at the time of the homicide (n=19, 29.2%). Of these, six engaged in problematic drug use only (see Figure 9).

Figure 9: Problematic substance use: Female homicide offender, male homicide victims (n=65)

Pregnancy

Three of the 65 female IPV homicide offenders were pregnant at the time of the homicide. All three women were the primary domestic violence victim who killed their abusive male partner.

Male IPV homicide victims, 2010–2018

A total of 71 men were killed by an intimate partner between 1 July 2010 and 30 June 2018. Of these, 65 were killed by a female intimate partner and six were killed by a male intimate partner. This section presents the demographic characteristics of the male IPV homicide victims.

Demographic details

Age

The average age of male IPV homicide victims who were killed by a female homicide offender was 38 years with a standard deviation of 11.24. The youngest male homicide victim was 18 years old and the oldest was 76 years old.

Of the male IPV homicide victims killed by a male offender, ages ranged from 34 to 62 with an average of 46 years and a standard deviation of 9.26.

Aboriginal and/or Torres Strait Islander status

Of the 65 male IPV homicide victims killed by a female partner, 25 identified as Aboriginal (38.5%) and one identified as Torres Strait Islander (1.5%). None of the men who were killed by a male intimate partner identified as Aboriginal or Torres Strait Islander.

As highlighted previously in this report, there are complex interrelated factors associated with family violence in Aboriginal and Torres Strait Islander communities. The consistent overrepresentation of Aboriginal and Torres Strait Islander people in this dataset demonstrates a greater need for services to work with these communities to improve family violence response and prevention.

Country of birth

Approximately 80 per cent of male IPV homicide victims who were killed by a female intimate partner were born in Australia (n=53, 81.5%). The remaining 12 were born outside of Australia (18.5%) in one of 10 different countries. This is lower than the national population statistics for people born outside of Australia (29.8%; Australian Bureau of Statistics, 2021a).

Of the six men who were killed by a male intimate partner, four were born in Australia and two were born outside of Australia.

Employment status

Of the 65 men who were killed by a female intimate partner, almost half were not engaged in paid employment at the time of their death (n=30, 46.2%).^[18] Twenty -five were employed (38.5%), two received a disability pension (3.1%) and one was retired or a pensioner (1.5%; see Table 11).

Table 11: Employment status: Male homicide victim, female homicide offender (n=65)

Employment status	IPV homicide incidents	% of male IPV homicide victims
Not in formal employment	30	46.2
Employed	25	38.5
Disability pension	2	3.1
Retired/pensioner	1	1.5
Unknown	7	10.8
Total	65	100

Of the six men who were killed by a male intimate partner, three were employed, two were not engaged in paid employment and one received a disability pension.

Disability

Five of the 65 male IPV homicide victims killed by a female homicide offender had a diagnosed disability or were recipients of a disability pension (8.5%). This is below the national statistic of 17.7 per cent of Australians living with disability (Australian Bureau of Statistics, 2019b) but should be interpreted with caution given the limitations around the identification and definition of disability in service data (see “Discussion”). Of the five men with disability, three had physical disability, two had psychiatric disability and one had intellectual disability.

One of the six men who were killed by a male intimate partner had physical disability.

IPV homicide and children

The dataset captures information on the number of children (biological, adopted or fostered) the homicide offenders and victims had together. The dataset also captures whether there were any stepchildren (from either homicide offender or victim) in the household; whether any biological, adopted, fostered or stepchildren witnessed domestic violence between the homicide offender and victim; and whether any children were killed in the homicide.

The homicide offenders and victims in this dataset were joint parents of at least 172 children under the age of 18 at the time of the homicide.^[19] In at least 133 cases children were exposed to violence between the homicide offender and victim (42.8% of all IPV homicides). Tragically, eight children were killed in the fatal episode.

Domestic violence characteristics preceding IPV homicides

Domestic violence death review teams are uniquely positioned to conduct in-depth analysis and reviews so as to identify discrete characteristics present within a relationship prior to an IPV homicide. This chapter examines data around characteristics relating to separation or intention to separate, including the initiation of family law proceedings; current or historical domestic violence orders; and the nature of domestic violence and abusive behaviours such as physical, sexual, emotional and psychological, social, verbal and financial abuse, and stalking that the primary abuser used prior to the homicide.

The data in this chapter is drawn from those jurisdictions with a formalised death review mechanism in place, namely New South Wales, the Northern Territory, Queensland, South Australia, Victoria and Western Australia. As previously mentioned, data from the Australia Capital Territory and Tasmania is excluded on the basis that, in the absence of an operational death review process, specialised information regarding histories of violence is not available.^[20] Similarly, Western Australia’s death review process commenced reviews of domestic violence homicides from July 2012 and consequently, data from Western Australia from 1 July 2010 to 30 June 2012 has not been included in this detailed analysis because of the limited information available about the relationship prior to the homicide. In addition, there are three cases from New South Wales that have been excluded as the source material was not available to be reviewed within the project timeframes. Accordingly, the numbers in this section differ slightly to those presented earlier in the report as this data is drawn from the smaller, focused subset of data.

The focused data findings in this section of the report are derived from 292 cases of IPV homicide that occurred across New South Wales, the Northern Territory, Queensland, South Australia and Victoria between July 2010 and June 2018, and Western Australia between July 2012 and June 2018. This focused subset includes 224 cases where a male homicide offender killed a female partner; 62 cases where a female homicide offender killed a male partner; and six cases where a male homicide offender killed a male partner (Table 12).

Table 12: Gender disaggregation of focused data subset (n=292)

Offender and victim gender	IPV homicide incidents
Male homicide offender, female victim	224
Female homicide offender, male victim	62
Male homicide offender, male victim	6
Total	292

Separation as a characteristic of IPV homicide

Specialised death review processes often have access to detailed information regarding a couple’s separation or intention to separate prior to the homicide. This includes details of how soon before the homicide separation occurred, or when an intention to separate was expressed and by whom.

Male-perpetrated IPV homicide and separation

Actual separation or an intention to separate was a feature in over half of the 224 cases where a male IPV homicide offender killed a female intimate partner (n=130, 58.0%).

In over a third of cases, the relationship between the male offender and the female victim had ended prior to the homicide (n=77, 34.4%). In 44 of these cases (57.1%), separation occurred within three months of the homicide and in 33 cases (42.9%) the separation occurred more than three months prior to the homicide (see Table 13).

Table 13: Separation proximity to homicide: Male homicide offender, female homicide victim (n=77)

Separation proximity to homicide	IPV homicide incidents	% of separated relationships
Separated less than 3 months before homicide	44	57.1
Separated more than 3 months before homicide	33	42.9
Total	77	100

In 147 cases the relationship between the male IPV homicide offender and female intimate partner was ongoing at the time of the fatal episode (65.6%). In 53 of these cases, however, while the relationship was still ongoing, one or both parties had indicated an intention to leave the relationship (36.1%). In the majority of cases the female victim had indicated an intention to leave the relationship (n=45, 84.9%) and in only three cases the male offender expressed an intention to leave the relationship (5.7%). In five cases both the male homicide offender and the female victim had indicated an intention to leave the relationship (9.4%). Figure 10 presents the number of offenders and victims who expressed their intention to separate within three months of, or more than three months prior to, the homicide.

Figure 10: Intention to separate prior to the homicide: Male homicide offender, female homicide victim (n=53)

	Male homicide offender	Female homicide offender	Both parties
Intention to separate more than 3 months prior to homicide	2	11	5
Intention to separate less than 3 months prior to homicide	1	34	N/A

Actual or intended separation was a feature in two of the six cases where a male IPV homicide offender killed a male intimate partner. In one of these cases, actual separation had occurred within three months of the homicide. In the other case, the relationship was ongoing but the male victim had expressed his intention to separate, within three months of the homicide.

Female-perpetrated IPV homicide and separation

Actual separation or an intention to separate was a feature in less than half of the cases where a female IPV homicide offender killed a male intimate partner (n=28, 45.2%).

In less than a quarter of cases, the relationship between the female homicide offender and the male homicide victim had ended prior to the homicide (n=14, 22.6%). As Table 14 demonstrates, five of these separations had occurred less than three months prior to the homicide (35.7%) and nine occurred more than three months prior to the homicide (64.3%).

Table 14: Separation proximity to homicide: Female homicide offender, male homicide victim (n=14)

Separation proximity to homicide	IPV homicide incidents	% of separated relationships
Separated less than 3 months before homicide	5	35.7
Separated more than 3 months before homicide	9	64.3
Total	14	100

Of the 48 relationships that were ongoing (77.4%), one or both parties in the relationship indicated an intention to leave the relationship in 14 cases (29.2%). Of these, in eight cases the female IPV homicide offender had expressed an intention to separate (57.1%), in five cases the male homicide victim had expressed their intention to separate (35.7%), and in one case both parties had expressed their intention to separate prior to the homicide. Figure 11 presents the number of offenders and victims who expressed their intention to separate within three months of or more than three months prior to the homicide.

Figure 11: Intention to separate prior to the homicide: Female homicide offender, male homicide victim (n=14)

	Male homicide offender	Female homicide offender	Both parties
Intention to separate more than 3 months prior to homicide	N/A	3	1
Intention to separate less than 3 months prior to homicide	5	5	N/A

Family law proceedings

The Family Court deals with complex legal family disputes, including cases relating to divorce or separation, financial and property disputes, and arrangements relating to parenting and care of children. Given the relatively high rates of actual or intended separation in this dataset, which was identified as a feature in a total of 160 of the 292 cases in the focused dataset (54.8%), active family law proceedings were found in a relatively small number of cases (n=11, 3.8%). This finding reflects that not all cases of separation or divorce are escalated to the family law court and in many cases, disputes relating to separation, property division or childcare arrangements are settled informally.

Of the 11 cases where family law proceedings were active at the time of the homicide, eight involved a male IPV homicide offender killing a female intimate partner and three involved a female IPV homicide offender killing a male intimate partner. None of the cases in which a male IPV homicide offender killed a male intimate partner were identified as having ongoing family law proceedings.

The Network has identified room for improvement in data collection and information sharing between the Family Court and domestic violence death review teams to better identify cases where family court proceedings had occurred at the time of or prior to domestic violence deaths. This work stemmed from findings of the Queensland Domestic and Family Violence Death Review and Advisory Board, which identified that there is no process whereby the family court system is notified when a domestic violence context death occurs involving parties engaged with the court. The Board was of the perspective that this represented a missed opportunity for the family law system to derive learnings from and develop practice improvements in response to such deaths (Domestic and Family Violence Death Review and Advisory Board, 2019). The Network has commenced engagement with the Family Court to address this gap and improve data-sharing mechanisms between the parties. This work will contribute to more detailed findings in the future regarding family law proceedings and domestic violence deaths.

Domestic violence orders

A domestic violence order is a civil order the object of which is to protect victims – or persons at risk – of domestic and family violence from another person who they are, or have been, in an intimate or familial relationship with. The following section presents data on domestic violence orders between the IPV homicide offender and victim. The data is disaggregated between orders which were current (or enforceable) at the time of the homicide and historical domestic violence orders between the offender and victim, that is, where the order had expired prior to the homicide. This section also includes data on who was the protected person under the order and who was the respondent (the person against whom the domestic violence order is made).

A range of national and jurisdiction-based research explores the complexities of identifying the person most in need of protection in domestic violence orders (Boxall et al., 2020; Mansour, 2014; Nancarrow et al., 2020; Ulbrick & Jago, 2018). Within this literature, systems abuse has been highlighted as an area of concern. Systems abuse involves the manipulation of legal systems by an abuser to exert power or control over a victim (Douglas & Chapple, 2021; Nancarrow et al., 2020). This may be done through the misapplication of domestic violence orders against a victim by, for example, making false allegations of violence used by the victim, taking out retaliatory domestic violence orders, or minimising the abuser’s own use of violence. Systems abuse has the potential to undermine a victim’s confidence in the legal system and restrict their access to support. Although systems abuse is not specifically explored or captured in this dataset, it is a factor worth considering when interpreting the data presented in this report relating to the person most in need of protection and the respondent to domestic violence orders.

