Journal of Family Violence https://doi.org/10.1007/s10896-018-9963-6

ORIGINAL ARTICLE

Counting Dead Women in Australia: An In-Depth Case Review of

Patricia Cullen1,2 & Geraldine Vaughan1 & Zhuoyang Li1 & Jenna Price3 & Denis Yu4 & Elizabeth Sullivan1,5

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract Gender-based fatal violence (femicide) is a preventable cause of premature death. The Counting Dead Women Australia (CDWA) campaign is a femicide census counting violent deaths of women in Australia from 2014. We conducted a cross-sectional in- depth review of CDWA cases Jan-Dec 2014 to establish evidence of antecedent factors and describe femicide in Australia. Victim (n = 81) and perpetrator (n = 83) data were extracted from the CDWA register, law databases and coronial reports. Mixed methods triangulation of socio-demographic and incident characteristics. Women ranged in age from 20 to 82 years of age (44 ± 15.4). There were 83 perpetrators, of which 13 were unknown (not yet apprehended). Known perpetrators (n = 70) ranged in age from 16 to 72 years of age (40 ± 12.7) and 89% were male (62/70). The location of the crime was most frequently the victim’shome (49/70). In cases where the relationship between the victim and perpetrator was known (n = 59), over half of were committed by intimate-partners (33/59). Intimate-partner perpetrators were more likely to have a history of violence and commit - than other perpetrators. Femicide is overwhelmingly perpetrated by males, with women most vulnerable in their own home and with their intimate partners. Furthermore, intimate-partner femicide is associated with modifiable risk factors, including previous violence and mental health issues, which represents opportunities for early intervention within healthcare settings as practitioners are well-placed to identify risk and provide support. In line with recommendations for multi-sectoral approach, future research should target identification of risk and protective factors, and improved coordination of data collection.

Keywords Femicide . Family violence . Intimate partner violence . Gender-based violence . . Women

