Family violence (FV) use by young people is increasingly recognised by clinicians, law enforcement, and acadmics as a significant public health and social problem that remains largely unreported (Fitz-Gibbon et al., 2018; Kuay & Towl, 2021). Youth FV is a broad term which involves abuse by young people aged 10–24 years toward relatives (e.g. parents, siblings, other relatives) and intimate partners, and includes both physical (e.g. physical assault, sexual assault, etc.) and non-physical (e.g. psychological abuse) behaviour.

Focusing research on young people who engage in FV behaviour, including how this behaviour differs in adolescence and young adulthood, can provide important cues for tailoring interventions to meet the needs of these young people and reduce further violence. The trajectory from early adolescence to young adulthood is a time of profound developmental change characterised by an increased engagement in risk-taking and antisocial behaviour, decreasing level of parental involvement, greater reliance on peers and intimate relationships, and elevated levels of mental health issues (Arnett, 2000; Johnson et al., 2015; Sawyer et al., 2018; Scott et al., 2016; Snyder & McCurley, 2008). Such changing needs and priorities inevitably have the potential to place stress upon the familial system and broader social context, potentially contributing to changes in FV risk with age.

Similarly, there is a small body of literature suggesting that the relative importance of certain dynamic risk factors/criminogenic needs (i.e., those factors known to be related to offending behaviour, such as substance abuse and unemployment/school truancy) to recidivism varies at different stages of adolescent development (van der Put et al., 2011; Put et al., 2012). This is known as the “age-risk factor paradox” (van der Put et al., 2011, p. 258) and is drawn from the general offending literature. The paradox suggests dynamic risk factors are least prevalent, but most predictive of, recidivism among those aged under 14 years old. In contrast, among those 14 years and over, there is a higher prevalence of dynamic risk factors, but they are less strongly predictive of recidivism (van der Put et al., 2011). For example, substance use is more prevalent among older adolescents who engage in offending behaviour than those in early adolescence (van der Put et al., 2011). However, there is currently a lack of understanding as to how such needs may differ between adolescent and young adult individuals, including for young people who use FV. Services within Australia and internationally are increasingly providing support for young people up to 25 years old given neuroscientific and psychosocial evidence of ongoing brain maturation into the third decade of life (Cohen et al., 2016; Scott et al., 2016), indicating a need for research to consider this young adult cohort alongside their adolescent counterparts.

Differences in the Characteristics of Youth who Use Family Violence by Age

There is limited research examining age-based variations in the sociodemographic, psychosocial, and FV-related characteristics of young people who come to police attention for engaging in FV. However, a small body of literature on the topic does exist.

Young people who come to police attention for FV are likely to be male and to target a female victims (Freeman, 2018; Phillips & McGuinness, 2020; Snyder & McCurley, 2008). There is some evidence among the general youth offending literature that intellectual disabilities are more prevalent among adolescents compared to adults (Richards, 2011), however no such research is available for samples of young people who use FV. Similarly, there is an absence of research comparing age-related variations in FV behaviour and offending according to socioeconomic status (SES).

The prevalence of psychosocial characteristics – including mental health issues, substance abuse, and education/employment issues – differ with age. Adolescent FV-users have been shown to experience more mental health problems – but less substance abuse issues – than adult FV-users (Phillips & McGuinness, 2020). Research from the general youth offending literature shows problems with education increase from early to late adolescence (van der Put et al., 2011; Put et al., 2012), while early adolescent offenders may display the highest rates of childhood victimisation (Jolliffe et al., 2017). There is also age-related variation in the FV characteristics of victims and young people. While victims of youth FV are disproportionately female regardless of the young person’s age (Boxall & Sabol., 2021; Phillips & McGuinness, 2020; Snyder & McCurley, 2008), child-to-parent abuse has been identified as most common among adolescent FV-users, while intimate partner abuse is most prevalent among young adults (Phillips & McGuinness, 2020; Snyder & McCurley, 2008).