Domestic violence orders as a characteristic of male-perpetrated IPV homicides

Domestic violence orders were a feature in over 40 per cent of the 240 cases where a male IPV homicide offender killed a female intimate partner (n=96, 42.9%). This demonstrates that in over two fifths of male-perpetrated IPV homicides, there was a history of police or court intervention due to domestic and family violence prior to the homicides. In 128 cases where a male offender killed a female victim there was no evidence of a current or historical domestic violence order (57.1%).

In 49 cases there was a current domestic violence order between the male offender and female victim at the time of the homicide (21.9%; see Figure 12). In the vast majority of these cases the order named the female victim as the protected person (n=44, 89.8%). In three cases, the order protected the male IPV homicide offender from the female intimate partner (6.1%). In two cases there were cross-orders, naming both the homicide offender and the homicide victim as the protected person (4.1%).

Figure 12: Protected person in current domestic violence orders: Male homicide offender, female homicide victim (n=49)

Historical domestic violence orders between the male IPV homicide offender and female intimate partner were a feature in 67 cases (29.9%; see Figure 13). Of these, the vast majority of cases named the female intimate partner as the protected person and the male homicide offender as the respondent (n=53, 79.1%). In 11 cases the historical orders named both the male homicide offender and female victim, either in cross-orders or in multiple separate orders (16.4%). In two cases the historical domestic violence orders protected the male homicide offender from the female victim (3.0%).

Figure 13: Protected person in historic domestic violence orders: Male homicide offender, female homicide victim (n=67)

There were no enforceable domestic violence orders at the time of the homicide in any of the six cases where a male homicide offender killed a male intimate partner. In one case, there was a historical domestic violence order protecting the homicide offender from the homicide victim.

Domestic violence orders as a characteristic of female-perpetrated IPV homicides

Domestic violence orders were a feature in 41 of the 62 cases where a female IPV homicide offender killed a male intimate partner (n=41, 66.1%). This demonstrates that in two thirds of female-perpetrated IPV homicides, there was a history of domestic violence and police or court intervention prior to the homicides.

In 21 cases there was a current domestic violence order between the female homicide offender and male homicide victim (33.9%; see Figure 14). Twelve of these orders protected the female homicide offender from the male homicide victim (57.1%) and eight named the male homicide victim as the protected person, and the female homicide offender as the respondent (38.1%). There was one cross-order naming both the female homicide offender and the male homicide victim as the protected person (4.8%).

Figure 14: Protected person in current domestic violence orders: Female homicide offender, male homicide victim (n=21)

Historical domestic violence orders were a feature in 29 of the 62 cases where a female homicide offender killed a male intimate partner (46.8%; see Figure 15). In more than half of these cases, the historical domestic violence orders named the female homicide offender as needing protection from the male homicide victim (n=17, 58.6%). In just over one third of these cases there had been cross-orders or multiple historical orders that named both the female homicide offender and the male homicide victim as the protected person (n=11, 37.9%). In one case the historical domestic violence order(s) named the male homicide victim as the protected person (3.4%).

Figure 15: Protected person in historic domestic violence orders: Female homicide offender, male homicide victim (n=29)

Domestic violence behaviours

Domestic violence is made up of a range of different behaviours intended to exercise power and control over a partner through the use of fear, coercion and intimidation (Council of Australian Governments, 2019). These behaviours include physical violence, emotional and psychological abuse, sexual violence, verbal abuse, social abuse, economic or financial abuse, and stalking.

In the focused dataset, the vast majority of the 224 male IPV homicide offenders who killed a female victim were identified as the primary domestic violence abuser in the relationship (n=212, 94.6%). Of the 62 cases where a female IPV homicide offender killed a male partner, only a small proportion of women were identified as the primary abuser against the male partner they killed (n=5, 8.1%). In half of the six cases where a male IPV homicide offender killed a male partner, the male offender was identified as the primary abuser in the relationship (n=3, 50%).

Information presented in the following section focuses on the abusive behaviours used by male primary domestic violence abusers who killed a female victim (n=212) as the sample size of cases where a female primary abuser killed a male partner and where a male primary abuser killed a male partner are too small for quantitative analysis of this kind.

The data presented draws on both reported and anecdotal accounts of abuse in the relationships prior to the homicide. Data sources include police records, prosecution briefs and witness statements of friends and family, records from various support services (health services, domestic and family violence services, housing services, child protection etc.) and coronial files.

Emotional and psychological abuse

Emotional and psychological abuse includes a broad spectrum of behaviours employed by abusers in order to frighten, belittle, humiliate, unsettle and undermine a victim’s sense of self-worth. These can include:

• verbally denigrating the victim
• making threats regarding custody of children as a means to control the victim
• blaming the victim for all adverse events
• gaslighting or exploiting a victim’s mental illness
• deliberately creating dependence.

Of the 212 male primary domestic violence abusers who killed female domestic violence victims, 173 exhibited emotionally and psychologically abusive behaviours against the female partners they killed (81.6%).

Physical violence

Physical violence pertains to an assault on the body, both with a weapon and/or assault without the use of a weapon, such as shaking, slapping, pushing, spitting, punching, non-fatal strangulation, kicking or pulling hair.

Physical violence was used in 169 of the 212 cases where a male primary domestic violence abuser killed a female domestic violence victim (79.7%).

Social abuse

Social abuse includes a range of abusive behaviours designed to prevent a person from spending time with family and friends and participating in social activities. Social abuse may include:

• controlling the extent to which the victim can interact with friends, family or colleagues
• publically denigrating the victim
• being threatening or rude to the victim’s friends and family
• forcibly relocating the victim away from support networks with the intention to isolate the victim
• restricting the victim’s access to transport
• controlling what the victim wears or eats.

These behaviours isolate victims and break down support networks, making it more difficult for a victim to seek help. Social abuse may also limit opportunities to gain employment or disrupt existing employment opportunities, which further limits access to support and may result in financial dependence on the abuser.

Social abuse was identified in 134 of the 212 cases where a male primary domestic violence abuser killed a female domestic violence victim (63.2%).

Stalking

Stalking involves a range of tactics whereby an abuser intimidates or controls a victim or makes them feel fearful through intentionally and persistently pursuing them or by monitoring their activities. It may include an abuser following the victim, loitering near the victim’s home or work, and breaking into the victim’s house. Stalking also includes acts of technology-facilitated abuse such as persistent text messaging, maintaining surveillance over the victim’s phone or email, covertly recording the victim’s activities, and engaging with the victim on social media/dating sites under a false identity.

Of the 212 cases where a male primary domestic violence abuser killed a female victim, the male homicide offender stalked the woman he killed in 88 cases (41.5%). In 71 cases, stalking occurred during the relationship (33.5%) and in 44 cases stalking occurred after the relationship ended (20.8%).

Financial abuse

Financial abuse encompasses a spectrum of abusive behaviours related to a partner’s or family member’s access to economic resources (including limiting access to finances, access to work etc.). This behaviour is intended to diminish a victim’s ability to support themselves and forces them to depend on the abuser financially.

Over a quarter of male primary domestic violence abusers who killed a female victim used economically or financially abusive tactics against the woman they killed (n=58, 27.4%).

Sexual violence

Sexual violence pertains to unwanted or non-consensual sexual behaviours used by an abuser against a victim. It may include rape, coercion, the use of physical violence during sex, sexual assault with implements, being forced to watch or engage in pornography, enforced prostitution, or being made to have sex with other people (Council of Australian Governments, 2019). It may also include the use of technology, including filming sexual acts without consent or sharing explicit images or videos of the victim without their consent.

Sexual violence was identified as being used by 34 male primary domestic violence abusers who killed a female homicide victim (16.0%).

The 2016 Personal Safety Survey (PSS) found that 17 per cent of all Australian women had experienced at least one incident of sexual assault since the age of 15 (Australian Bureau of Statistics, 2017). Further analysis of the PSS found that 36.2 per cent of women who had experienced physical assault had also experienced sexual assault since the age of 15 (Cox, 2015, p. 126).

Approximately one in six women had not told anyone about their most recent experience of sexual assault (Cox, 2015). Research has found that reporting of sexual violence to police, services and/or friends and family may be hindered due to:

• stigma and shame around sexual assault and sexual violence
• the victim’s perception of risk or fear of the abuser
• low confidence in the justice system to respond to sexual violence
• assumptions that sexual violence is a private matter
• minimisation of sexual violence or not regarding it as a serious/criminal matter (Australia’s National Research Organisation for Women’s Safety, 2019; Lievore, 2003; Wall, 2012a, 2012b).

Due to the potential underreporting of sexual violence, it is likely that this is an undercount of the true incidence of sexual violence in these cases.

Discussion

The data presented in the “Results” section of this report provides a national picture of the characteristics present in IPV homicides in Australia. This discussion section will consider some of the strengths and limitations of the dataset and explore some of the key findings in this report.

As noted previously, due to the case inclusion criteria determined by the Network’s Homicide Consensus Statement, there are challenges in using this dataset to accurately reflect the rate of IPV homicides over the eight-year reporting period (see the “Methods” chapter). Further, it is also important to recognise that, due to the federalist structure of Australia, many legal and policy matters concerning domestic and family violence are the responsibility of the individual states and territories (Bugeja et al., 2013; Butler et al., 2017). As such, it is not appropriate for this report to make direct policy recommendations, as these are best undertaken by the jurisdictional domestic and family violence death review mechanisms.

Dataset strengths and limitations

The Network represents a unique collaboration between domestic and family violence death review mechanisms across Australia. Network members are located within coroners courts, ombudsman’s offices or government agencies, and have substantial specialist knowledge of domestic and family violence, the relevant policy frameworks, and the domestic and family violence service landscape in their jurisdiction. It is through this extensive expertise that the Network was able to establish a nationally consistent definition of domestic violence deaths and develop the protocols for national data collection on IPV homicide in the NMDS.^[21]

The strength and quality of the data within the NMDS is also due to the diversity of the primary source materials the Network members have access to. Due to the location of death review mechanisms within relevant state and territory government agencies, death review teams have access to a broad range of primary data relating to a domestic violence death, including coronial files, briefs of evidence, police reports, media reporting, sentencing remarks, and services or agency records. Many of these data sources are not always available to national research agencies exploring domestic violence deaths. This sets the Network’s NMDS apart from other studies exploring homicide in Australia.

The combined expertise of Network members and unique primary source data available to the Network offers a greater opportunity to explore the trends and behaviours throughout a victim’s and an offender’s life course, including in the lead-up to and after the IPV homicide.

However, all datasets that rely on service data have some limitations. These were briefly set out in the “Methods” section of this report and are expanded on in more detail below.

Limitations for identifying IPV homicides

As per the case inclusion criteria set out in the methods chapter, the data presented in this report focuses on IPV homicides, that is, homicides between intimate partners that were preceded by an identifiable history of domestic and family violence. Intimate partner homicides where there was no identifiable reported or anecdotal history of domestic and family violence do not form a part of this dataset.

The history of domestic and family violence in the cases included in the dataset may have been formally reported or anecdotal. A reported or unreported history of domestic and family violence was identified through the review of police documents, sentencing remarks, coronial files, medical records (where available), service records and witness statements from people who knew the couple prior to the homicide. However, given the high proportion of domestic and family violence that is not disclosed or reported, it is acknowledged that the figures in this report may present an undercount of the true incidence of domestic and family violence-context homicides in Australia (Australian Bureau of Statistics, 2017; Cox, 2015).

Limitations in source data about Aboriginal and Torres Strait Islander status

This report draws on service data, such as government records, health records, police reports and support services records, to identify Aboriginal and Torres Strait Islander peoples in this dataset. The consistency and quality of this administrative data as a means of identifying Aboriginal and Torres Strait Islander status can be compromised by structural issues – such as the services not explicitly asking the client or making assumptions about their identity – and by a client’s choice not to self-identify as Aboriginal and/or Torres Strait Islander. Accordingly, there is the potential for underreporting of IPV homicides involving Aboriginal and Torres Strait Islander peoples.