Introduction death; however the perpetration of gender-based fatal violence continues to be a pervasive violation of the human rights of is a global public health problem, women. which at its most extreme, results in femicide – homicide of In Australia, the estimated femicide rate for 2010–2012 a woman regardless of the gender or intent of the perpetrator was 0.8 per 100,000 Australian women, of which two thirds (Mouzos 1999). Femicide is a preventable cause of premature were domestic that involved family members or intimate partners (Bryant and Cussen 2015). Women are vast- ly over-represented as victims in domestic homicides and in particular account for over 75% of all intimate partner homi- * Elizabeth Sullivan cides (Bryant and Cussen 2015): they are 3-times more likely [email protected] than men to be murdered in their home by a current or former 1 Australian Centre for Population and Public Health Research, The partner. Faculty of Health, University of Technology Sydney, The strongest predictors of intimate partner femicide are: Ultimo, Australia history of intimate partner violence (IPV) and; during relation- 2 The George Institute for Global Health, The University of New South ship estrangement, post-separation and at re-partnering (Beyer Wales, Sydney, Australia et al. 2013; Commonwealth of Australia 2009, 2016b;Du 3 School of Communication, Faculty of Arts and Social Sciences, Plat-Jones 2006; Petrosky et al. 2017). Other risk factors in- University of Technology Sydney, Ultimo, Australia clude perpetrator unemployment and residing with a child 4 Steinhardt School of Culture, Education, and Human Development, who is not biologically linked to the perpetrator, while prior New York University, New York, USA arrest of a perpetrator was found to be a protective factor 5 PO Box 123, Broadway, NSW 2007, Australia (Beyer et al. 2013). In contrast to intimate partner femicide, JFamViol there is little literature available on non-intimate partner far reaching physical, psychological, social and economic femicide and its risk factors, although it is considered to be consequences of violence (Commonwealth of Australia quite distinct from other types of femicide. Perpetrator crimi- 2009, 2016a). Furthermore, there are calls for a more coordi- nal history and concomitant alcohol and drug use bear some nated and responsive approach to supporting women through relationship with homicide generally, yet there is not clear the integration of specialist and mainstream services, includ- evidence of the associations with femicide (Bryant and ing within healthcare settings (Commonwealth of Australia Cussen 2015). 2009, 2016b). It is likely that healthcare practitioners have Despite the known risks to women, there is limited empir- early contact with victims and perpetrators and are well- ical data reporting femicide in Australia, however in placed to recognise risk; however there are insufficient data May 2014, the Counting Dead Women Australia (CDWA) that quantifies femicide in Australia and limited information campaign was launched by researchers of Destroy The Joint, available to practitioners about the women, the perpetrators following a UK model. CDWA serves as a contemporaneous and social context of femicide. Here, we provide a compre- repository of femicide data and addresses gaps in data beyond hensive review of the CDWA census from 2014, to better family and to encompass all violent deaths of understand antecedents and modifiable risk factors to raise women, regardless of perpetrator gender. Between January awareness of femicide in Australia in order to prevent 2014 and December 2016, CDWA counted 234 femicides; femicide. approximating to one woman being violently murdered every 4.7 days in Australia across that time period. Through reporting and monitoring femicide, CDWA intends to focus Methods the attention of both the public and policy makers on strategies for prevention. Design Violence against women has been described by the World Health Organisation (WHO) as a violation of human rights A cross-sectional in-depth review of femicide cases (n =81)in and a public health epidemic (World Health Organisation Australia was undertaken. The Counting Dead Women 2013); however the prevalence of violence against women Australia census commenced in 2014, however more recent remains pervasive and unacceptable in both developed and femicide cases have very limited publicly available data due to developing countries. The Australian Bureau of Statistics the lengthy time between the crime and the court case being (ABS) report that 40% of women experience interpersonal finalised. Thus we have limited the review to cases between violence across their lifetime, with one in three women 1st January and 31st December 2014. experiencing physical violence and one in five experiencing sexual violence (Australian Bureau of Statistics 2012). There Data Collection and Analysis is recognition in Australia of both the public health and eco- nomic costs of violence against women with an estimated cost A retrospective review of case data to analyse the following of 13.5 billion in 2008–09, which is primarily borne by the characteristics of femicide: 1) Victim (socio-demographic); 2) victims/survivors in terms of pain, suffering and premature Perpetrator (socio-demographics, substance use, mental death (Bugeja et al. 2015; KPMG 2009). Beyond physical health, violence and criminal history); and 3) Incident (mur- injury and premature death, women who have experienced der-suicide, additional victims, type of death, location of violence have increased rates of mental health disorders in- crime and relationship between victim and perpetrator). The cluding anxiety, depression, post-traumatic stress disorder, relationship between the victim and perpetrator was classified drug and alcohol misuse. Furthermore, the burden of femicide using relational distance categories: (1) Intimate partner (cur- can extend to corollary victims, which can include new part- rent and former partners); (2) parent-child; (3) other family; ners, children, other friends and family and first responders (4) friends and acquaintances (including carer, colleague, (Smith et al. 2014). neighbour); and (5) stranger (victim and perpetrator were not Increasingly, governments have seen the need for a public known to each other 24 h prior to incident). Aboriginal and health approach to preventing violence against women, in- Torres Strait Islander women were not able to be reliably iden- cluding the identification of underlying determinants, risk tified in the data and thereby we have not reported the propor- and protective factors (Walden and Wall 2014). The tion of femicide victims that identify as Aboriginal and/or Australian Coalition of Governments (COAG) has identified Torres Strait Islander. violence against women as a public health priority and en- All data was collected from publicly available sources with dorsed the National Plan to Reduce Violence against Women triangulation of victim, perpetrator and incident data. Initial and their Children 2009–2021 (Commonwealth of Australia data were extracted from the CDWA register of cases. These 2009). This plan calls for community action rather than plac- data were confirmed and further explored through interroga- ing the burden of responsibility on women and recognises the tion of law databases, court documents and state coronial JFamViol reports. Where necessary, the data was supplemented by print Contextual Characteristics media sources. A mixed methods approach involved sourcing and Relationship Between Victim and Perpetrator reviewing case data to identify frequently occurring anteced- ent factors, which were then incorporated into the dataset it- In cases where the relationship was known (n =59),theper- eratively with frequent consultation among the research team petrator was known to the victim in 93% of cases (55/59; to review emerging contextual factors and co-construct the Fig. 2). Of these, over half were committed by intimate part- dataset. Construction of the dataset was considered complete ners (32/59), all were male, and the location of both the crime when no new concepts emerged from the case reviews and no and death was often in the woman’s own home (47/59; additional case data could be sourced. The quantitative anal- Table 1). There were four femicides perpetrated by strangers ysis of the dataset involved descriptive statistical analysis (4/59), and none were committed in a home. (counts and percentage for nominal data), and the mean and There was greater relational distance for female perpetra- standard deviation for scale measures. Analysis was per- tors (n = 8), most of which were acquaintances or strangers formed with the Statistical Package for Social Sciences for (6/8). Seven women were murdered by their own child all of Windows version 20.0 (SPSS Inc., Chicago, Illinois) whom had a history of mental health issues; these matricides software. were most frequently perpetrated by the victim’s son (6/7) in the victim’shome(7/7).