Family Violence Recidivism

Compared to the adult FV literature, there is a distinct lack of research examining variation in recidivism rates among adolescent and young adult FV-users. Based on the limited available literature, FV recidivism rates for those under 24 years old appear to vary between 20.8% and 58% (Boxall & Morgan, 2020; Boxall & Sabol, 2021; Pooley et al., 2021; Sheed et al., 2022; Spivak et al., 2021). Pooley et al. (2021) found over half of all young people (58%) aged 13–17 years had been reported to police for a further FV incident by 23 years old, while the 6-month recidivism rate for youth FV has been shown to vary between 20.8% and 28% (Boxall & Morgan, 2020; Spivak et al., 2021). In their examination of time to FV recidivism, Boxall and Morgan (2020) found the highest risk period for FV recidivism (among 12-18-year-olds) occurred approximately one month after the index incident, with the probability of a repeat FV event declining sharply after this period. Rates of youth FV recidivism are broadly consistent with the adult literature, in which the 6-month FV recidivism is approximately 23% (Hulme et al., 2019; Spivak et al., 2021), while studies with longer periods of follow-up (i.e., approximately four to five years) found 32–51% of individuals were reported for another FV incident (Hilton & Radatz, 2021; Hulme et al., 2019). Comparing FV recidivism rates of young people (Boxall & Morgan, 2020) and adults (Morgan et al., 2018), it appears youth reoffend more quickly within the first month following the index incident than their adult counterparts.

The Present Study

This study represents the first to examine how characteristics of young FV-users differ between early adolescence, late adolescence, and young adulthood using a population-based sample. It aimed to: (1) provide a descriptive overview of a population cohort of young people (aged 10–24 years) who came to police attention for using FV; (2) determine whether there were age-related differences in the sociodemographic, psychosocial, and FV-related characteristics of young people at their index FV incident (i.e., the incident leading to inclusion in the study), and (3) identify whether there were age-related differences in time to FV recidivism. The three age groups were developed based on the key developmental periods of early adolescence (10–14 years), late adolescence (15–19 years), and young adulthood (20–24 years) identified by the World Health Organisation (WHO, 2021).

It is expected that most FV-users will be male and most victims will be female (Freeman, 2018; Simmons et al., 2018; Snyder & McCurley, 2008). The proportion of male FV-users, and the prevalence of substance abuse and intimate partner abuse, is anticipated to be greatest among those in late adolescence and young adulthood (Snyder & McCurley, 2008; Spruit et al., 2017; van der Put et al., 2011). Similarly, child-to-parent abuse is hypothesised to be higher among early- and late-adolescent FV-users, compared with those in young adulthood (Snyder & McCurley; Phillips & McGuinness, 2020). There is insufficient prior research examining age variations in the other sociodemographic, psychosocial, and FV-related characteristics – and time to FV recidivism between age groups – from which to make other informed hypotheses.

Methodology

The study used a pseudo-prospective follow-up design employing administrative data from Victoria Police databases. Victoria Police are the sole policing agency for the Australian state of Victoria (population 6.63 million at the time of the study; 67% of whom live in the state’s capital city of Melbourne; Australian Bureau of Statistics, 2021) and record all reported incidents of FV as a matter of policy, regardless of whether criminal charges were laid. Not all forms of FV identified under the Act constitute a criminal offence (e.g. no specific charges are associated with psychological abuse or coercion in Victoria at the time of publication). In Victoria, only half (N = 47,468, 50.8%) of all FV incidents reported to the Police between July 2020 and June 2021 involved a criminal offence for which charges were laid (Crime Statistics Agency, 2021).

Definitions

The present study uses the terms young person who uses family violence and family violence-user (FV-user) in recognition of the need to consider young people as more than their behaviour, and to encourage a person-centred approach to conceptualisation of youth FV. Family violence was defined in the present study according to Victoria’s Family Violence Protection Act 2008. The Act identifies FV as involving abuse toward relatives (e.g. parents, siblings, children, other relatives, carers) and intimate partners, and includes both physical (e.g. physical assault, sexual assault, etc.) and non-physical (e.g. coercion, threats) behaviour, and which causes the family member to fear for their – or another’s – safety and wellbeing. The behaviour of a FV-user does not need to constitute a criminal offence to be considered FV.

The developmental periods of interest to the present study were derived from the World Health Organisation (WHO, 2021). The WHO defines ‘young people’ as those aged 10–24 years and identifies those aged 10–14 years as being in a period of ‘early adolescence’, while those aged 15–19 years are in a period identified as ‘late adolescence’ (WHO, 2021). ‘Young adulthood’ included those aged 20–24 years, consistent with previous literature examining youth FV (Simmons et al., 2022).