Services have a responsibility to provide a culturally safe and responsive environment for Aboriginal and Torres Strait Islander clients. As such, it is crucial that services are implementing consistent procedures to collect accurate service data. However, within the Australian context, data collection on Aboriginal and Torres Strait Islander status has evolved against the backdrop of colonial structures, values and assumptions that have historically been used for exclusionary purposes, and which have failed to recognise Aboriginal and Torres Strait Islander peoples’ right to self-determination (Griffiths et al., 2019). Within this context, services need to not only implement consistent procedures to collect data on Aboriginal and Torres Strait Islander status, but also to address the structural concerns and barriers that may reduce people’s confidence in disclosing their identity to the service. This includes addressing institutional racism, discrimination and stereotyping; ensuring cultural safety; providing clear guidance on information use and privacy; and considering the most appropriate way to collect this data (e.g. who should be asking, where, how and why; NSW Aboriginal Affairs, 2015).

Limitations for identifying people with disability

Similarly to the challenges around identification of Aboriginal and Torres Strait Islander peoples, this report’s reliance on service data (government records, health records, police reports and support services records) to identify people with disability is acknowledged to be a limitation. There are critical gaps and inconsistencies in data collection around disability, particularly in relation to administrative datasets that are collected as a by-product of service delivery. Many service providers still do not identify disability at all, and for those that do there are stark variations in the definitions of disability. The Victorian Royal Commission into Family Violence and the current Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability have both identified the need to improve collection and reporting of domestic violence data so that governments can better understand and respond to violence and abuse against people with disability (Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability, 2020). These issues have likely resulted in a significant underreporting of the experiences of IPV homicide for people with disability, who are known to experience IPV and abuse at much higher rates than people without disability (Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability, 2020).

Limitations for identifying IPV homicides in LGBTQ communities

This research draws on service data from police, judicial and coronial systems to identify cases where an intimate partner homicide has occurred. It is possible that cases where the relationship between the homicide offender and victim was not disclosed or was otherwise not evident to those external to the relationship are misclassified as other forms of homicide. For example, a couple may be identified as housemates rather than intimate partners. As a result, there is the potential for underreporting of IPV homicides. This may be especially relevant for LGBTQ relationships, whether cohabiting or not, where one or both parties may not have disclosed the relationship to family or friends and consequently the relationship is not acknowledged or recognised in system data.

Limitation on reporting of children

The data sources used in this report do not always accurately capture information on surviving children, especially in the case of stepchildren (of either the homicide victim or offender) or children who are living outside the home. Due to this limitation, the data presented in this report on surviving children may represent an undercount, based on the known number of children identified in the source material.

Limiting the scope of research to intimate partner homicides

This dataset focused on intimate partner homicides as one specific form of domestic and family violence-related death. Domestic and family violence homicides can also occur between other family members, such as between parents and children, between siblings, or between other extended family members or kinship ties. The Network has plans to extend its data reporting to encompass other types of domestic and family violence-related deaths in the future.

Key findings

Gendered nature of IPV homicide

Between 1 July 2010 and 30 June 2018, the majority of IPV homicides that followed a reported or anecdotal history of domestic violence were perpetrated by a male homicide offender against his female intimate partner (n=240, 77.2%). This accounted for over three quarters of the 311 IPV homicides in this dataset.

Of the male offenders who killed a female victim, the vast majority of these men had been the primary user of domestic violence abusive behaviours against the female partner they killed (n=227, 94.6% of male IPV homicide offenders who killed a female victim).

In far fewer cases, a female homicide offender killed a male partner (n=65, 20.9% of all IPV homicides). The majority of these cases involved a woman who was the primary domestic violence victim in the life of the relationship who killed a male partner who was violent towards her (n=46, 70.8%).

There were six cases that involved a male IPV offender killing a male intimate partner and no cases where a female homicide offender killed a female intimate partner.

Domestic violence orders prior to IPV homicides

Detailed data provided by the Australian jurisdictions with a formalised death review mechanism demonstrates that domestic violence orders were a feature in a significant proportion of IPV homicides. These orders (current or historical) were a feature in two thirds of cases where a female homicide offender killed a male partner (n=41, 66.1%) and two fifths of cases where a male homicide offender killed a female partner (n=96, 42.9%).

The data consistently demonstrates that in the majority of cases, notwithstanding who was ultimately killed, these orders named the female party as the person in need of protection from their male intimate partner. This was the case in over half of the 21 cases where there was a current order between a female IPV homicide offender and the male partner she killed at the time of the homicide (n=12, 57.1%). In a similar proportion of cases where there was a historical domestic violence order between the parties, the female homicide offender had been named as the person in need of protection from the male partner she ultimately killed (n=17, 58.6% of the 29 historical protection orders). This demonstrates that in over half of the cases where a female IPV homicide offender killed a male partner, there had been a reported history of domestic violence against the woman by her male partner and there had been prior police or court intervention as a result of his violence.

Domestic violence orders were a feature in just less than half of the cases where a male offender killed a female victim. It is evident that in the majority of cases, the female homicide victim was named as the person in need of protection from the male partner who killed her. This was evident in almost 90 per cent of the 49 cases where there was a current order at the time of the homicide (n=44, 89.8%) and in almost 80 per cent of 67 cases where there was a historical order (n=53, 79.1%). Accordingly, while domestic violence orders were a feature of just less than half of the male-perpetrated IPV homicides against women, the vast majority of those orders were imposed to protect the female homicide victim from the male offender prior to her death.

In 128 cases there was no evidence of a current or historical domestic violence order between the male homicide offender and female victim (57.1%). There are numerous reasons why there may not have been a domestic violence order in place for a relationship that exhibited a history of domestic violence. In many cases, the violence may never have been disclosed to police. Analysis of the Australian Bureau of Statistics’ Personal Safety Survey demonstrates that only one in three women who have been physically assaulted and one in five women who have been sexually assaulted by a male cohabiting partner reported the most recent incidence of violence to the police (Cox, 2015, p. 105). It is far more common for women to disclose these types of assaults to informal supports such as family, friends, colleagues or religious ministers (Cox, 2015). In some cases, it was evident that there was no domestic violence order in place because the police did not adequately recognise and respond to the domestic violence by applying for an order, or the order may not have been granted by the Court.

Separation as a characteristic of IPV homicide

Actual separation or an intention to separate was a feature in more than half of the cases where a male IPV homicide offender killed a female intimate partner (n=130, 58.0%). In over a third of cases, the relationship between the male offender and the female victim had ended prior to the homicide (n=77, 34.4%), and the majority of relationships had ended within three months of the homicide (n=44, 57.1% of separated couples). In over a third of cases where the relationship was ongoing, one or both parties had expressed their intention to separate (n=53, 36.1% of the 147 ongoing relationships). In the vast majority of these cases, the female homicide victim had expressed her intention to separate from the male offender (n=50, 94.3% of cases where an intention to separate had been expressed). Further, 38 of these 50 female homicide victims had expressed their intention to separate within three months of the homicide. This accounts for 76 per cent of female victims who expressed their intention to separate from the homicide offender and over 25 per cent of ongoing relationships between the male offender and female victim (25.9%). This data demonstrates that the period prior to and immediately after separation may be particularly dangerous for women and reinforces the need to strengthen support and protection for women who intend to separate or have recently separated from their abusive partner.

Separation rates for incidents of female-perpetrated IPV homicide and male-perpetrated IPV homicide against a male partner are relatively small, making it difficult to derive meaningful disaggregation of this nature.

Domestic violence behaviours used by domestic violence abusers

Abusive men who killed their female intimate partners employed a range of different tactics to dominate and control their partners prior to the homicide. While physically violent behaviours are often the most common form of abuse associated with domestic violence, the findings reveal a range of non-physical forms of violence as a feature in domestic violence relationships prior to IPV homicide. Of the 212 men who were the primary abusers of the female partner they killed, the most frequently identified domestic violence behaviour was emotional and psychological abuse, such as verbally denigrating, threatening, blaming or gaslighting the victim (n=173, 81.6%). Physically violent behaviours, such as slapping, punching, pushing or strangling the victim, were the next most frequently identified type of abuse (n=169, 79.7%). Tactics of social abuse – including isolating the victim from support networks and controlling her movements – were evident in over 60 per cent of cases (n=134, 63.2%), and economic or financial abuse to create dependency on the abuser was evident in just over a quarter of cases (n=58, 27.4%). Sexual violence was a feature in 34 cases (16.0%).

Of the 212 cases where a male primary domestic violence abuser killed a female victim, the male homicide offender stalked the woman he killed in 88 cases (41.5%). Stalking includes a range of behaviours used by an abuser to intimidate or control a victim or to make them feel fearful. It includes intentional and persistent behaviours such as following a victim, surveillance of the victim at their home or workplace, sending the victim harassing messages and monitoring their phone or email. In 71 cases, the male homicide offender stalked the female victim during the relationship (33.5%) and in 44 cases stalking occurred after the relationship ended (20.8%).

This dataset demonstrates the range of physical and non-physical abuse employed by abusers in a domestic violence relationship to maintain dominance and control over their victims. The high prevalence of non-physical forms of violence, such as emotional or psychological abuse, social abuse and stalking, highlights the importance for services and first responders to recognise the pattern of abusive and controlling behaviour used by male abusers and demonstrates that any relationship that exhibits domestic violence, whether physical or non-physical, is embedded with a risk of lethality. Services need to work with domestic and family violence victims and survivors and abusers and across agencies to improve response and prevention efforts, while holding abusers to account.

Conclusion

This report presents the national data on IPV homicide in Australia from July 2010 to June 2018, with a specific focus on the characteristics that precede these types of homicides. It draws on the expertise of the Network and the extensive primary source materials available to Network members to create a national minimum dataset on IPV homicide. This primary source material sets this dataset apart from other national studies exploring homicide in Australia, as the Network has greater access to government agency files and records to analyse the characteristics and behaviours that occurred throughout the homicide victims’ and offenders’ life courses.

The data demonstrates the highly gendered nature of IPV homicide, with the majority of cases involving a male offender and female victim. In terms of the relationship details, the data demonstrates that in the majority of cases, domestic violence is used by the man against his female partner. In most cases this results in the male abuser killing his female partner, however, there is also evidence that in many cases women kill partners who have been abusive towards them. In only a handful of cases, the woman is the primary abuser in the relationship.

The homicide characteristics demonstrate that the victim’s residence, whether shared with their partner or not, is generally the most common site of an IPV homicide, although homicides may also occur in other places such as the offender’s residence, public spaces, someone else’s residence or workplaces.

Analysis of the demographic details of homicide offenders and victims demonstrate that IPV homicide occurs across a broad age range, from as young as 16 through to over 80 years of age. Across all cohorts, the majority of offenders and victims were born in Australia, with the rates for those born outside of Australia consistent with or below national migration population statistics.

There is a significant overrepresentation of Aboriginal and Torres Strait Islander people in the dataset, as both homicide victims and offenders. While not examined in depth, this report has sought to highlight the complex interrelated factors that must be considered when examining these cases, such as colonisation, dispossession, child removal, and institutional and structural violence, which impact Aboriginal and Torres Strait Islander peoples’ experiences of family violence and access to appropriate support services (Adams et al., 2017; Cripps & Adams, 2014; Langton et al., 2020a). Understanding these complexities provides an important framework to better understand approaches to respond to and prevent family violence, which must be undertaken in collaboration with Aboriginal and Torres Strait Islander men and women (Adams et al., 2017). Further, services need to do more to respond to the needs of Aboriginal and Torres Strait Islander clients, to ensure they collect consistent and accurate data, and to address structural barriers including bias, racism and privacy concerns which may reduce a victim’s and survivor’s confidence in accessing and disclosing to the service (Langton et al., 2020b; NSW Aboriginal Affairs, 2015).

Only approximately one third of all IPV homicide offenders and victims were engaged in paid employment at the time of the homicide (n=225 out of 622, 36.2%). As highlighted throughout the report, this is of significance because workplaces can offer a site of intervention for domestic and family violence (Domestic Violence Death Review Team, 2020; Special Taskforce on Domestic and Family Violence in Queensland, 2015).

