Trends & issues in crime and criminal justice ISSN 0817-8542    No. 568 February 2019

International and Australian research indicates that children aged under five years make up the largest proportion of filicide victims (Bourget & Gagné 2005; Brown, Tyson & Fernandez Arias 2014; Dixon, Krienert & Walsh 2014; Kunz & Bahr 1996; Mariano, Chan & Myers 2014; Mouzos & Rushforth 2003; West, Friedman & Resnick 2009). Infants are at particular risk—between one quarter and one-third of all filicide victims are under one year of age (Bourget & Gagné 2005; Dixon, Krienert & Walsh 2014; Kunz & Bahr 1996; Mariano, Chan & Myers 2014). Filicide, however, involves not only child victims. While the incidence of adult filicide victimisation is less well documented, one study from the US estimated 18 percent of filicide victims over a 32-year period were aged 18 years or over (Mariano, Chan & Myers 2014).

Early research on filicide, pioneered by Resnick (1970, 1969), was dominated by an examination of the motives of filicide offenders. Motive continues to be a common theme in filicide research, yet some scholars argue that categorising offenders according to motive is simplistic and misleading (Sidebotham 2013; Stroud 2013).  Motives are rarely expressed, or expressed clearly, by filicidal parents (Brown, Tyson & Fernandez Arias 2014; Mouzos & Rushforth 2003) and its emphasis disregards the personal, familial and social circumstances that may lead to or characterise a filicide. 

More recent filicide research has therefore examined the effects of some of these circumstances, such as -

The Monash Filicide Research Project investigated filicide deaths that occurred in Victoria between 2000 and 2009 (Brown, Tyson & Fernandez Arias 2014). Fifty-two children were identified as being killed in the state over the 10-year period. Information on 42 of the 52 children was available from closed case files held by the Victorian Coroner’s Office, including witness statements, police records, court transcripts and criminal records for the offender.

In this study 16 children were killed by their mother, 15 by their father, nine by a step-father and one by both parents. Of the 36 filicides in which risk factor information was available, the majority followed parental separation (72%; n=26) and involved an offender with a mental illness (66%; n=24).

A history of domestic violence or child abuse characterised a quarter (25%; n=9) and a fifth (22%; n=8) of filicides respectively.

This study drew on the Monash Filicide Research Project to produce a picture of filicide across Australia. The data used for analysis came from the Australian Institute of Criminology’s National

Homicide Monitoring Program (NHMP). The NHMP collects data on homicide incidents, victims and offenders recorded in Australia since 1989–0.

NHMP data are derived from (a) offence records created by Australian state and territory police and (b) state coronial records (including toxicology and post-mortem reports). State and territory police provide homicide incident, victim and offender data to the NHMP using a data collection form developed by the AIC in consultation with police agencies. Coronial data are accessed from the National Coronial Information System using victim name, jurisdiction and incident year identifiers drawn from police data. Autopsy, toxicology and coronial findings reports are used to verify incident and victim data provided by police, such as cause of death.

Court transcripts and media reports are used where suitable to supplement offence records and coronial data sources. NHMP data are linked to these sources using victim and offender name(s) provided in offence records.

The analytical framework for the study was guided by the findings from the Monash Filicide Research Project (described above) and constructed based on the range of descriptive variables captured in the NHMP.

Data were analysed as incident, victim and offender characteristics, and victim relationship with the primary offender.

Between 2000–01 and 2011–12, Australian state and territory police recorded 238 incidents of filicide, in which 284 victims were killed by their parent(s). Ninety-six percent of filicide victims (n=274) were aged 0–17 years. Of the 10 adult victims, the youngest was 18 years and the oldest 33 years.

The 238 filicide incidents were committed by 260 offenders. This group of homicide offenders accounted for 16 percent of all domestic homicide offenders (n=1,484) and six percent of all identified homicide offenders (n=3,771) over the 12-year period. Eighty-three percent of filicide offenders (n=216) committed the homicide on their own. The remaining offenders committed the filicide with another person—that is, 22 homicide incidents involved two co-offenders.

The primary offender rate fluctuated during the period, with the highest rate—0.2 offenders per 100,000—recorded in 2001–02 (see Figure 1). Offender rates were similarly variable across selected states (see Figure 1), largely affected by the small number of filicide offenders recorded in each jurisdiction each year.