Sample

The present study examined all police-reported incidents of FV (including multiple incidents for the same FV-user) involving a unique dyad in which a young person (aged 10–24 years) was identified as the FV-user (or Respondent, in police parlance) over the four-month period between 1 and 2019 and 31 December 2019 (index period; N = 5014).

In cases where a young person had engaged in FV in multiple unique dyads (e.g., child-to-parent, intimate partner, sibling, other family) during the index period, a new case was recorded for each unique dyad. However, in cases where a young person was reported multiple times for using FV within the same dyad during the index period, only the first incident of family violence that was entered into the LEAP system was recorded. There were 433 (8.64%) FV-users who were listed as using FV in more than one index dyad. This included 44 (7.41%) repeat FV-users aged 10–14 years, 166 (8.84%) aged 15–19 years, and 223 (9.82%) aged 20–24 years. The results provided in this study include repeat FV-users, as data analysis is based on unique dyads of abuse (rather than incidents of FV where each FV-user is counted only once).

Data were extracted from a wider population sample of all 24,419 FV reports recorded by police during the same period, of which 358 (1.50%) had missing age data and were excluded from selection for this sample. The index incident data obtained for each FV-user were linked with historical data held in police databases and FV recidivism data collected over a 6-month follow-up period.

Over two thirds of young people who used FV were identified as male (n = 3528, 70.40%), while 29.61% (n = 1484) were female. There was no sex information for two young people (0.04%), who were excluded from analyses involving the young FV-user’s sex. The mean age of the sample was 19.19 years (SD = 3.51) at the time of the index incident.

Approvals and Ethics Clearances

The study was approved by the Swinburne University Human Research Ethics Committee (SUHREC: November 30, 2020, reference: 20204231-5617) and the Victoria Police Research Coordinating Committee (Project 968).

Data Source

All information relating to FV incidents is recorded using FV reports and is stored on Victoria Police’s Law Enforcement Assistance Program (LEAP) database, which is used by Victoria Police to record all known offences and police involvements (FV- and non-FV-related) for an individual, regardless of outcome (e.g. arrested, charged, convicted). Whenever Victoria Police members respond to an incident of FV, they record characteristics of the incident, the victim, the person using FV, and their relationship as part of a FV report. The FV report used by police contains demographic information of FV-users and victims, as well as 39 separate risk factors associated with future FV or lethal FV incidents, allowing Victoria Police to routinely collect information on a range of evidence-based factors related to future FV events (McEwan et al., 2019). All available data from the FV reports involving a unique relationship dyad during the index period were linked to historical and outcome data for each young FV-user and victim. Historical data included information relating to involvement in past police-reported FV incidents for both the FV-user and the victim, history of non-FV offending by the young person, the presence of restraining orders (either past or current) against the young person, and history of police-reported FV victimisation experienced by the young person, including during childhood (aged 0–11 years). Family violence recidivism was recorded over a 6-month period following the index incident.

Sociodemographic, Psychosocial, and Family Violence-Related Characteristics

Sociodemographic Variables

The sex of young FV-users and the presence of any accessibility needs are recorded on the FV report at the time of the index incident. Accessibility needs of the FV-user form a component of the FV report and request responding police officers to identify whether the young person has issues relating to vision, hearing, mobility, understanding, communication, or memory. These items are scored according to police questioning and discretion (i.e. the officer may notice the young person has one or more of these issues, or they may ask the victim, FV-user, or third party whether the young person has any of these difficulties). Due to low prevalence of accessibility needs being recorded by police (n = 102, 2.03%), these items were grouped together and coded as a binary variable (i.e. either present or absent).

An approximation of socioeconomic status (SES) was coded using the ABS (2018) Index of Relative Socio-Economic Advantage and Disadvantage (IRSAD; using Victoria-specific rankings). The socioeconomic status of young FV-users was identified using the postcode recorded on the FV report and its corresponding IRSAD decile obtained from ABS (2018) data. Three groups were then created for the present study, with the first group comprising those in the lowest 20% SES (deciles one and two), middle 60% SES (deciles three to seven), and highest 20% (deciles eight to ten).