This report also explores data relating to domestic violence characteristics, including separation, family law proceedings, domestic violence orders and abusive behaviours. This data is derived from a focused data subset, which includes the cases from jurisdictions where a domestic violence death review mechanism operates.

The data demonstrates that there is a heightened vulnerability for women who separate or intend to separate from their partners prior to the homicide, with actual or intended separation being a feature in over half of the male-perpetrated IPV homicides. The majority of actual separations in this cohort occurred within three months of the homicide. Similarly, where an intention to separate had been expressed, the majority involved the female partner expressing her intention to separate within three months of when she was killed. It is important that services recognise the heightened vulnerability women may experience when separating from an abusive partner.

Given the relatively high rates of actual or intended separation in this dataset, active family law proceedings were found in only a small number of cases (n=11, 3.8%). It is important to recognise that not all cases of separation or divorce are escalated to the family law court. In many cases, disputes relating to separation, property division or childcare arrangements are settled informally.

Over 20 per cent of female IPV homicide victims were named as the protected person in a domestic violence order at the time of their death. These orders were imposed to protect her from the male partner who killed her. A history of domestic violence orders was also evident in almost 30 per cent of male-perpetrated IPV homicides against a female partner. The data demonstrates that in over 40 per cent of male-perpetrated IPV homicides against a female partner, there was a history of police or court intervention due to domestic and family violence prior to the homicide.

Current or historical domestic violence orders were even more prevalent in cases where a female IPV offender killed a male intimate partner, which were present in almost two thirds of these cases. In over half of these cases, the female homicide offender was protected from the male homicide victim (57.1% of current orders and 58.6% of historical orders). There was a higher rate of cross-orders protecting both parties from each other in female-perpetrated IPV homicide incidents. These results should be read in conjunction with the emerging literature on systems abuse and research on the misidentification of the person most in need of protection (Boxall et al., 2020; Nancarrow et al., 2020; Ulbrick & Jago, 2018).

There was also a significant proportion of cases where no domestic violence order was evident, either at the time of the offence or historically. This may be due to the violence never being disclosed to authorities, the police not issuing an order when called out to an incident, or the Court not granting an order.

Finally, this report explored the types of abusive behaviours used prior to the fatal assault. This dataset demonstrates the range of physical and non-physical abuse employed in a domestic violence relationship to maintain dominance and control over the victims. The high prevalence of emotional and psychological abuse (such as verbally denigrating, threatening, blaming or gaslighting the victim) and social abuse (such as isolating the victim from support networks and controlling her movements) demonstrates the need for services and first responders to recognise, beyond the use of physical violence, the pattern of coercive and controlling behaviours that are used by domestic violence abusers. Further, the diverse range of abusive tactics identified in this dataset, including physical, emotional, social, financial and sexual violence and stalking, suggests that any relationship that exhibits domestic violence, whether physical or non-physical, is embedded with a risk of lethality.

Footnotes

1. This accords with the definition of family violence contained in the Family Law Act 1975 (Cth), which is adopted by the Network. ↑
2. Domestic homicide is a classification used by the Australian Institute of Criminology’s National Homicide Monitoring Program. The term refers to “incidents involving the death of a family member or other person in a domestic relationship. Domestic homicide incidents include: intimate partner homicide; filicide; parricide; siblicide; and other family homicide” (Australian Bureau of Statistics, 2018). ↑
3. The next stage will be to develop an NMDS on filicides that were preceded by an anecdotal or reported history of domestic and family violence. ↑
4. While it is not within the scope of this research to explore these complexities in their entirety, some examples are explored throughout the report alongside the relevant data on Aboriginal and Torres Strait Islander offenders and victims. ↑
5. Domestic homicide is a classification used by the Australian Institute of Criminology’s National Homicide Monitoring Program. The term refers to “incidents involving the death of a family member or other person in a domestic relationship. Domestic homicide incidents include: intimate partner homicide; filicide; parricide; siblicide; and other family homicide” (Australian Bureau of Statistics, 2018). ↑
6. Data from these sources has been drawn from the National Coronial Information System (NCIS). ↑
7. The type of data available on NCIS can vary from one jurisdiction to another. For many of the cases drawn from NCIS, the data would generally consist of sentencing remarks, autopsy and toxicology reports, and a police report from the incident. ↑
8. Western Australia’s death review team provided data for July 2012 to June 2018, with data for July 2010 to June 2012 extracted from NCIS. ↑
9. Data from Western Australia for the focused dataset is for July 2012 to June 2018. ↑
10. For example, convictions include assault occasioning bodily harm or death, or dangerous operation of a motor vehicle. ↑
11. There are various reasons that may lead to an acquittal or charges not being laid. For example, in some cases the victim’s pre-existing health conditions made it difficult to establish the cause of death to the criminal standard. ↑
12. Drawing on the Australian Institute of Health and Welfare’s definition, a psychiatric disability includes “clinically recognisable symptoms and behaviour patterns frequently associated with distress that may impair functioning in normal social activity. Psychiatric disability may be associated with schizophrenia, affective disorders, anxiety disorders, addictive behaviours, personality disorders, stress, psychosis, depression and adjustment disorders, but dementias, specific learning disorders (such as attention deficit disorder) and autism are excluded” (Australian Institute of Health and Welfare, 2021). ↑
13. Workplaces may also be a site of continuing violence, for example, in circumstances where a homicide offender and victim are employed by the same organisation. ↑
14. People who are not engaged in formal paid employment may be engaged in other types of work including home duties, care responsibilities and volunteer work. ↑
15. For the purpose of this study, a disability pension is a financial support allowance provided to a person with a physical, intellectual or psychiatric condition that stops them from working. A retiree or pensioner refers to a person who receives an age pension in circumstances where they are no longer working, or a person who is otherwise retired and no longer working in a paid capacity. ↑
16. For the purpose of this dataset, the term “drug” refers to a pharmacological or non-therapeutic substance used for non-medical purposes. Drugs may include illicit or other non-illicit substances that are being used in ways contrary to their intended medical or other purpose. ↑
17. One case was finalised through coronial findings as the homicide offender suicided after the homicide. ↑
18. People who are not engaged in formal paid employment may be engaged in other types of work including home duties, care responsibilities and volunteer work. ↑
19. The term “joint parents” refers to children that the offender and victim had together, including biological, adopted and foster children. Data on children from the source material is not always accurate or complete, especially in the case of stepchildren. This means that the number of surviving children may be underreported. ↑
20. The Australian Capital Territory is in the process of establishing a death review mechanism. ↑
21. This was identified as a priority by the Australian Human Rights Commission (2016). ↑
22. Adopting the definition in J.M. Last (Ed.). (2001). A Dictionary of Epidemiology (4th ed.). Oxford University Press. ↑
23. Cussen, T., & W. Bryant, W. (2015). Domestic/family homicide in Australia. Australian Institute of Criminology. ↑
24. Australian Human Rights Commission. (2017). A National System for Domestic and Family Violence Death Review. AHRC. ↑
25. Recognising the specific vulnerabilities associated with people residing in rural and remote locations, or challenges associated with different service systems in these areas, members may need to consider a way to standardise and code this information across jurisdictions to allow for appropriate analysis, while retaining privacy and confidentiality of individual cases. ↑

References

Adams, M., Bani, G., Blagg, H., Bullman, J., Higgins, D., Hodges B., Hovane, V., Martin-Pederson, M., Porter, A., Sarra, G., Thorpe, A., & Wenitong, M. (2017). Towards an Aboriginal and Torres Strait Islander violence prevention framework for men and boys. White Ribbon. https://www.whiteribbon.org.au/awcontent/whiteribbon/documents/White-Ribbon-Australia-Towards-an-Aboriginal-and-Torres-Strait-Islander-violence-prevention-framework.pdf

Australia’s National Research Organisation for Women’s Safety. (2019). Intimate partner sexual violence: Research synthesis (2nd ed.; ANROWS Insights, 08/2019). https://www.anrows.org.au/publication/intimate-partner-sexual-violence-research-synthesis/

Australian Bureau of Statistics. (2017). Personal Safety, Australia, 2016 (Cat. no. 4906.0). https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/4906.0main+features12016

Australian Bureau of Statistics. (2018). National Homicide Monitoring Program. Directory of Family, Domestic, and Sexual Violence Statistics, 2018. https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4533.0~2018~Main%20Features~National%20Homicide%20Monitoring%20Program~19

Australian Bureau of Statistics. (2019a). Aboriginal and Torres Strait Islander women’s experiences of family and domestic violence. National Aboriginal and Torres Strait Islander Social Survey, 2014-15. https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4714.0~2014-15~Feature%20Article~Aboriginal%20and%20Torres%20Strait%20Islander%20women’s%20experiences%20of%20family%20and%20domestic%20violence%20(Feature%20Article)~10100

Australian Bureau of Statistics. (2019b). Disability, ageing and carers, Australia: Summary of findings. https://www.abs.gov.au/statistics/health/disability/disability-ageing-and-carers-australia-summary-findings/latest-release

Australian Bureau of Statistics. (2019c). Estimates and projections, Aboriginal and Torres Strait Islander Australians. https://www.abs.gov.au/statistics/people/aboriginal-and-torres-strait-islander-peoples/estimates-and-projections-aboriginal-and-torres-strait-islander-australians/2006-2031#australia

Australian Bureau of Statistics. (2021a). Migration, Australia 2019–2020. https://www.abs.gov.au/statistics/people/population/migration-australia/2019-20#key-statistics

Australian Bureau of Statistics. (2021b). Table 6: Population, by age and sex – States and territories [at 30 June 2020]. National, state and territory population, December 2020. https://www.abs.gov.au/statistics/people/population/national-state-and-territory-population/dec-2020

Australian Domestic and Family Violence Death Review Network. (2018). Data report 2018. https://www.coronerscourt.vic.gov.au/sites/default/files/2018-11/website%2Bversion%2B-%2Badfvdrn_data_report_2018_.pdf

Australian Human Rights Commission. (2016). A national system for domestic and family violence death review. https://humanrights.gov.au/sites/default/files/document/publication/AHRC_2016_12_19_Expanding_DV_Death_Review.pdf

Australian Institute of Health and Welfare. (2019). Family, domestic and sexual violence in Australia: Continuing the national story 2019. https://doi.org/10.25816/5ebcc837fa7ea

Australian Institute of Health and Welfare. (2021). Mental health services in Australia. https://www.aihw.gov.au/reports/mental-health-services/mental-health-services-in-australia/report-contents/psychiatric-disability-support-services/key-concepts#10_psychiatric_disability

Boxall, H., Dowling, C., & Morgan, A. (2020). Female-perpetrated domestic violence: Prevalence of self-defensive and retaliatory violence. Trends & Issues in Crime and Criminal Justice (no. 584). Australian Institute of Criminology. https://doi.org/10.52922/ti04176

Bricknell, S., & Doherty, L. (2021). Homicide in Australia 2018–19 (Statistical Report no. 34). Australian Institute of Criminology. https://doi.org/10.52922/sr78153

Bugeja, L., Butler, A., Buxton, E., Ehrat, H., Hayes, M., McIntyre, S.-J., & Walsh, C. (2013). The implementation of domestic violence death reviews in Australia. Homicide Studies, 17(4), 353–374. https://doi.org/10.1177/1088767913494787

Butler, A., Buxton-Namisnyk, E., Beattie, S., Bugeja, L., Ehrat, H., Henderson, E., & Lamb, A. (2017). Australia. In M. Dawson (Ed.), Domestic homicides and death reviews: An international perspective (pp. 125–158). Palgrave Macmillan.