Note: Excludes Tas, NT and the ACT due to very small numbers. Offender rates for all jurisdictions are shown in Table A1

Source: AIC NHMP 2000–01 to 2011–12 [computer file]; Table A1

Males and females comprised around half of primary filicide offenders each (52% (n=124) vs 48% (n=114); see Table 1). However, male offender rates were higher than female offender rates for all but four of the years from 2000–01 and 2011–12 (see Figure 2). Some variation in the proportion of male and female filicide offenders was observed across the jurisdictions. Higher proportions of male filicide offenders were recorded in Victoria (62%; n=29) and Western Australia (60%; n=12), as well as in Tasmania and the Northern Territory, although based on very small offender numbers.

In contrast, three-quarters of filicide offenders in South Australia were female (76%; n=19).  The reason(s) for the predominance of female filicide offenders in South Australia is unclear.

Source: AIC NHMP 2000–01 to 2011–12 [computer file]; Table A2

Tas 2 100 0 0 2

Note: Table refers to primary filicide offenders only. Percentages may not total 100 due to rounding

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

The median age of filicide offenders was 32 years; the youngest offender was 17 years of age and the oldest 75 years. The largest proportion of female filicide offenders were aged 25–34 years (42%; n=47). Another third were aged 35–49 years (34%; n=38). Similar proportions of male offenders were aged 25–34 (37%; n=45) and 35–49 years (32%; n=39). There was no statistically significant difference in the age of male and female filicide offenders.

Nine in 10 primary filicide offenders were non-Indigenous (89%; n=217). The non-Indigenous offending rate ranged from 0.04 to 0.19 per 100,000 between 2000–01 and 2011–12 (see Figure 3). The Indigenous offending rate was up to six times the non-Indigenous offending rate depending on the year considered but fluctuated markedly due to the small numbers of Indigenous filicide offenders recorded each year (range=0–3). Overall, 21 of the 238 primary filicide offenders in 2001–02 to 2011–12 were Indigenous. Approximately equal proportions of non-Indigenous offenders were male and female, whereas nearly two-thirds of Indigenous filicide offenders were male (62%; n=13).

Source: AIC NHMP 2000–01 to 2011–12 [computer file]; Table A3

Divorced 2 2 2 2 4 2

a: Registered marriage

b: Includes same-sex relationships

c: Includes separations from registered and de facto marriages

Note: Excludes 15 offenders for whom marital status was unknown. Percentages may not total 100 due to rounding

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

Most filicide offenders did not attempt (93%; n=200) or commit suicide (81%; n=190) following the homicide incident. Among the 14 primary offenders who attempted suicide, 12 were female.  Of those who committed suicide, 53 percent (n=24) were male and 47 percent (n=21) were female.

Because the relationship recorded in the NHMP is the victim’s relationship to the offender, the following data are presented as victim rather than offender counts. Therefore, in homicides where an offender killed two or more of their children, the offender will be counted multiple times. The number of single victim and multiple victim incidents by parental status is shown in Table 3. The majority (n=201; 84%) were single victim incidents.

Note: percentages may not total 100 due to rounding

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

Three-quarters (76%; n=215) of the 284 filicide victims between 2000–01 and 2011–12 were killed by a custodial parent—46 percent (n=133) by their custodial mother and 29 percent (n=81) by their custodial father (see Table 4). Smaller proportions of children were killed by a step-parent (14%; n=41) or non-custodial parent (10%; n=28). All but one of the victims of a non-custodial parent were killed by their father and all victims of step-parents were killed by a step-father.

More males than females were killed by their parent or parent equivalent—56 percent (n=158) vs 44 percent (n=125). Most children killed by their father, irrespective of the custodial arrangements, were male (see Table 4). Males comprised 54 percent (n=44) of children killed by a custodial father, 63 percent (n=17) of those killed by a non-custodial father and 66 percent (n=27) of those killed by a step-father. Just over half of children killed by a custodial mother were male (52%; n=69).

a: Excludes 1 case where victim gender was unknown

Note: Percentages may not total 100 due to rounding

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

The median age of filicide victims was two years. Two-thirds of victims were under five years old when they were killed—31 percent (n=90) under 12 months and 35 percent (n=100) aged one to four years. Two-thirds of children (n=141) killed by a custodial parent were 0–4 years (see Table 4).

Custodial mothers killed similar proportions of children under one year and children aged one to four years (35% (n=46) vs 33% (n=44)). Almost four in 10 children killed by their custodial father were less than one year (37%; n=30) and a quarter were aged one to four years (26%; n=21).

The majority of children killed by a step-father were one to four years (61%; n=25). Children killed by a non-custodial father tended to be older. A third were five to nine years (33%; n=9) and another third one to four years (33%; n=9).