Psychosocial Variables

Variables related to the presence of mental health issues, substance abuse, and unemployment/school truancy were recorded in a binary manner (i.e. present or absent). These factors were primarily ascertained by police questioning, which typically involves obtaining information from multiple sources, including from the victim, young FV-user, and any relevant third parties (e.g., asking the young person and victim if they use substances, asking relevant third parties about whether the young person is substance-affected, and police noting whether the young person appears substance affected). Police scoring of psychosocial data could also be informed by their own observations at the scene (i.e., noticing whether the young person appeared substance affected or to be suffering from mental health-related phenomena). The presence of a formal diagnosis was not required to be scored in the affirmative. As a result, prevalence of these issues in the present study are likely to represent relatively gross estimates. Data relating to victimisation history of the young FV-user were obtained from historical FV reports on LEAP in which the index FV-user was listed as a victim of FV, including between 0 and 11 years old (inclusive).

Family Violence-Related Variables

Victim characteristics, including victim sex, age, and presence of a disability or mental health issues, are recorded in a binary manner (i.e. male or female; presence of a disability, presence of mental health issues) on the FV report at the time of the index incident. The relationship of abuse was recorded by the responding police officer according to the type of relationship between the victim and FV-user (i.e. child-to-parent abuse, intimate partner abuse, sibling abuse, other family abuse) at the time of the index incident. The category relating to abuse of other family members may include grandparents, uncles, aunts, cousins, carers, or the child of the young FV-user.

Family Violence Recidivism

Data pertaining to FV recidivism of the young person were obtained from any FV reports uploaded to LEAP in the six months following the index incident in which the young person was reported for using FV again toward any person. This was coded in a binary manner (i.e., yes, or no).

Data Analysis

Data were analysed using IBM SPSS Statistics, version 28 (2020). Family violence-users were first grouped into one of three developmental periods based on their age at the index incident (i.e., early adolescence, late adolescence, and young adulthood). To address the first aim, descriptive statistics were then provided for the sample overall and by age group.

The second aim was addressed using a series of chi-square analyses with odds ratios as a measure of effect size. Multiple comparisons were conducted for key sociodemographic, psychosocial, and FV-characteristics, with a Bonferroni-Holm correction applied to control for Type I error. Analyses related to the relationship in which abuse occurred at the index incident (child-to-parent abuse, intimate partner abuse, sibling abuse, child maltreatment, and other family abuse) and SES (i.e., lowest 20% SES) used binary dummy variables. For example, when determining whether child-to-parent abuse was more common among early adolescent compared with young adult FV-users, child-to-parent abuse was coded as 1 and all other relational dyads were coded as 0.

To address the third aim, a Kaplan-Meier survival curve was used to compare time to FV recidivism among early adolescent, late adolescent, and young adult FV-users. A log rank test was used to compare survival distributions across the three age groups.

Results

The results section is organised according to three segments which align with the aims of the study. The first segment provides a descriptive overview of the sample. The second segment provides results pertaining to differences in sociodemographic, psychosocial, and FV-related characteristics among young people. The third segment examines differences in time to FV recidivism according to age group, respectively.

Descriptive Overview

The characteristics of young FV-users and their FV behaviour at the time of the index incident are provided in Table 1. Of the total sample, approximately one in ten were aged 10–14 years, over one third were aged 15–19 years, and approximately half were aged 20–24 years. Regardless of age, young people who engaged in FV were predominantly male and there was an over-representation of youth in the lowest socioeconomic quintile. There was a high proportion of young people with mental health issues across all age groups, while substance abuse issues and unemployment/school truancy became more prevalent with age. The proportion of youth with histories of childhood FV victimisation was highest among early adolescent FV-users.

Table 1 Characteristics of young people who use family violence and their family violence behaviour at the time of the index incident
Full size table

Child-to-parent abuse and intimate partner abuse were the most common relationships of abuse at the index incident. The proportion of child-to-parent abuse incidents was highest among early- and late-adolescent FV-users, while the proportion of young people using intimate partner abuse increased with age and was most common among young adults. Despite this, approximately one in twenty early adolescent FV-users engaged in intimate partner abuse, and one quarter of young adults engaged in child-to-parent abuse. Over one quarter of all youth had ever been abusive across more than one type of relationship, while over one third were reported to police for FV recidivism within the subsequent six months.