Campo, M., & Tayton, S. (2015). Intimate partner violence in lesbian, gay, bisexual, trans, intersex and queer communities. Australian Institute of Family Studies. https://aifs.gov.au/cfca/sites/default/files/publication-documents/cfca-resource-dv-lgbti-2020.pdf

Carlson, B., Day, M., & Farrelly, T. (2021). What works? Exploring the literature on Aboriginal and Torres Strait Islander healing programs that respond to family violence (Research report, 01/2021). ANROWS. https://www.anrows.org.au/publication/what-works-exploring-the-literature-on-aboriginal-and-torres-strait-islander-healing-programs-that-respond-to-family-violence/

Council of Australian Governments. (2011). The National Plan to Reduce Violence against Women and their Children 2010–2022. Commonwealth of Australia. https://www.dss.gov.au/women/programs-services/reducing-violence/the-national-plan-to-reduce-violence-against-women-and-their-children-2010-2022

Council of Australian Governments. (2019). The Fourth Action Plan of the National Plan to Reduce Violence Against Women and their Children 2010–2022. Commonwealth of Australia. https://www.dss.gov.au/women-publications-articles-reducing-violence/fourth-action-plan

Cox, P. (2015). Violence against women: Additional analysis of the Australian Bureau of Statistics’ Personal Safety Survey, 2012. ANROWS. https://www.anrows.org.au/publication/violence-against-women-in-australia-additional-analysis-of-the-australian-bureau-of-statistics-personal-safety-survey-2012/

Cripps, K.A., & Adams, M . (2014). Indigenous family violence: Pathways forward. In R. Walker, P. Dudgeon, & H. Milroy (Eds.), Working together: Aboriginal and Torres Strait Islander mental health and wellbeing principles and practice (pp. 399–416). Australian Government Department of the Prime Minister and Cabinet & Telethon Kids Institute/Kulunga Aboriginal Research Development Unit. http://aboriginal.telethonkids.org.au/kulunga-research-network/working-together-2nd-edition-(2014)/

Domestic and Family Violence Death Review and Advisory Board. (2019). Domestic and Family Violence Death Review and Advisory Board: 2018–19 Annual Report. Queensland Courts. https://www.courts.qld.gov.au/__data/assets/pdf_file/0006/630159/domestic-and-family-violence-death-review-and-advisory-board-annual-report-2018-19.pdf

Domestic Violence Death Review Team. (2020). NSW Domestic Violence Death Review Team report 2017–2019. Coroners Court New South Wales. https://www.coroners.nsw.gov.au/coroners-court/resources/domestic-violence-death-review.html

Douglas, H., & Chapple, K. (2021). National Domestic and Family Violence Bench Book.

Douglas, H., & Fitzgerald, R. (2018). The domestic violence protection order system as entry to the criminal justice system for Aboriginal and Torres Strait Islander people. International Journal for Crime, Justice and Social Democracy, 7(3), 41–57. https://doi.org/10.5204/ijcjsd.v7i3.499

Griffiths, K., Coleman, C., Al-Yaman, F., Cunningham, J., Garvey, G., Whop, L., Pulver, L.J,. Ring, I., & Madden, R. (2019). The identification of Aboriginal and Torres Strait Islander people in official statistics and other data: Critical issues of international significance. Statistical Journal of the IAOS, 35(1), 91–106. https://doi.org/10.3233/SJI-180491

Koleth, M., Serova, N., & Trojanowska, B. K. (2020). Prevention and safer pathways to services for migrant and refugee communities: Ten research insights from the Culturally and Linguistically Diverse Projects with Action Research (CALD PAR) initiative (ANROWS Insights, 01/2020). ANROWS. https://www.anrows.org.au/publication/prevention-of-violence-against-women-and-safer-pathways-to-service/

Langton, M., Smith, K., Eastman, T., O’Neill, L., Cheesman, E., & Rose, M. (2020a). Family violence policies, legislation and services: Improving access and suitability for Aboriginal and Torres Strait Islander men. ANROWS. https://www.anrows.org.au/publication/family-violence-policies-legislation-and-services-improving-access-and-suitability-for-aboriginal-and-torres-strait-islander-men/

Langton, M., Smith, K., Eastman, T., O’Neill, L., Cheesman, E., & Rose, M. (2020b). Improving family violence legal and support services for Aboriginal and Torres Strait Islander women. ANROWS. https://www.anrows.org.au/project/improving-family-violence-legal-and-support-services-for-indigenous-women/

Lievore, D. (2003). Non-reporting and hidden reporting of sexual assault: An international literature review: Australian Institute of Criminology. https://www.aic.gov.au/publications/archive/archive-135

Mansour, J. (2014). Women defendants to AVOs: What is their experience of the justice system? Women’s Legal Services NSW. http://www.wlsnsw.org.au/wp-content/uploads/womendefAVOsreport.pdf

Nancarrow, H., Thomas, K., Ringland, V., & Modini, T. (2020). Accurately identifying the “person most in need of protection” in domestic and family violence law (Research report, 23/2020). ANROWS. https://www.anrows.org.au/publication/accurately-identifying-the-person-most-in-need-of-protection-in-domestic-and-family-violence-law/

NSW Aboriginal Affairs. (2015). Aboriginal identification: The way forward – An Aboriginal peoples’ perspective. NSW Government. https://www.create.nsw.gov.au/wp-content/uploads/2010/10/Aboriginal-identification-in-NSW-the-way-forward.pdf

Olsen, A., & Lovett, R. (2016). Existing knowledge, practice and responses to violence against women in Australian Indigenous communities: Key findings and future directions. ANROWS. https://www.anrows.org.au/publication/existing-knowledge-practice-and-responses-to-violence-against-women-in-australian-indigenous-communities-state-of-knowledge-paper/

Our Watch. (2018). Changing the picture: Background paper – Understanding violence against Aboriginal and Torres Strait Islander women. https://media-cdn.ourwatch.org.au/wp-content/uploads/sites/2/2020/09/20231756/Changing-the-picture-Part-1-AA.pdf

Royal Commission into into Violence, Abuse, Neglect and Exploitation of People with Disability. (2020). Issues paper: Violence and abuse of people with disability at home. https://disability.royalcommission.gov.au/system/files/2020-12/Issues%20paper%20-%20Violence%20and%20abuse%20of%20people%20with%20disability%20at%20home.pdf

SNAICC National Voice for Our Children, National Family Violence Prevention Legal Services Forum, & National Aboriginal and Torres Strait Islander Legal Services. (2017). Strong families, safe kids: Family violence response and prevention for Aboriginal and Torres Strait Islander children and families. https://www.snaicc.org.au/wp-content/uploads/2017/09/Strong_Families_Safe_Kids-Sep_2017.pdf

Special Taskforce on Domestic and Family Violence in Queensland. (2015). Not now, not ever: Putting an end to domestic and family violence in Queensland. Queensland Government. https://www.publications.qld.gov.au/dataset/not-now-not-ever/resource/533db62b-b2c9-43cc-a5ff-f9e1bc95c7c7

Stark, E. (2007). Coercive control: How men entrap women in personal life: Oxford University Press.

Ulbrick, M., & Jago, M. (2018). “Officer she’s psychotic and I need protection”: Police misidentification of the “primary aggressor” in family violence incidents in Victoria. Women’s Legal Service Victoria. https://womenslegal.org.au/files/file/WLSV%20Policy%20Brief%201%20MisID%20July%202018.pdf

Ussher, J. M., Hawkey, A., Perz, J., Liamputtong, P., Marjadi, B., Schmied, V., Dune, T., Sekar, J.A., Ryan, S., Charter, R., Thepsourinthone, J., Noack-Lundberg, K., & Brook, E. (2020). Crossing the line: Lived experience of sexual violence among trans women of colour from culturally and linguistically diverse (CALD) backgrounds in Australia (Research report, 14/2020). ANROWS. https://www.anrows.org.au/publication/crossing-the-line-lived-experience-of-sexual-violence-among-trans-women-of-colour-from-culturally-and-linguistically-diverse-cald-backgrounds-in-australia/

Voce, I., & Bricknell, S. (2020). Female perpetrated intimate partner homicide: Indigenous and non-Indigenous offenders (Statistical Report no. 20). Australian Institute of Criminology. https://doi.org/10.52922/sr04220

Wall, L. (2012a). Asking women about intimate partner sexual violence [ACSSA resource sheet]. Australian Centre for the Study of Sexual Assault, & Australian Institute of Family Studies. https://aifs.gov.au/publications/asking-women-about-intimate-partner-sexual-violence

Wall, L. (2012b). The many facets of shame in intimate partner sexual violence [ACSSA research summary]. Australian Centre for the Study of Sexual Assault, & Australian Institute of Family Studies. https://aifs.gov.au/publications/many-facets-shame-intimate-partner-sexual-violence

Workplace Gender Equality Agency. (n.d). Family and domestic violence. Retrieved 25 November 2021 from https://www.wgea.gov.au/take-action/family-domestic-violence

Appendix A:
Australian Domestic and Family Violence Death Review Network: Terms of reference

Background and position summary

Domestic and family violence has a devastating impact on individuals and communities. It is a complex phenomenon and includes child abuse, violence between siblings, violence by adolescents against parents, elder abuse, carer abuse, violence between same-sex partners, and violence perpetrated by women against their male intimate partners. However, in the overwhelming majority of cases, domestic and family violence is perpetrated by males against their female intimate partner.

Domestic and family violence can also be fatal. A significant proportion of all homicide victims are killed by a person with whom they share or have shared a domestic relationship, i.e. a current or former intimate partner or family member. Women are significantly overrepresented in this category of homicide.

Domestic and family violence deaths rarely occur without warning. In many fatal cases, there have been repeated incidents of abuse prior to the homicide, as well as identifiable indicators of risk. There have typically also been many opportunities for individuals or agencies to intervene before the death. When viewed as the escalation of a predictable pattern of behaviour, domestic and family violence deaths can be seen as largely preventable.

Domestic and Family Violence Death Review context

Background to establishment

For well over a decade, domestic and family violence death review processes have been operational in a number of international jurisdictions, most notably in the United States where domestic violence fatality review teams were first established in the early 1990s. Since that time, domestic and family violence death reviews have also been established in Canada, the United Kingdom and New Zealand.

The broad objective of these reviews is to identify potential areas for improvement in systemic responses to domestic and family violence. Domestic and family violence death reviews operate with a view to identifying patterns and commonalities between deaths for the purposes of reform. Such processes are effective in identifying and addressing weaknesses in service delivery and systems related to domestic and family violence.

In the mid-2000s, there was a call for the establishment of domestic and family violence death review processes in Australia. Over the past 12 years, Victoria, Queensland, New South Wales, South Australia, Western Australia and the Northern Territory have each implemented a domestic and family violence death review function with dedicated resources. In 2015 a pilot death review process was commenced in the Australian Capital Territory and it is currently in the process of establishing a permanent death review process.

The national policy context

The establishment of the Network aligns with Strategy 5.2 of the national policy agenda as detailed in the National Plan to Reduce Violence against Women and their Children 2010–2022. This mandates states and territories to work together to:

Strengthen leadership across justice systems.

Action 2: Drive continuous improvement through sharing outcomes of reviews into deaths and homicides related to domestic violence.

Immediate national initiatives: Monitor domestic violence-related homicide issues to inform ongoing policy development, including the Australian Institute of Criminology’s National Homicide Monitoring Program to research domestic violence-related homicides, risk factors and interventions.

Australian Domestic and Family Violence Death Review mechanisms

Victoria

The Victorian Systemic Review of Family Violence Deaths (VSRFVD) was established in 2009.

Positioned within the Coroners Court of Victoria and operating under the provisions of the Coroners Act 2008 (Vic), the VSRFVD assists with open coronial investigations of family violence-related deaths involving children and adults.

The VSRFVD has five main aims, which are to:

• examine deaths suspected to have resulted from family violence
• identify risk and contributory factors associated with deaths resulting from family violence
• identify trends and patterns in deaths resulting from family violence
• identify trends and patterns in responses to family violence
• provide coroners with information obtained through the exercise of the above functions.

The VSRFVD’s definitions of “family violence” and a “family member” are aligned with the Family Violence Protection Act 2008 (Vic) and the Victorian Indigenous Family Violence Taskforce Report (2003).