Results Additional Victims

Characteristics of the Women There were nine additional victims across seven cases – the women’s children1 (n = 4) and/or partner (n = 5). Thirteen In 2014, 81 women were violently murdered in Australia, cases involved murder- or attempted murder suicides approximating one woman murdered every 4.5 days. The by male perpetrators, of which 12 were intimate partners women (n = 81) were aged from 20 to 82 years with a mean (Table 1). age of 44 years (SD 15.4) and median age of 45.5 years; 53% of women were (43/81) and according to the Socio- Economic Indexes for Areas (SEIFA) quintiles, more than one Antecedent Factors third lived within the most disadvantaged quintile (29/81; Fig. 1). One in five women were born outside of Australia Antecedent data (previous relationship violence, substance (17/81). Apart from residential location and age, there was use, mental health and criminal history) was unavailable in minimal information available for one in five women who many cases. In cases with accessible data, previous intimate were unnamed (17/81), however six were murdered by an partner violence and male perpetrator mental health issues intimate partner; three were mothers. were salient factors (Table 1). Almost one third of perpetrators were known to have a relevant criminal history, and this was higher for female perpetrators of whom half were known to Characteristics of the Perpetrators have a criminal history (Table 1). Previous history of violence was unavailable for 10 inti- The database contained 83 perpetrators. A group of 13 perpe- mate partner perpetrators; however among those with acces- trators was excluded from this study because they had not sible data, two thirds had a history of violent behaviour (15/ been apprehended and therefore, no details collected. Nine 22; Table 1), all but one was within the context of the relation- perpetrators were known but unnamed for legal reasons; data ship with the victim. Only two of the victims were known to from these perpetrators was included in our analysis. Thus the have a history of mental health issues; however there was final sample contained 70 perpetrators. unavailable mental health data for more than half of the wom- Known perpetrators (n = 70) were aged from 16 to 72 years en (41/81). Mental health history data was unavailable for 14 with a mean age of 40 years (SD 12.7) and a median age of intimate partner perpetrators; however among those with ac- 38 years; 89% were male (62/70) and according to the SEIFA cessible data, almost two thirds were known to have a history quintiles, approximately one quarter fell within the most dis- of mental health issues (12/18; Table 1). All seven matricide advantaged quintile (18/70; Fig. 1). The country of birth was perpetrators had a history of mental health issues (Table 1). not documented for the majority of known perpetrators (n = 43/70); however almost one quarter of known perpetrators were born outside of Australia (n =17/70). 1 In one such case the woman’s three children were additional victims JFamViol

Fig. 1 Distribution of women and 40 perpetrators across Socio- Economic Indexes for Areas 35 (SEIFA) quintiles according to place of residence 30 25