Regardless of the young FV-user’s age, victims of youth FV incidents were predominantly female and aged in their thirties. One fifth of all victims were reported to have a mental health issue, while a minority of victims were identified as a having a disability.

Age-Related Differences in Youth Family Violence

The results of analyses testing for significant age-related differences in key sociodemographic, psychosocial, and FV characteristics are presented in Table 2. The multiple comparison analyses for some variables presented in Table 1 are not presented in Table 2 due to space constraints, however the results of these analyses have been provided in supplemental materials.

Table 2 Multiple comparisons examining sociodemographic, psychosocial, and index family violence characteristics according to FV-user age
Full size table

Sociodemographic Characteristics

Significant age-related differences in the young FV-user’s sex, socioeconomic status (SES) and accessibility needs were observed. The odds of young adult FV-users being male were significantly higher than for their adolescent counterparts, while early adolescent FV-users displayed a higher proportion of females compared to young adults FV-users. In contrast, early adolescent FV-users appeared to differ significantly from their young adult counterparts in relation to SES and accessibility needs. Compared to those aged 15 years and over, the odds of early adolescent FV-users being in the lowest SES quintile (i.e. lowest 20%) were significantly lower, however their odds of experiencing accessibility needs were 1.80–2.66 times higher.

Psychosocial Characteristics

No significant age-related differences in the likelihood of a young FV-user having mental health issues were observed, however age-related differences were found in relation to substance abuse, unemployment/school truancy, and police-reported FV victimisation in childhood. Substance abuse issues became more prominent as the young FV-user’s age increased, while the proportion of young people who had experienced FV victimisation in childhood (aged 0–11 years) was greatest among early adolescent FV-users.

Family Violence Characteristics

Significant age-related differences were observed for victim sex and the relationship in which the abuse occurred during the index incident. The odds of early adolescent FV-users aggressing against a female victim – and engaging in child-to-parent abuse – were significantly higher than their late adolescent and young adult counterparts. The odds of an early adolescent FV-user engaging in child-to-parent abuse were 6.01 times greater than for young adults, and 1.89 times greater than late adolescent FV-users. In contrast, the odds of a young adult FV-user engaging in intimate partner abuse were 24.45 times higher than early adolescents and 3.39 times higher than those in late adolescence.

Time to Family Violence Recidivism

No significant age-related differences were observed for rates of FV recidivism. Figure 1 depicts the Kaplan-Meier curve examining time to FV for early adolescent, late adolescent, and young adult FV-users. A log-rank test showed no statistically significant differences in the survival distribution for the three age groups (χ2(2) = 3.77, p = .152).

Fig. 1
figure 1

Time to family violence recidivism for early adolescent (10–14 years), late adolescent (15–19 years), and young adult (20–24 years) FV-users over a 6-month period

Full size image

Discussion

The present study sought to provide a descriptive overview of young FV-users and determine whether differences exist in sociodemographic, psychosocial, and FV-related characteristics – including time to recidivism – according to the age of the FV-user. To the authors’ knowledge, this represents the first study to examine the characteristics of young FV-users across early adolescence, late adolescence, and young adulthood. Consistent with the literature, results from this study show most FV-users were male and targeted a female victim (Freeman, 2018; Phillips & McGuinness, 2020; Snyder & McCurley, 2008), with many also having been the victim of police-reported FV and coming from low SES areas (Fitz-Gibbon et al., 2018; Phillips & McGuinness, 2020). The findings support existing research showing a high prevalence of mental health issues (Phillips & McGuinness, 2020), while issues with unemployment/school truancy and substance use were higher among those in late adolescence (Phillips & McGuinness, 2020) and young adulthood.

Family violence-users were disproportionately male, and victims were predominantly female across all age groups. Existing research has consistently shown adolescents who engage in FV to be more likely to come from single-parent households which are headed by a female, which may results in females being the most common victims simply due to opportunity (Cottrell & Monk, 2004; Lyons et al., 2015). Similarly, scholars have suggested that factors associated with single-parent households, such as financial hardship and social isolation, may be associated with increased conflict between the young person and parent (Pagani et al., 2003).