New South Wales

The Domestic Violence Death Review Team (DVDRT) was established in 2010 under the Coroners Act 2009 (NSW) to review deaths occurring in the context of domestic violence in New South Wales. Adopting both qualitative and quantitative review processes, the DVDRT aims to develop intervention and prevention strategies so as to reduce the likelihood of future deaths and to improve the response to domestic violence more generally.

Convened by the NSW State Coroner, the DVDRT is a multiagency committee constituted by representatives from key government stakeholders, including law enforcement, justice, health and social services, as well as four representatives from non-government agencies. The DVDRT is staffed by a secretariat constituting a manager and a research analyst.

The core legislative functions of the DVDRT are to:

• review and analyse individual closed cases of domestic violence deaths (as defined in the Coroners Act 2009)
• establish and maintain a database so as to identify patterns and trends relating to such deaths
• develop recommendations and undertake research that aims to prevent or reduce the likelihood of such deaths.

The DVDRT reports biennially to the NSW Parliament, setting out findings from the case and data analyses and the recommendations which are derived from these analyses.

The DVDRT secretariat also works with coroners on open cases of domestic violence-related deaths.

Queensland

The Domestic and Family Violence Death Review Unit (DFVDRU) was established in the Coroner’s Court of Queensland in January 2011 and provides assistance to coroners investigating domestic and family violence-related deaths under the Coroners Act 2003 (Qld). In 2014 the scope of the DFVDRU was expanded to include the deaths of children who were known to the child protection system.

The DFVDRU undertakes research in relation to domestic and family violence, which can be used to contextualise and inform coronial findings and recommendations. The DFVDRU assists coroners to formulate preventative recommendations for those investigations that proceed to inquest. The DFVDRU also maintains a dataset of domestic and family violence-related homicides and suicides.

The DFVDRU’s definitions align with the Domestic and Family Violence Protection Act 2012 (Qld). The DFVDRU is also responsible for the provision of administrative, secretariat and research support to the independent, multidisciplinary Domestic and Family Violence Death Review and Advisory Board, which was established in 2016 to enhance the systemic review of these types of deaths. Under the Coroners Act 2003 the Board is empowered to make recommendations and must submit an annual report to the Attorney-General on the performance of its functions.

South Australia

In response to election commitments made by the South Australian Government, the Office for Women and the South Australian Coroner’s Court have undertaken a partnership to research and investigate domestic violence-related deaths. The position of Senior Research Officer (Domestic Violence) was established in January 2011 as an initiative of the South Australian “A Right to Safety” (ARTS) reform agenda.

This position works collaboratively with the ARTS reporting and advisory structure and reports on outcomes to the Chief Executive Group (chaired by the Minister for the Status of Women) that oversees ARTS outcomes.

The position is based within the South Australian Coroner’s Office and works as part of the coronial investigation team to:

• identify deaths with a domestic violence context in order to assist in the investigation of the adequacy of system responses and/or interagency approaches which may prevent deaths occurring within that context
• review files, provide interim reports and have specific input into coronial inquests which relate to domestic violence
• develop data collection systems in order to inform coronial processes and identify demographic or service trends, gaps or improvements more broadly
• conduct specific retrospective research projects relevant to building a domestic violence death review evidence base.

The legislative basis for this position sits within the Coroners Act 2003 (SA). The definition of “domestic violence context” is aligned with the Intervention Orders (Prevention of Abuse) Act 2009 (SA).

Western Australia

On 1 July 2012, the Ombudsman commenced a new role to review family and domestic violence fatalities. For the purposes of this jurisdiction, a family or domestic relationship has the same meaning as given to it under s 4 of the Restraining Orders Act 1997 (WA).

The Ombudsman has a number of functions in relation to the review of family and domestic violence fatalities:

• reviewing the circumstances in which and why family and domestic violence deaths occur
• identifying patterns and trends that arise from reviews of family and domestic violence deaths
• making recommendations to public authorities about ways to prevent or reduce family and domestic violence deaths.

The Ombudsman reports comprehensively on family and domestic fatalities.

Northern Territory

The position of Research Officer (Family Violence) commenced in 2016 and is based within the Northern Territory Coroner’s Office.

The position operates under the provisions of the Coroners Act 1993 (NT) to assist open coronial investigations of domestic and family violence-related deaths by examining the context in which the death occurred and the adequacy of system responses to domestic and family violence to inform coronial findings and recommendations.

The position also maintains an evidence base so as to identify patterns and trends from reviews of family and domestic violence deaths. Currently that dataset is limited to intimate partner domestic violence-related deaths, but it is intended that the data collection will also extend to include other familial relationships where the death has been identified as domestic and family violence-related.

Common elements of review teams

The following are common elements across all existing Australian domestic and family violence death review mechanisms:

• Each is underpinned by the view that domestic and family violence-related deaths are largely preventable.
• Each operates in accordance with state-based legislation and state-determined governance structure.
• Each state clearly defines relationships and behaviours that amount to domestic and family violence.
• Each adopts review criteria which facilitate the review of homicides, homicide/suicides and suicides where such deaths have occurred in a context of domestic and family violence.
• Each reviews individual deaths with a domestic violence context as well as identifying data trends and patterns across multiple deaths.

ADFVDR Network overview

Following the implementation of domestic and family violence death review mechanisms in several Australian jurisdictions in recent years, the Australian Domestic and Family Violence Death Review Network (the Network) was established in March 2011. The Network comprises permanent representatives from each of the established Australian death review teams, namely the:

• Victorian Systemic Review of Family Violence Deaths (Vic)
• Domestic Violence Death Review Team (NSW)
• Domestic and Family Violence Death Review Unit (Qld)
• Domestic Violence Unit (SA)
• Reviews Team (WA)
• Family Violence Death Review Unit (NT).

The Network recognises that Tasmania and the Australian Capital Territory are exploring the implementation of a death review mechanism within their respective jurisdictions, and as such have not consolidated a final model of operating.

Representatives of these jurisdictions are also considered standing members of the Network where such a trial is being undertaken.

Special observer membership of the ADFVDRN

Special observers are invited to participate in discussions and Network processes but do not have decision-making authority. The addition of special observers recognises that domestic and family violence death review processes are established and operational outside of Australia and can contribute to the knowledge and development of the work undertaken by the Network.

Purpose

The overarching goals of the Network are to:

• improve knowledge regarding the frequency, nature and determinants of domestic and family violence deaths
• identify practice and system changes that may improve outcomes for people affected by domestic and family violence and reduce these types of deaths
• identify, collect, analyse and report data on domestic and family violence-related deaths
• analyse and compare domestic and family violence-related deaths
• analyse and compare domestic and family violence death review findings and recommendations.

Scope

The scope of the activities of the Network includes:

• using the learning and outcomes of state-based review processes to benefit the work of other Network members. This shall include comparing and reporting on findings across jurisdictions
• defining minimum case inclusion criteria and developing standardised minimum data sets across each jurisdiction to contribute to the development of minimum standard national data in relation to domestic and family violence-related deaths
• sharing information and evidence relating to the identification of domestic and family violence risk indicators and/or case characteristics.

Some key areas of consideration may include:

• identifying common risk indicators, case characteristics and/or system failures in the lead-up to a death
• the development of policies and recommendations to state and federal governments.

Governance

Membership

• Membership consists of persons or agreed representatives from each state-based domestic and family violence death review.
• Membership is closed and new membership and special observer requests will be determined by standing members of the Network, based on the compatibility of the function or unit with the purpose of the Network.
• Membership decisions will be formally documented and relayed to the requesting person or authority in writing by the Chairperson.
• Network meetings are restricted to Network members, officially recognised special observers and, by agreement, invited guests.
• The Network can, by agreement, request advice, support and/or consult with outside agencies or individuals as required.

Confidentiality provisions

• Maintaining confidentiality is critical to the functioning of the Network. Due to the sensitive nature of the information discussed, information discussed in the Network is confidential and non-disclosure requirements apply.
• Where the state-based death review is involved in reviewing open coronial matters there will be specific legislative confidentiality provisions required of each participant. It is the responsibility of individual members to be aware of and adhere to their particular legislative requirements regarding confidentiality.

Decision-making

• Each member state is responsible for making decisions in line with their employment and legislative responsibilities. This includes seeking appropriate permission, advice and authority to advance information or participate in decision-making where necessary.
• The Network operates within a consensus decision-making framework, which recognises the autonomy, and differing operating models, of each jurisdiction.
• As an underlying principle, this model will focus on identifying, and as much as practicable, addressing any individual member’s concerns to achieve the agreement of all jurisdictions.
• Where full agreement cannot be achieved on a particular course of action by the Network, but majority consensus has been reached, then this will be documented, but will not restrict the Network from undertaking a particular course of action.
• The Chairperson will document all decisions and actions arising from each Network meeting.

Meeting frequency

• Meetings will be held, either by teleconference or face-to-face, at least four times per year. Meetings may occur more frequently as determined by the needs of the Network.

Roles and responsibilities

Members

• All members are responsible for seeking relevant permissions, advice or authority before participating in decision-making and agree to adhere to the statutory or legislative requirements of their role.
• All members agree to contribute and cooperate in good faith and declare any conflict of interest or other disclaimers at the first possible opportunity or realisation of that conflict.
• All members may submit agenda items and papers for consideration by the Network and should endeavour to do so in a timely fashion for inclusion in the meeting agenda.
• Each member is responsible for keeping their own records of discussion from meetings.

Chairperson

The position of Chairperson will rotate between members on an annual basis. Appointment of the Chairperson will be by agreement of the Network members at the end of each calendar year and should not be undertaken in consecutive years by any representative from the same state.

The roles and responsibilities of the Chairperson include:

• preparing and disseminating the meeting agenda and relevant documents in a timely manner
• ensuring the Network operates in a manner consistent and in alignment with the terms of reference
• moderating decision-making processes
• minuting all decisions and actions arising from each meeting and distribution of these minutes to members as soon as practicable after the conclusion of each meeting
• maintaining a history of all documents produced as part of the Network and transferring that catalogue of information to the next nominated Chairperson
• with prior agreement by the Network, distributing information about the Network, making comment on Network matters (as appropriate), responding to enquiries and correspondence, requests for membership or meeting attendance and other such matters.

Partner project with Australia’s National Research Organisation for Women’s Safety (ANROWS)

ANROWS is an independent, not-for-profit organisation established by the Commonwealth and all state and territory governments of Australia as an initiative under Australia’s National Plan to Reduce Violence against Women and their Children 2010–2022 (the National Plan). ANROWS was formally constituted in February 2013. It was established to build the evidence base through a program of nationally relevant research, and to facilitate the take-up of evidence in policy and practice to support effective implementation of the National Plan. ANROWS is based in New South Wales.

In 2020, ANROWS and the Network established a memorandum of understanding, valid for two years, to facilitate collaboration between the parties to produce the deliverables for the Australian Domestic and Family Violence Death Review National Data Update project (the project).

ANROWS has been commissioned by the Department of Social Services (DSS) to establish a dedicated program of research to support the Fourth Action Plan of the National Plan to Reduce Violence against Women and their Children 2010–2022. The project will be funded under this arrangement.

The project will include three deliverables:

1. Produce the next iteration of the Australian Domestic and Family Violence Death Review Network Data Report, to include IPH data 1 July 2010 to 30 June 2018.
2. Analyse data held by the Network to identify risk factors present in IPHs in Australia.
3. Develop a national minimum data set for filicide.

The MoU will be monitored by a Network Project Steering Committee (the Steering Committee), which includes representatives from the Network and ANROWS and meets monthly to monitor the currency of the MOU and the effectiveness of collaborations, and to seek to resolve any issues of concern to either of the parties.

The Network member states and territories will retain data ownership as outlined by the Network data sharing protocols. All collated, de-identified data will remain the intellectual property of the Network member states and territories.

Last updated July 2021

Appendix B:
Australian Domestic and Family Violence Death Review Network: Domestic and Family Violence Homicide Consensus Statement

Background and purpose

Following the implementation of domestic and family violence death review mechanisms in several Australian jurisdictions, the Australian Domestic and Family Violence Death Review Network (the Network) was established in March 2011. The Network comprises representatives from each of the established Australian death review teams, namely:

• Domestic Violence Death Review Team (New South Wales)
• Domestic and Family Violence Death Review Unit (Queensland)
• Domestic and Family Violence Death Review (South Australia)
• Victorian Systemic Review of Family Violence Deaths
• Review Team Ombudsman Western Australia
• Family Violence Death Review Unit (Northern Territory).