20

15 Proportion (%) 10

5

0 Quintile 1 (most Quintile 2 Quintile 3 Quintile 4 Quintile 5 (least disadvantaged) disadvantaged)

Women Perpetrators

Discussion consistent with the National Homicide Monitoring Program (NHMP), which reported the average age of victims as Eighty one women were violently murdered in Australia in 38.6 years in 2010–2012; however a victim’s age is known 2014. The majority of these preventable and premature deaths to vary according to the type of homicide e.g. victims were committed in a home, perpetrated by males who were (parents killed by children) tend to be older (Bryant and well known to the woman. Femicide victims and perpetrators Cussen 2015). were distributed across age categories, geographic and socio- Previous literature has identified that social determinants economic circumstances. A high proportion of intimate part- are risk factors for femicide depending on the relationship ner perpetrators were known to have a history of perpetrating classification (Beyer et al. 2013; Beyer et al. 2015). For ex- violence against the victim and had known mental health is- ample, a recent study in the US reported neighbourhood level sues. A greater proportion of victim-perpetrator dyads were socioeconomic disadvantage to be predictive of non-intimate located within the most disadvantaged communities, although partner femicide while disrupted neighbourhood level cohe- a considerable proportion of dyads resided in the least disad- sion (the proportion of people who were in a different house vantaged locations, challenging the notion that socioeconomic five years prior) predicted intimate partner femicide (Beyer disadvantage increases women’s vulnerability to violent et al. 2015). In terms of relationship classification, our analysis deaths. found high rates of intimate partner perpetrators, which is The age range of victims in our study challenges beliefs consistent with prior research whereby intimate partners per- that femicide more frequently impacts younger women; the petrated more femicides than all other relational categories average age was 44 years (median 45.5 years) and almost one (Campbelletal.2007; Dawson 2016;Garciaetal.2007; third of victims were aged 50 years or over. This finding is Moreschi et al. 2016;Mouzos1999). Among these studies, intimate partner femicides accounted for 50–93% of all Stranger femicides and were predominately perpetrated by male inti- 7%(n=4) mate partners. The alarmingly high proportion of femicides perpetrated by male intimate partners is observed globally in

Other: both high and low income countries (Bugeja et al. 2015; neighbour, Campbell et al. 2007;Garciaetal.2007). Moreover, while acquaintence, caretaker homicide rates have tended to decrease in high income coun- 20%(n=12) tries, the overrepresentation of women in domestic homicides Intimate partner has remained fairly stable (Bryant and Cussen 2015; Bugeja 54%(n=32) et al. 2015;Campbelletal.2007;Garciaetal.2007). Mouzos Other family 7%(n=4) Child (1999) reviewed femicide cases in Australia from 1989 to (perpetrator) 12%(n=7) 1998 and reported that three out of five femicides were perpe- trated by intimate partners of which 99% were male. Stranger perpetrators, dominated by male perpetrators, are responsible for a small proportion of femicides (Mouzos Fig. 2 Relational distance between victim and perpetrator for cases where ć the relationship was documented (n =59) 1999;Mufti and Baumann 2012), data which are consistent JFamViol