Children killed by a parent who then attempted or committed suicide were most commonly killed by a custodial parent (n=13 (93%) for attempted suicide and n=37 (86%) for suicide). Offenders who did not suicide were most likely to have killed children aged under one year (n=78; 41%), whereas offenders who did suicide were most likely to have killed children aged one to four years (n=18; 40%).

Over half of children killed by a step-father were beaten (54%; n=21) and a quarter died from being shaken (18%; n=7; see Table 5). In contrast, a quarter of children killed by a non-custodial father were strangled or suffocated (26%; n=7) and a fifth were drowned (22%; n=6). The largest proportion of children killed by a custodial mother also died from strangulation or suffocation (20%; n=25). Beating was the most common method of killing used by custodial fathers (32%; n=26), followed by stab wounds (16%; n=13).

Drowning/

Strangulation/

Shaken baby

Poisoning/

Note: Excludes 12 cases where cause of death was unknown. Percentages may not total 100 due to rounding

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

An average of 20 filicide incidents was recorded each year in Australia between 2000–01 and 2011–12. The incidence remained steady over this period, unlike rates of domestic homicide and homicide overall, which declined by 13 percent and 20 percent respectively (Brown et al. 2019). Few studies have reported national filicide statistics, and data that are available are not comparable to

Australian estimates. Homicide data collected in England and Wales, Canada and the United States

provide a useful reference point for national incident and victimisation rates, but filicide rates are

based on different victim age populations. Similar issues affect the only study comparing rates of child

homicide across multiple countries (Pritchard, Davey & Williams 2013).

The characteristics of Australian filicide offenders, however, can be compared with those of filicide

offenders identified in other industrialised Western countries and are generally comparable. There

was a predominance of male offenders (if parent and parent equivalents are counted together), and

a relationship between offender gender and victim age. A history of domestic violence involving

the offender was also prevalent. Parental separation, however, did not feature as a common factor

among filicidal parents, and rates of mental illness among female filicide offenders, compared with

male filicide offenders, were higher than those reported in the international literature.

Most offenders were custodial parents, with the largest number of children killed by a custodial

mother. Mothers were slightly more likely to kill a male child while fathers (custodial, non-custodial or

step-fathers) mostly killed male children. Further, gender and custodial relationship were associated

with different methods of killing. Beating deaths were associated with custodial fathers and stepfathers,

while strangulation and suffocation deaths were associated with custodial mothers.

One factor of increasing interest in international research (Pritchard, Davey & Williams 2013;

Sidebotham et al. 2016) is the criminal history of offenders, which was examined in this study.

Forty-three percent (n=83) of filicide offenders had a criminal history. Male offenders were twice

as likely as female offenders to have a previous conviction, and step-fathers almost twice as likely

to have a criminal history compared with custodial fathers. The criminal histories of female filicide

offenders in Australia, which largely involved drug offending but with some violent offending, provide

important context on previous offending among filicide offenders that has not been found in other

studies (Pritchard, Davey & Williams 2013). Comparable national data on criminal histories of all

male and female offenders are not publicly available. However, the ratio of male to female offenders

for the years where national offender data are available (ie 2007–08 to 2011–12; ABS 2018) was approximately four to one.

The results in this report should be interpreted in light of the following limitations.

each individual victim’s relationship with the offender is counted separately. Therefore, in

homicides where an offender was responsible for the death of two or more of their children,

the offender was counted multiple times.

on the primary offender. For filicide incidents involving a single victim and offender,

identifying the primary offender is straightforward. However, where there are multiple

victims and/or offenders involved in a filicide incident, identifying the primary offender

becomes more complicated. The primary offender is hence determined based on the closest

known relationship between any victim–offender pairing (known as the principal/primary

relationship). Identifying the closest relationship can be difficult, as different relationships may

exist between different pairs of victims and offenders. Further, there may be some instances

where, for example, a step-parent is more culpable than a custodial parent in the homicide

incident; however, the custodial parent is classified as the primary offender because they

had a closer relationship with the victim. This is because the NHMP dataset does not indicate

offender culpability where multiple offenders were involved in the one incident.

violence between intimate partners and is not indicative of a victim dying in the context of a

family or domestic violence incident.

judgement of the victim and/or offender or by incapacitating the victim in some way

(intentionally or not). The NHMP dataset identifies the presence of alcohol and other drugs but

not the impact of their use on the victim, offender or situation. For offenders, this information

is based on police observations.