The overarching goals of the Network are to, at a national level:

• improve knowledge regarding the frequency, nature and determinants of domestic and family violence deaths
• identify practice and system changes that may improve outcomes for people affected by domestic and family violence and reduce these types of deaths
• identify, collect, analyse and report data on domestic and family violence-related deaths
• analyse and compare domestic and family violence death review findings and recommendations.

These goals align with the National Plan to Reduce Violence Against Women and their Children 2010–2022.

Definitions

This Consensus Statement defines the inclusion criteria adopted by the Network for domestic and family violence homicide. While there is no universally agreed definition of the behaviours that comprise domestic and family violence, in Australia it includes a spectrum of physical and non-physical abuse within an intimate or family relationship. Domestic and family violence behaviours include physical assault, sexual assault, threats, intimidation, psychological and emotional abuse, social isolation, and economic deprivation. Primarily, domestic and family violence is predicated upon inequitable relationship dynamics in which one person exerts power and coercive control over another. This accords with the definition of family violence contained in the Family Law Act 1975 (Cth), which is adopted by the Network.

The definition of homicide adopted by the Network is broader than the legal definition of the term. “Homicide”, as used by the Network, includes all circumstances in which an individual’s intentional act, or failure to act, resulted in the death of another person, regardless of whether the circumstances were such as to contravene provisions of the criminal law.

Surveillance

The World Health Organization defines surveillance as “systematic ongoing collection, collation and analysis of data and the timely dissemination of information to those who need to know so that action can be taken”.^[22]

Surveillance processes produce data that describe the frequency and nature of mortality and morbidity at the population level. This serves as a first step to the identification of risk factors to target preventive intervention. The Network applies these principles to ensure a consistent and standardised approach to data collection and analysis. To identify the target population and opportunities for intervention, surveillance of domestic and family violence homicide incidents is conducted both retrospectively and prospectively.

Categorisation

Identification and classification of domestic and family violence deaths is complex and needs to be conducted cautiously. The key considerations in this area are:

1. the case type
2. the role of human purpose in the event resulting in a death (intent)
3. the relationship between the parties (i.e. the deceased–offender relationship)
4. the domestic and family violence context (i.e. whether or not the homicide occurred in a context of domestic and family violence).

Consideration 1: Case type

Determination of case type (i.e. external cause, natural cause, unknown cause) is the first consideration for classification. An external cause death is any death caused, directly or indirectly, by an offender through the application of assaultive force or by criminal negligence. In cases where the cause of death is unknown, the death is monitored until further information is available.

Case type	Definition	Inclusion
External cause	Any death resulting directly or indirectly from environmental events or circumstances that cause injury, poisoning and/or other adverse effect	Yes
Unexplained cause	Deaths for which it is unable to be determined whether it was an external or natural cause	No
Natural cause	Any death due to underlying natural causes. Includes chronic illness due to long-term alcohol abuse/smoking	No

Consideration 2: Intent

The second consideration is to establish the role of human purpose in the event resulting in the external cause death. In accordance with the WHO International Classification of Disease (ICD-10), the intent is coded according to the following categories.

Intent	Definition	Inclusion
Assault*	Injury from an act of violence where physical force by one or more persons is used with the intent of causing harm, injury, or death to another person; or an intentional poisoning by another person. This category includes intended and unintended victims of violent acts (e.g. innocent bystanders)	Yes
Complications of medical or surgical care	Death which occurred due to medical misadventure, accidents or reactions in the administration of medical or surgical care drugs or medication	No
Intentional self-harm	Injury or poisoning resulting from a deliberate violent act inflicted on oneself with the intent to take one’s own life or with the intent to harm oneself	No
Legal intervention/operations of war	Death which occurred due to injuries that were inflicted by police or other law-enforcing agents (including military on duty), in the course of arresting or attempting to arrest lawbreakers, suppressing disturbances, maintaining order or other legal action	Yes (only where DV context present)
Still enquiring	Death under investigation whereby the intent or case type is not immediately clear based on the level of information available	No
Undetermined intent	Events where available information is insufficient to enable a person to make a distinction between unintentional, intentional self-harm and assault	No
Unintentional	Injury or poisoning that is not inflicted by deliberate means (that is, not on purpose). This category includes those injuries and poisonings described as unintended or “accidental”, regardless of whether the injury was inflicted by oneself or by another person	No
Unlikely to be known	Upon case completion, the coroner was unable to determine whether the death was due to natural or external causes, therefore unable to make a determination on intent	No

* Mortality classification systems refer to “homicide” as “assault”.

Consideration 3: Relationship

The third consideration for classification is whether a domestic or familial relationship existed between the deceased and the offender. The Network recognises the various state and federal legislative instruments that define and address deceased–offender relationship. In particular, it is acknowledged that the member jurisdictions operate within the following legislative frameworks:

• Coroners Act 2009 (NSW)
• Domestic and Family Violence Protection Act 2012 (Qld)
• Family Violence Protection Act 2008 (Vic)
• Intervention Orders (Prevention of Abuse) Act 2009 (SA)
• Restraining Orders Act 1997 (WA) and Parliamentary Commissioner Act 1971 (WA)
• Domestic and Family Violence Act 2007 (NT).

Each review team recognises current or former intimate partners (heterosexual and homosexual), family members (adults and children) and kin as relevant relationships. To standardise the inclusion and categorisation of relationship type, the following definitions are adopted by the Network.

Relationship type	Definition	Inclusion
Intimate**	Individuals who are or have been in an intimate relationship (sexual or non-sexual)	Yes
Relative***	Individuals, including children, related by blood, a domestic partnership or adoption	Yes
Aboriginal and/or Torres Strait Islander kinship relationships	A person who under Aboriginal and/or Torres Strait Islander culture is considered the person’s kin	Yes
No relationship	There is no intimate or familial relationship between the individuals	Yes (only where DV context present)
Unknown	Relationship is unknown	No

** This includes current and former intimate relationships irrespective of the gender of the individuals.
*** This includes formal and informal family-like relationships, and explicitly includes extended family-like relationships that are recognised within that individual’s cultural group.

Consideration 4: Domestic and family violence context

Having determined that a homicide has occurred and that a domestic relationship exists between the deceased and offender, the final consideration for classification is whether the homicide occurred in a domestic or family violence context. Deaths that fulfil these criteria are defined as domestic and family violence homicides and are subject to review by each jurisdiction.

Each jurisdiction can also review deaths where no direct domestic relationship exists between the deceased and offender but the death nonetheless occurs in a context of domestic and family violence. For example, this might include a bystander who is killed intervening in a domestic dispute or a new partner killed by their current partner’s former abusive spouse.

Similarly, the Network recognises that the existence of an intimate or familial relationship between a deceased and offender does not, in itself, constitute a domestic and family violence homicide. In these deaths, other situational factors determine the fatal incident, such as the offender experiencing an acute mental health episode. These deaths do not feature many of the characteristics known to define domestic and family violence, such as controlling, threatening or coercive behaviour; having previously caused the other person to feel fear; or evidence of past physical, sexual or other abuse.

Case inclusion criteria for the National Minimum Dataset on intimate partner homicides

Additional inclusion criteria was agreed upon for the development of the second edition of the Australian Domestic and Family Violence Death Review Network Data Report, to be published in 2021. This report focuses on intimate partner homicides, which narrows the “relationship type” inclusion criteria set out previously in the third consideration for classification. This report will include cases where:

• the death was as a result of a homicide that occurred in Australia between 1 July 2010 and 30 June 2018
• the homicide victim and homicide offender were either in a current or former intimate partner relationship
• there was an identifiable history of violence between the homicide victim and homicide offender
• the coronial or criminal proceedings in that homicide were complete on or before 31 December 2020.

Last updated July 2021

Appendix C:
Australian Domestic and Family Violence Death Review Network: Data sharing protocols

Purpose

The purpose of this document is to establish governance arrangements to allow for the sharing of data across jurisdictions to support the establishment of a National Minimum Dataset on domestic and family violence deaths.

It briefly discusses the policy landscape and national impetus for the development of a dataset, recognises the different governance processes within each jurisdiction that allow this data to be shared, and establishes specifications for which all jurisdictions that participate within this process agree to adhere to, for the purposes of appropriate data collection, storage and dissemination.

Background

For well over a decade, domestic and family violence death review processes have been operational in a number of international jurisdictions, most notably in the United States, where domestic violence fatality review teams were first established in the early 1990s.

Since that time, domestic and family violence death reviews have also been established in Canada, the United Kingdom, Australia and New Zealand, as well as in other jurisdictions. The broad objective of these reviews is to identify potential areas for improvement in systemic responses to domestic and family violence. Domestic and family violence death reviews operate with a view to identifying patterns and commonalities between deaths for the purposes of reform. Such processes are effective in identifying weaknesses in service delivery and systems, and opportunities to improve responses to domestic and family violence across the service system.

In the mid-2000s, after a long period of sector advocacy, there was a call for the establishment of domestic and family violence death review processes in Australia. Within the past decade, Victoria, Queensland, New South Wales, South Australia, Western Australia and the Northern Territory have each implemented a domestic and family violence death review function with dedicated, permanent resources.

The Australian Capital Territory is in the process of establishing a death review mechanism. There is currently no death review process in Tasmania, however, the Chief Coroner has granted the Network access to relevant case files on NCIS for the purpose of progressing the Network’s work.

Following the implementation of domestic and family violence death review mechanisms in several Australian jurisdictions, the Australian Domestic and Family Violence Death Review Network (the Network) was established in March 2011. The establishment of the Network aligned with Strategy 5.2 of the national policy agenda as detailed in the National Plan to Reduce Violence against Women and their Children 2010–2022 (the National Plan).

As detailed in Action 19 of the Second Action Plan, one of the overarching goals of the Network is to identify, collect, analyse and report data on domestic and family violence-related deaths, and share information, for the purposes of improving knowledge regarding these types of deaths.

To achieve this work, the Network has taken a number of steps to be able to comprehensively report these data at a national level.

This has included the establishment of a nationally consistent definition of a “domestic and family violence homicide” through the Homicide Consensus Statement which defines the inclusion criteria adopted by all members of the Network for implementation within their respective jurisdictional review mechanisms.

The Consensus Statement sets out the processes for identifying and classifying domestic and family violence homicides, taking into consideration the case type, the intent, the relationship between the deceased and the offender, and the domestic and family violence context of the death.

Further, building upon this standardised definition, the Network has also established data collection protocols to develop a staged, standardised, national dataset for domestic violence homicides, with the intent to ultimately extend data collection to include homicides within a family relationship, “bystander” homicides, and suicides that have been identified as domestic and family violence related.

To accommodate jurisdictional differences and mandates that govern the way in which the death review processes are conducted, this preliminary data collection covers all closed intimate partner domestic violence context homicides from 2008 onwards to allow for consistency in reporting across jurisdictions.

This dataset identifies specific data variables for collation which include homicide details; demographic details, and other characteristics for the deceased and offender; case characteristics; histories of violence; and relationship characteristics between the deceased and the offender.

Jurisdictional governance

With the majority of domestic and family violence death review mechanisms in Australia embedded within coronial jurisdictions, this paper recognises the legislative landscape which governs the management of data and information in relation to these types of deaths in each state or territory.

Each jurisdiction currently has processes in place to allow for the collection of data and information in relation to domestic and family violence deaths, which includes strict provisions as to when, how and why this information may be shared.

Queensland

In Queensland, data and information pertaining to domestic and family violence deaths is generated through a two-tiered review process, either through supporting coroners in their investigation of a relevant reportable death (Tier 1) or through the Domestic and Family Violence Death Review and Advisory Board, who are responsible for the systemic review of these types of deaths (Tier 2).