Table 1 Contextual characteristics of femicide cases Perpetrator with known perpetrators A according to the type of Intimate partner (n = 32) Female (n =8) Male (n = 62) Total (N = 70) perpetrator (intimate partner, male and female) Perpetrator mental health history Yes 12(37.5) 1(12.5) 27(43.5) 28(40.0) No 6(18.8) 3(37.5) 9(14.5) 12(17.1) Unknown 14(43.8) 4(50.0) 26(41.9) 30(42.9) Perpetrator problematic drug and alcohol use Yes 1(3.1) 3(37.5) 9(14.5) 12(17.1) No 27(84.4) 3(37.5) 45(72.6) 48(68.6) Unknown 4(12.5) 2(25.0) 8(12.9) 10(14.3) Perpetrator criminal history Yes 5(15.6) 4(50.0) 16(25.8) 20(28.6) No 15(46.9) 2(25.0) 25(40.3) 27(38.6) Unknown 12(37.5) 2(25.0) 21(33.9) 23(32.9) Perpetrator history of violence Yes 15(46.9) 0(0.0) 21(33.9) 21(30.0) No 7(21.9) 5(62.5) 19(30.6) 24(34.3) Unknown 10(31.3) 3(37.5) 22(35.5) 25(35.7) Location of crime Victims home 21(65.6) 2(25.0) 39(62.9) 41(58.6) Other home 5(15.6) 1(12.5) 7(11.3) 8(11.4) Other location 6(18.8) 5(62.5) 16(25.8) 21(30.0) Type of death Knife 8(25.0) 3(37.5) 19(30.6) 22(31.4) Gun 2(6.3) 0(0.0) 6(9.7) 6(8.6) Strangulation 4(12.5) 1(12.5) 11(17.7) 12(17.1) Other/Unknown 18(56.3) 4(50.0) 26(41.9) 30(42.9) Murder suicide Yes/Attempted 12(37.5) 0(0.0) 13(21.0) 13(18.6) No 15(46.9) 8(100.0) 40(64.5) 48(68.6) Unknown 5(15.6) 0(0.0) 9(14.5) 9(12.9) Additional victimsB Yes 1(3.1) 0(0.0) 7(11.3) 7(10.0) No 31(96.9) 8(100.0) 55(88.7) 63(90.0)