While the disproportionate number of male FV-users and female victims was evident across all age groups, in relative terms, sex differences existed across the three groups. Young adult FV-users were significantly more likely to be male and target male victims compared to their early- and late-adolescent counterparts, while those in early adolescence had a higher proportion of female FV-users and a higher proportion of female victims, although the effect sizes were small. At first glance these results may seem counterintuitive. Given the gendered nature of FV is most prominent among adults (Snyder & McCurley, 2008) it might be expected that the proportion of female victims increased alongside the increasing prevalence of male FV-users as young people aged. These findings are likely explained by the higher proportion of fathers being targeted by older youth (i.e. those in late adolescence and young adulthood; Simmons et al., 2018), the decline in prevalence of female FV-users with age (Simmons et al., 2018; Snyder & McCurley, 2008), and the tendency for females to target female victims (Boxall & Sabol., 2021; Freeman, 2018; Snyder & McCurley, 2018), although this latter finding has not been consistently reported in other studies (Simmons et al., 2018).

The high rates of unemployment/school truancy and substance abuse among late adolescent and young adult FV-users, in addition to the consistent rates of FV recidivism across all age groups, may be partially explained using the “age-risk factor paradox” (van der Put et al., 2011, p. 258) from the general offending literature. Given the present paper only examines two criminogenic needs it is not possible to determine whether the age-risk factor paradox is present in youth FV, however the variation in dynamic risk factors among young FV-users provides an interesting avenue for future research. The rates of school truancy among young people who use FV is significantly higher than the among the general population, with approximately 36% of young FV-users having issues with school truancy/unemployment compared to approximately 15% of the general Victorian population who displayed school attendance issues in 2020 (Victorian Department of Education, 2022). Similarly, substance abuse issues among young FV-users (ranging between 9 and 38%) are significantly higher than the general population (approximately 0.4% among those aged 0–24 years; Australian Institute of Health and Welfare (AIHW, 2022), although there is a high margin of error in this AIHW statistic and so it should be interpreted with caution.

Additional age-based analyses related to mental health issues, accessibility needs, and history of childhood FV victimisation indicates the presence of age-related variation in non-criminogenic needs that may impact upon a young person’s responsivity to intervention. Mental health needs and FV victimisation occurred at high rates across all age groups, indicating important responsivity considerations when attempting to intervene with this cohort, regardless of age. The results suggest mental health issues exist at a substantially higher rate for young FV-users than the general population, with data from the AIHW (2022) indicating that approximately 16.3% of those aged 0–24 years’ old experience mental and behavioural problems. While the rates of accessibility needs (i.e., a police indicator of disability) in the present study were found to be approximately similar to the general population – whereby 3 – 4.4% of the general population aged 5–24 years were identified as having a disability (Australian Bureau of Statistics, 2022) – it is possible the rates of disability among young FV-users is higher than what is reported here, as police are not typically trained to assess or screen for disability. It is important to note that the base rates provided here of mental health issues and accessibility needs – as well as of the school truancy and substance use issues described above – in the general population are representative of different age brackets and typically assess diagnosed conditions. In contrast, the presence psychosocial characteristics in the present study were ascertained through the use of police questioning and discretion. Given this, direct comparison of these findings with broader population base rates is difficult. Further research is needed to explore the prevalence of psychosocial characteristics such as mental health issues, disability, and associated needs among young people who use FV, as they represent important responsivity factors relevant to assessment and intervention.

Limitations

The present study is limited in several respects. First, the use of official police records to determine FV will have underestimated the true extent of youth FV and recidivism, as not all cases are reported to police. Relatedly, recidivism data were limited to 6-months follow-up and outcomes were also likely impacted by the intervention of police and other services. While these issues limit the generalisability of findings, and prevent conclusions being drawn for recidivism beyond six months, the results of the present study are particularly relevant to stakeholders that rely on FV being officially reported in order to intervene, including police, youth justice, and other community services. Second, the results are based on police-reported incidents of youth FV, not necessarily incidents in which a criminal offence or arrest has occurred. This potentially limits comparison with other studies which do use offence and/or arrest data, although it also ensures a broader spectrum of FV behaviours is examined. Third, while this research examined characteristics of FV among young people of different age groups, it cannot be used to make statements about the onset of FV behaviour nor how a young person’s FV behaviour may change over time. Fourth, the lack of a non-FV control group and limited information about the base rate of certain characteristics within the broader population limits our understanding of how characteristics may correlate to risk of future FV.