Under the Coroners Act 2003 (Qld) the State Coroner is responsible for approving the release of any data or information held in relation to the coronial jurisdiction. The Act specifies when and how this information may be shared, and what the State Coroner needs to consider when making a determination to release data or information.

Applicable provisions also allow for the State Coroner to specify how long a person may have access to coronial information and also provides for the State Coroner to withdraw their approval.

The Act further specifies that access to investigation documents must be de-identified except if the State Coroner is satisfied that the opportunity for increased knowledge that may result from the research outweighs the need to protect the privacy of any living or dead person.

While this Act mainly pertains to investigation documents generated through a coronial investigation, the principles outlined within the Act are extended to apply to data and information generated through the death review process as part of the coronial investigation.

Data in relation to these types of deaths are stored within a secure server, with access restricted to staff at the Coroners Court who are bound by relevant confidentiality requirements to ensure the safe storage of this type of information.

New South Wales

In New South Wales, data and information pertaining to domestic violence deaths is collected by the Domestic Violence Death Review Team (DVDRT) secretariat and housed in a purpose-built secure database. The DVDRT is convened by the NSW State Coroner.

The DVDRT was established with the insertion of Chapter 9A of the Coroners Act 2009 (NSW) and information sharing is governed by a number of sections within this Chapter. Under s 101F(4),

the Convenor may enter into an agreement or other arrangement for the exchange of information between the Team and a person or body having functions in another state or territory that are substantially similar to the functions of the Team, being information relevant to the exercise of the functions of the Team or that person or body.

Information sharing is also anticipated under s 101M of the Act, which provides exceptions to the strict confidentiality provisions governing the DVDRT’s operation and allows the Convenor to share data and information pursuant to an agreement or arrangement made under the Chapter.

South Australia

In South Australia, data and information relating to domestic and family violence deaths is gathered through the coronial investigation of a relevant reportable death. The Senior Research Officer (Domestic Violence) supports the Coroner to investigate deaths and produces detailed reports and analysis on all homicide deaths with a domestic violence context.

As well as informing the active coronial investigation, specific data and information, relating to South Australian homicides and suicides, is collected in the Coronial Domestic Violence Information System (CDVIS). The CDVIS is a purpose-built secure database used to house data and produce reports relating to the prevalence and context of homicides in South Australia. This data is reported in the State Coroner’s Annual Report.

The Coroners Act 2003 (SA), under s 38, provides discretion for the State Coroner, for the purposes of research, education, public policy development or for any other sociological purpose, to permit a person or body access and use of information derived from records of the Coroner’s Court. Furthermore, the provision of this information may be subject to such conditions as the State Coroner thinks fit.

Victoria

In Victoria, data and information pertaining to family violence deaths is collected by the Coroners Court of Victoria.

The Coroners Court of Victoria maintains a secure purpose-built Surveillance Database of all reviewable and reportable deaths in Victoria. The Victorian Coroners Court’s Victorian Homicide Register was established to draw from this database as the basis for the identification and collection of data which is utilised by the Victorian Systemic Review of Family Violence Deahs (VSRFVD).

Section 115(2) of the Coroners Act 2008 (Vic) provides that a Coroner may release a document to:

1. an interested party if the Coroner is satisfied that the party has a sufficient interest in the document
2. a statutory body if the Coroner is satisfied that the release of the document is required to allow the statutory body to exercise a statutory function
3. a police officer for law enforcement purposes
4. a person who is conducting research if the Coroner is satisfied that the research has been approved by an appropriate human research ethics committee
5. any person if the Coroner is satisfied that the release is in the public interest
6. a person specified in the rules as being a person to whom documents may be released.

The Coroners Act 2008 (Vic) also provides that a Coroner may impose conditions on the release of any document. Penalties apply if a person to whom a document has been released fails to comply with any condition placed on that release.

Western Australia

The Ombudsman commenced an important role to review all family and domestic violence fatalities on 1 July 2012. In doing so, the Ombudsman has all the powers provided for in the Parliamentary Commissioner Act 1971 (WA) (the Act) and all of the powers of a standing Royal Commission. In addition to information relating to the Ombudsman’s role to review family and domestic violence fatalities, significant information, data, collation and analysis regarding family and domestic violence arising from reviews undertaken is reported annually to Parliament.

The Ombudsman also undertakes major investigations of his own motion in relation to family and domestic violence fatalities. The first major own motion investigation, Investigation into issues associated with violence restraining orders and their relationship with family and domestic violence fatalities, was tabled in Parliament in November 2015. The report of the investigation contains extensive reporting and analysis of data and information regarding family and domestic violence fatalities in Western Australia and 54 recommendations to prevent or reduce family and domestic violence fatalities.

The Ombudsman also undertakes reporting of the steps taken to give effect to the recommendations arising from major own motion investigations. A report on giving effect to the recommendations arising from the Investigation into issues associated with violence restraining orders and their relationship with family and domestic violence fatalities was tabled in Parliament in November 2016.

Furthermore, subject to the relevant provisions of s 23(1b) of the Act, the Ombudsman may disclose information, or make a statement, to any person or to the public or a section of the public if, in his opinion, it is in the interests of any department or authority to which the Act applies or of any person, or is otherwise in the public interest.

Northern Territory

In the Northern Territory, data and information relating to domestic and family violence deaths is gathered through the coronial investigation of a relevant reportable death.

As well as informing the active coronial investigation, specific data and information relating to domestic and family violence-related homicides is collected in the Northern Territory coronial database which has restricted access.

There is no express provision in the Coroners Act 1993 (NT) that provides for the release or sharing of any data or information held in relation to coronial investigations.

However, in line with the overarching goal of the Network to collect, analyse and report on domestic and family violence related deaths at a national level, the Northern Territory agrees to provide such data required for the purposes of achieving its goal including for the development of the national minimum dataset.

All Northern Territory data is de-identified to ensure the protection of the privacy of individuals involved in coronial investigations.

Partnership project with ANROWS

In 2020, ANROWS and the Network established a memorandum of understanding, valid for two years, to facilitate collaboration between the parties to produce the deliverables for the Australian Domestic and Family Violence Death Review National Data Update project (the project).

The project will include three deliverables:

1. Produce the next iteration of the Australian Domestic and Family Violence Death Review Network Data Report, to include IPH data 1 July 2010 to 30 June 2018.
2. Analyse data held by the Network to identify risk factors present in IPHs in Australia.
3. Develop a national minimum data set for filicide.

The MoU sets out the data sharing protocols between the Network and ANROWS. In particular, it states that the parties may exchange confidential information relevant to projects and activity under the MoU.

Each party undertakes to treat as confidential all confidential information obtained from the other party and undertakes not to divulge any confidential information to any person without first obtaining the consent of the other party in writing.

Each party will take such reasonable steps to provide for the safe custody of any and all confidential information in its possession and to prevent unauthorised access thereto or use thereof.

At any time upon the written request of a party, the other party must return any documents which embody confidential information and must not keep any copies in any form.

Issues

Systemic monitoring and surveillance of relevant reportable death categories are a core component of any death review mechanism.

While they are a necessary first step in identifying cases that may benefit from a more detailed review, they also assist in developing an understanding of the prevalence and incidence of these types of deaths within any locality or jurisdiction. They may further assist in the identification of risk indicators or cohorts who may be at increased risk of harm, which enables a more targeted approach to prevention activities.

Despite the prevalence of deaths that occur in the context of domestic and family violence, there has not, until recently, been a mechanism for the systematic review of these deaths across all Australian jurisdictions.

Limitations with current processes for the collection of homicide data have been identified in a range of national reports. For example, the Australian Institute of Criminology has recently highlighted that qualitative incident-specific analysis is required to understand the nuances of precipitating events, personal characteristics of offenders and victims, and motives of perpetrators pertaining to domestic and family violence homicides.^[23]

This is not achievable through existing national data collection mechanisms.

The Australian Human Rights Commission^[24] has further identified that there is a lack of reliable reporting, in line with consistent definitions of domestic and family violence homicides. In particular, it was noted that the National Homicide Monitoring Program (NHMP) does not report on the context of domestic violence limiting the ability of this function to report on the nuances of this type of death.

Likewise, the National Coronial Investigation System (NCIS) does not reliably report on the context of how a person has died, focusing on the medical cause of death. As a data storage system for coronial information, the NCIS is not a system that is designed to support more nuanced analysis of these types of deaths.

While combining data generated through the death review process is not research in and of itself, there are key learnings that can be adopted from established research guidelines which can inform the consideration of how to administratively manage and share such information, including from the Australian Code for the Responsible Conduct of Research (the Code).

The Code promotes integrity in research, and describes the principles and practices for encouraging the responsible conduct of research for administrators, institutions and researchers. Applicable to this initiative they highlight areas for consideration by institutions for the management of data, and the publication and dissemination of research findings that have been used to inform the development of these protocols for the sharing of data across jurisdictions.

Notably, upon review of these guidelines, the legislative basis within which all of our respective death review mechanisms operate, and the existing jurisdictional mechanisms for the storage and retention of data and information generated through the review process, already supersede processes that are put in place to guide the conduct of responsible research.

In this regard, it is acknowledged at the outset that all members are required to comply with any governing legislation, policies and procedures applicable to their jurisdiction for the appropriate collation, storage and dissemination of data generated through their respective death review processes.

While individual processes may vary across jurisdictions, these protocols aim to instead establish a national standard for the storage, ownership and dissemination for data to be shared across jurisdictions for the sole purpose of the development of a national database on domestic and family violence-related deaths, with the ultimate aim of preventing future deaths.

Shared specifications

All data and information provided to inform the development of a national picture of domestic homicides is strictly confidential and will be treated as such, until such point as all members have formally agreed to its release.

While jurisdictions are empowered under their own legislative framework to manage their data as they consider it appropriate to do so, the following points apply to the custodianship and management of data provided by other jurisdictions to inform this initiative.

Data storage

Each member must take all necessary steps to ensure that data provided by any other member for the purposes of informing a national picture of domestic and family violence homicides is secure at all times.

This must include, but not be limited to, storage on a secure server with access restricted to members hosting the data storage.

As a general principle, where such data is transmitted electronically, this should only be communicated by means of a formal government department, agency or authority email, or encrypted data storage device and password protected. The password should be communicated and stored separately to this communication.

Data will be provided in a de-identified format only. This includes the removal of the following: name of offender, name of deceased, address of death,^[25] and identifying details pertaining to the specific circumstances of the death.

Data ownership

Data is provided by members for the purposes of improving knowledge regarding the frequency, nature and determinants of these types of deaths, and as such data cannot be used for any other purpose without the express permission of each contributing member.

Members retain all intellectual property rights and permissions to data that they have provided, including the right to withdraw their consent for this data and information to be stored or accessed by other members.

Should they make a determination to do so, member jurisdictions must advise the Network in writing that they withdraw their consent for this data and information to be accessed. In this event, every other jurisdiction must, as soon as practicable, take all steps necessary to permanently delete or destroy any information or data held by them that had been provided by the requesting jurisdiction. They must then confirm to the requesting jurisdiction that this has been completed in writing.

The exception to this specification are documents that are within the public domain, and that the requesting jurisdiction has previously provided consent to release publicly.

Ownership of the contributed data remains the property of the individual contributing member. As such each member must be consulted with, and agree to, the use of their data for inclusion in any project, document or report, or through presentation in any forum.

In the event that a member makes a determination that their data and information should not be included within any report or activity undertaken by the member, then this should not restrict other members from participating within this activity or report. It is preferable to note within any documentation produced by the members, that the report does not reflect the full membership of the Network.

Data dissemination

Members are expressly prohibited from referencing, or releasing, any data or information provided by another member without their express written consent.

Members commit to taking all reasonable steps to ensure that any data or findings are accurate and properly reported. Should members become aware of misleading or inaccurate statements about the data they have contributed they must take action to correct this as soon as practicable, including to notify the Network chair as soon as possible.

Review

This document will be reviewed annually to ensure it accords with the Network’s priorities, and can be reviewed at any time as requested by a participating jurisdiction.

Last updated July 2021