AThere were seven female perpetrators but eight victims as one perpetrator was responsible for the of two women B Additional victims were murdered by the same perpetrator in the same incident with our findings. Public misconceptions about the prevalence Our analysis showed that perpetrator gender was salient to of stranger femicides may stem from bias in media reporting, femicides that involve additional victims and murder-suicides; which tends to over-report random acts of violence against these perpetrators were primarily male and intimate partners, women perpetrated by strangers (Sutherland et al. consistent with previous reports (Muftić and Baumann 2012; 2015). In contrast to male perpetrators, we found that Salari and Sillito 2016; Smith et al. 2014). Salari and Sillito female perpetrators were more likely to be acquain- (2016) reported that intent was typically homicidal rather than tances of the victim, than intimate partners or strangers. suicidal, and a known history of intimate partner violence was This is consistent with previous research, which sug- most common in younger victim-perpetrator dyads; however gests that perpetrator gender interacts with relational substance misuse and mental health history were not salient distance such that male perpetrators are more likely than fe- antecedent factors, as we observed. The authors’ previous male perpetrators to be at the extremes of the relational dis- qualitative analysis revealed that murder-suicides preceded tance spectrum (Mouzos 1999). by a history of intimate partner violence were both JFamViol premeditated and homicidal in intention, with suicide as a remain safe rather than perpetrators to self-care. In Australia, Bsecondary decision^ (Salari and Sillito 2016). there is a strong move toward implementing recommenda- Perpetrator mental health history and previous violence tions for integration of services between frontline specialist toward a victim emerged in our analysis as salient services and mainstream healthcare (Murphy et al. 2016; antecedent factors, particularly in domestic femicides Phillips and Vandenbroek 2014). While integrated services involving matricide and intimate partner perpetrators. All provide a soft entry point for victims and perpetrators, it is matricide perpetrators had significant mental health histories, critical that healthcare practitioners are trained to identify and among intimate partner perpetrators, the majority had a signs of violence and are sufficiently supported to provide this history of violence against the victim. Moreover, the data type of care (Plat-Jones 2006). Indeed, there is evidence that frequently depicted relationships that were beset with practitioners may be uncomfortable or may feel they do not unrelenting torment and unsuccessful attempts by the victim have the necessary skills or competency to provide this inte- to seek safety and/or legal intervention. Although Campbell grated model of care (Plat-Jones 2006; Ramsay et al. 2012). et al. (2007) reported that intimate partner femicide is preced- Thereby, practitioner perspectives and consultation are essen- ed by violence in 65–70% of cases, with prior partner violence tial to understand the barriers and facilitators to integrating as the major risk factor for femicide, there is a lack of consen- services for victims and perpetrators of violence within main- sus regarding how intimate partner violence culminates in stream healthcare. While this study offers insight into anteced- femicide. One theory is that violence within the relationship ent factors, there is a need to understand the progression of escalates, possibly in response to stressors for example preg- non-fatal violence, the relationship with mental health and the nancy, estrangement or substance misuse (McPhedran and potential points of intervention. Accordingly, the present re- Baker 2012). An alternative theory posits that specific perpe- sults provide impetus for qualitative research with perpetrators trator typologies exist that are characterised by a pervasive and proxies for femicide victims to better understand the so- pattern of anti-social and violent behaviour that is not neces- cial context of femicide. Similarly, women who have experi- sarily specific to their relationship with the victim (McPhedran enced near-fatal violence and the perpetrators of such violence and Baker 2012). While our data does not distinguish between may provide important insight into risk and protective factors. these hypotheses, there is support for the latter with evidence This study should be considered within the context of sev- of multiple forms of perpetrator anti-social behaviour includ- eral limitations. The primary limitation is missing data, which ing stalking, controlling, jealousy and violence toward others. was most frequently related to cases with unknown perpetra- Overall, high rates of antecedent family violence, mental tors and cases that were pending, in progress or not recorded health issues, anti-social behaviour and prior contact with the in legal databases. Where possible we supplemented legal data justice system represent missed opportunities for intervention with media sources; this methodology is used by the National targeting the victim, perpetrator or dyad. Cases in our sample Homicide Monitoring Program (Bryant and Cussen 2015). revealed multiple points of contact for both victims and per- This limitation of the present study serves to highlight the petrators including within the health system; this has similarly challenges in collating femicide data and is indicative of the been described in prior research with up to 40% of femicide scant empirical evidence that investigates the context of victims seeking medical help for injury, physical or mental femicide; consequently in many countries, particularly in health in the year prior to their murder (Sharps et al. 2001). low resource settings, the incidence of femicide is not moni- Murphy et al. (2016) reviewed intimate partner homicides for tored or is not readily accessible (Bugeja et al. 2013; the period 2000–2008 in Victoria, Australia, and found that McPhedran and Baker 2012; Weil 2016). A further limitation three quarters of victims and/or perpetrators had contact with of the present study is small number of cases upon which this either the justice or health system in the 12 months prior to the review is based as we have limited the review to cases from femicide, with the health system frequently the first and pre- 2014, which was the first year that the CDWA recorded ferred contact. Perpetrators were more likely than victims to femicides in Australia. This period was selected due to the have contact and the majority of contacts were within one lengthy time between the crime and the court case being month of the femicide. This indicates that both victims and finalised; therefore more recent femicide cases have very lim- perpetrators are seeking help, particularly from the health sys- ited publicly available data. We recommend that future re- tem, which represents an important opportunity for search spanning a longer period of time will address this and intervention. provide greater insight into the context of femicide. As part of a multi-sectoral approach to violence against While the National Homicide Monitoring Program collates women, health practitioners can provide a non-threatening femicide data in Australia, McPhedran and Baker (2012)de- point of contact and are well-placed to identify risk and assess scribe specific data challenges that need to be addressed to safety for both victims and perpetrators. It is critical that per- better understand the context of gendered violence and the petrators be supported to seek help as this shifts the burden of specific pathways from non-fatal violence to fatal violence. responsibility, which is inexplicably placed upon women to This includes a more consistent and coordinated approach to JFamViol data collection and dissemination, with implementation of Bugeja, L., Dawson, M., McIntyre, S.