At the same time, the study does assist professionals across a variety of services by identifying that offender circumstances and characteristics differ according to the offender’s relationship with the victim. The problems associated with each of the offender types (custodial parents, non-custodial parents and step-parents) are likely to require different service responses. For example, step-fathers with a history of prior offending are likely to be engaged with criminal justice services. Therefore, a range of services that potential offenders may be involved with extends beyond the traditional child welfare services and, in particular, beyond the statutory child protection services. The services with which they are involved may not be focused on offering therapeutic intervention. Nevertheless, it is important that adult service providers remain alert to the potential dangers to children. Thus, mental health services, general health services, family law courts, criminal justice services, post-separation services and domestic violence services should continue to adopt interventions that consider the safety of their clients’ children.

Complex service networks are not perfect mechanisms and, although many improvements have been made over decades to inter-professional communication and coordinated service delivery, areas identified for improvement and the need to trial a range of practice approaches remain. Although the study did not seek to develop intervention policies and programs, it provides information that will be used in future research to fashion more precise proposals for intervention.

ABS. http://www.abs.gov.au/ausstats/abs@.nsf/mf/4519.0

Australian Institute of Criminology

Brown T & Tyson D 2012. An abominable crime: Filicide in the context of parental separation and divorce.

publications/sr/sr002

Butler A & Buxton E 2013. Trends and issues in filicide deaths: A review of all NSW filicide deaths in the context

of domestic violence. Addressing Filicide: Inaugural International Conference for Cross National Dialogue, Prato,

Italy, 29–30 May

Daly M & Wilson M 1994. Some differential attributes of lethal assaults on small children by stepfathers versus

Dixon S, Krienert JL & Walsh J 2014. Filicide: A gendered profile of offender, victim, and event characteristics in a

Eriksson L, Mazerolle P, Wortley R & Johnson H 2014. Maternal and paternal filicide: Case studies from the

Frederico M, Jackson A & Dwyer J 2014. Child protection and cross-sector practice: An analysis of child death

Friedman SH, Hrouda DR, Holden CE, Noffsinger SG & Resnick PJ 2005. Filicide-suicide: Common factors in

Johnson CH 2008. Intimate partner homicide and familicide in Western Australia. Keynote Paper. Australian

Institute of Criminology. International Conference on Homicide. Surfers Paradise, Queensland, 3–5 Dec

Paper no. 8. Melbourne: Domestic Violence Resource Centre Victoria. http://www.dvrcv.org.au/knowledgecentre/

our-publications/discussion-papers/just-say-goodbye

Mariano TY, Chan HCO & Myers WC 2014. Towards a more holistic understanding of filicide: A multidisciplinary

255. Canberra: Australian Institute of Criminology. https://aic.gov.au/publications/tandi/tandi255

Putkonen H, Amon S, Eronen M, Klier CM, Almiron MP, Cederwall JY & Weizmann-Henelius G 2010. Child

murder and gender differences: A nationwide register-based study of filicide offenders in two European

126(10): 1414–1420

University of Sydney & Family Court of Australia. http://www.familycourt.gov.au/wps/wcm/connect/fcoaweb/

reports-and-publications/reports/

Sidebotham P, Brandon M, Bailey S, Belderson P, Dodsworth J, Garstang J, Harrison E, Retzer A & Sorensen P

Final report May 2016. London: UK Department for Education. https://www.gov.uk/government/publications/

analysis-of-serious-case-reviews-2011-to-2014

Stroud J 2013. Responding to and understanding filicide: Challenges and psychosocial perspectives. Addressing

Filicide: Inaugural International Conference for Cross National Dialogue, Prato, Italy, 29–30 May

Stroud J & Pritchard C 2001. Child homicide, psychiatric disorder and dangerousness: A review and an empirical

Weekes-Shackelford VA & Shackelford TK 2004. Methods of filicide: Stepparents and genetic parents kill

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

2000–01 0.10 0.10 0.10

2009–10 0.08 0.09 0.09

2010–11 0.12 0.11 0.11

Source: AIC NHMP 2000–01 to 2011–12 [computer file]

General editor, Trends & issues in crime and criminal justice series: Dr Rick Brown, Deputy

Director, Australian Institute of Criminology. Note: Trends & issues in crime and criminal justice

ISSN 0817-8542

©Australian Institute of Criminology 2019

GPO Box 1936

Canberra ACT 2601, Australia

Tel: 02 6268 7166

Trends & issues in crime and criminal justice

Australian Institute of Criminology

aic.gov.au

2002–03 0.32 0.16

2003–04 0.31 0.15

2005–06 0.87 0.16

2007–08 0.58 0.13

2011–12 0.24 0.10