-J., & Walsh, C. (2015). Domestic/ well-understood measures of partner and family violence. Family Violence Death Reviews: An International Comparison. Trauma, Violence, & Abuse, 16(2), 179–187. https://doi.org/10. 1177/1524838013517561. Campbell, J. C., Glass, N., Sharps, P. W., Laughon, K., & Bloom, T. Conclusion (2007). Intimate Partner Homicide: Review and Implications of Research and Policy. Trauma, Violence, & Abuse, 8(3), 246–269. https://doi.org/10.1177/1524838007303505. The present study has emphasised that women are most at risk Commonwealth of Australia. (2009). Time for Action: The National of fatal violence in their own home, at the hands of a perpe- Council’s Plan for Australia to Reduce Violence against Women trator who is well known to them, likely male, and with a and their Children, 2009–2021 (ISBN9781921380310). history of perpetrating violence; however there is insufficient Commonwealth of Australia. (2016a). COAG Advisory Panel on – insight into the complex social dynamic that sees non-fatal Reducing Violence against Women and their Children Final Report (978-1-925238-35-8). In Department of the Prime Minister violence progress to femicide. Indeed risk factors for femicide and Cabinet. vary by the type of homicide, and expanded data collection is Commonwealth of Australia. (2016b). Third Action Plan 2016–2019 of urgently required to understand the inherent complexities such the National Plan to Reduce Violence against Women and their – – – that targeted strategies can be implemented as part of a com- Children 2010 2022. In ISBN: 978 1 925318-47-0. Dawson, M. (2016). Punishing femicide: Criminal justice responses to bined justice and public health approach. This is critical to the killing of women over four decades. Current Sociology, 64(7), implementing recommendations for integration of services 996–1016. https://doi.org/10.1177/0011392115611192. within mainstream healthcare to address the burden of vio- Du Plat-Jones, J. (2006). : the role of health profes- lence against women. Furthermore, effective integration of sionals. Nursing Standard, 21(14), 44–48. https://doi.org/10.7748/ ns2006.12.21.14.44.c6392. services requires extensive consultation between the justice Garcia, L., Soria, C., & Hurwitz, E. L. (2007). Homicides and Intimate and health sector to ensure that practitioners have the re- Partner Violence: A Literature Review. Trauma, Violence, & Abuse, sources and competency to provide support within healthcare 8(4), 370–383. https://doi.org/10.1177/1524838007307294. settings. KPMG. (2009). The cost of violence against women and their children: Report commissioned by the National Council to Reduce Violence against Women and their Children. Retrieved from Canberra: https:// Acknowledgements We acknowledge the immense and important work www.dss.gov.au/sites/default/files/documents/05_2012/vawc_ of Pat Bradley, Sue McLeod and Debra Smith and the Counting Dead economic_report.pdf Women Australia researchers of Destroy The Joint. The research team also acknowledges the contribution of Sanela Begani who assisted with McPhedran, S., & Baker, J. (2012). Lethal and Non-Lethal Violence data collection. Against Women in Australia: Measurement Challenges, Conceptual Frameworks, and Limitations in Knowledge. Violence Against Women, 18(8), 958–972. https://doi.org/10.1177/ Compliance with Ethical Standards 1077801212456755. Moreschi, C., Da Broi, U., Zamai, V., & Palese, F. (2016). Medico legal Declarations of Interest None. and epidemiological aspects of femicide in a judicial district of north eastern Italy. Journal of Forensic and Legal Medicine, 39,65–73. https://doi.org/10.1016/j.jflm.2016.01.017. References Mouzos, J. (1999). Femicide: The Killing of Women in Australia 1989– 1998. Retrieved from Muftić, L. R., & Baumann, M. L. (2012). Female Versus Male Australian Bureau of Statistics. (2012). Personal Safety Survey.(ABS Perpetrated Femicide: An Exploratory Analysis of Whether Cat. No. 4906.0). In Canberra Retrieved from http://www.abs.gov. Offender Gender Matters. Journal of Interpersonal Violence, au/ausstats/[email protected]/Lookup/4906.0Chapter2002012. 27(14), 2824–2844. https://doi.org/10.1177/0886260512438282. Beyer, K. M. M., Layde, P. M., Hamberger, L. K., & Laud, P. W. (2013). Murphy, B., Liddell, M., & Bugeja, L. (2016). Service Contacts Characteristics of the residential neighborhood environment differ- Proximate to Intimate Partner Homicides in Victoria. Journal of entiate intimate partner femicide in urban versus rural settings. The Family Violence, 31(1), 39–48. https://doi.org/10.1007/s10896- Journal Of Rural Health: Official Journal Of The American Rural 015-9738-2. Health Association And The National Rural Health Care Petrosky, E., Blair, J., Betz, C., Fowler, K. A., Jack, S., & Lyons, B. Association, 29(3), 281–293. https://doi.org/10.1111/j.1748-0361. (2017). Racial and Ethnic Differences in Homicides of Adult 2012.00448.x. Women and the Role of Intimate Partner Violence — United Beyer, K. M. M., Layde, P. M., Hamberger, L. K., & Laud, P. W. (2015). States, 2003–2014. MMWR. Morbidity and Mortality Weekly Does Neighborhood Environment Differentiate Intimate Partner Report, 66,741–746. https://doi.org/10.15585/mmwr.mm6628a1. Femicides From Other Femicides? Violence Against Women, Phillips, J., & Vandenbroek, P. (2014). Domestic, family and sexual vio- 21(1), 49–64. https://doi.org/10.1177/1077801214564075. lence in Australia: an overview of the issues.Retrievedfromhttp:// Bryant, W., & Cussen, T. (2015). Homicide in Australia: 2010–11 to parlinfo.aph.gov.au/parlInfo/download/library/prspub/3447585/ 2011–12: National Homicide Monitoring Program report. (23). upload_binary/3447585.pdf;fileType=application/pdf Canberra: Australian Institute of Criminology. Plat-Jones J, D. (2006). Domestic violence: the role of health profes- Bugeja, L., Butler, A., Buxton, E., Ehrat, H., Hayes, M., McIntyre, S.-J., sionals. Nursing Standard, 21(14 –16), 44–48. & Walsh, C. (2013). The Implementation of Domestic Violence Ramsay, J., Rutterford, C., Gregory, A., Dunne, D., Eldridge, S., Sharp, Death Reviews in Australia. Homicide Studies, 17(4), 353–374. D., & Feder, G. (2012). Domestic violence: knowledge, attitudes, https://doi.org/10.1177/1088767913494787. and clinical practice of selected UK primary healthcare clinicians. JFamViol

British Journal of General Practice. https://doi.org/10.3399/ women and their children: State of knowledge paper. Retrieved from bjgp12X654623. Sydney : Australia’s National Research Organisation for Women’s Salari, S., & Sillito, C. L. (2016). Intimate partner homicide–suicide: Safety Limited (ANROWS): Perpetrator primary intent across young, middle, and elder adult Walden, I., & Wall, L. (2014). Reflecting on primary prevention of vio- age categories. Aggression and Violent Behavior, 26,26–34. lence against women: The public health approach (1833–7864). https://doi.org/10.1016/j.avb.2015.11.004. Retrieved from Melbourne: https://aifs.gov.au/publications/ Sharps, P. W., Koziol-McLain, J., Campbell, J., McFarlane, J., Sachs, C., reflecting-primary-prevention-violence-against-women/export & Xu, X. (2001). Health Care Providers' Missed Opportunities for Weil, S. (2016). Making femicide visible. Current Sociology, 64(7), Preventing Femicide. Preventative Medicine, 33,373–380. 1124–1137. https://doi.org/10.1177/0011392115623602. Smith, S. G., Fowler, K. A., & Niolon, P. H. (2014). Intimate Partner World Health Organisation. (2013). Global and regional estimates of Homicide and Corollary Victims in 16 States: National Violent violence against women: prevalence and health effects of intimate – Death Reporting System, 2003 2009. American Journal of Public partner violence and non-partner sexual violence. Switzerland: – Health, 104(3), 461 466. https://doi.org/10.2105/ajph.2013. Retrieved from Geneva http://www.who.int/reproductivehealth/ 301582. publications/violence/9789241564625/en/. Sutherland, G., McCormack, A., Pirkis, J., Easteal, P., Holland, K., & Vaughan, C. (2015). Media representations of violence against