Fifth, the recording of data was completed by responding police officers at the time of the index FV incident, which may increase the risk of recording errors given the dynamic and often high-stress situations in which FV occurs. Similarly, the data pertaining to mental health and substance abuse issues was ascertained using police questioning and discretion (i.e. asking the young person if they use substances, asking the victim if the young person has mental health issues, noticing the young person appears substance affected), meaning the results in the present study may be relatively gross estimates of the prevalence of these issues. However, the use of police-reported data to examine prevalence of mental health and substance abuse issues has been used previously (Millsteed & Coghlan, 2016; Phillips & McGuinness, 2020). Sixth, the present study has examined correlates of FV and how these differ across groups, however specific analyses examining whether these correlates relate to FV behaviours and recidivism were not conducted. Given this, further research is needed examining whether the drivers of FV recidivism differ across developmental groups. This may include examination of the influence of criminogenic needs on FV recidivism at various ages (see van der Put et al., 2011, 2012 for example), and exploration of how the prevalence and influence of risk factors in recidivism change as a young person gets older.

Seventh, the lack of data pertaining to race and ethnicity, as well as other sociodemographic and psychosocial information (e.g., household structure, whether the young FV-user is living with the victim at the time of the index incident), represent important limitations of the present study. Unfortunately, there is no reliable way for Victoria Police to ascertain the ethnicity of individuals involved in a FV incident (McEwan et al., 2019), and such additional sociodemographic and psychosocial information is not routinely collected so could not be analysed.

Implications

The findings represent preliminary evidence for the similarities and differences in the characteristics of young FV-users and their associated FV behaviour across early adolescence, late adolescence, and young adulthood. While additional research in this area is undoubtedly needed, the findings may assist police and other services to screen for vulnerabilities of both victims and FV-users, with this information then being used to assist with identifying key referral pathways. For example, it may be important to prioritise efficient mental health referral pathways for young FV-users, given a significant minority of this cohort, regardless of age, were found to experience mental health-related issues. Similarly, the higher rates of substance abuse and unemployment/school truancy issues among those in late adolescence and young adulthood suggest the presence of important intervention targets for these youth, indicating a need for services to be able to screen for, and respond to, such issues.

Additionally, findings suggest interventions for young people in late adolescence and early adulthood (i.e., 15–24 years) will likely need to target a wider range of factors than interventions for those in early adolescence. These findings are consistent with the general youth offending research (van der Put et al., 2011; Put et al., 2012) and indicate interventions may need to concurrently address mental health issues, substance abuse, history of FV victimisation, and unemployment/school truancy. In contrast, interventions for early adolescent FV-users (10–14 years) may be able to focus on a smaller number of issues, but will need to be particularly cognisant of responsivity factors, including the presence of accessibility needs and a comparatively higher level of historical police-reported FV victimisation in childhood.

Future Research

Further research examining age-related differences in the characteristics of FV-users and their associated behaviour is needed to assist in informing developmentally-sensitive risk assessment, management, and intervention strategies. Such research may further explore the influence of various criminogenic needs on FV recidivism at various ages (see van der Put et al., 2011, 2012 for example), examine whether the effectiveness of various risk management strategies (e.g., intervention orders, diversion) are more effective for young FV-users at different ages, and further explore the intervention needs of FV-users at different developmental stages.

Similarly, the needs of victims must also be examined. Existing research suggests the nature of the relationship between the victim and FV-user, and the age of the young FV-user, influence the likelihood of reporting FV (Kang & Lynch, 2014). It therefore appears reasonable to suggest that these factors, particularly the age of the young FV-user, could also influence other victim-related factors, such as likelihood of the victim being injured or the victim’s perceived ability to comply with police-imposed risk management strategies (e.g., intervention orders).

Conclusion

The present study adds to existing evidence by examining how the sociodemographic, psychosocial, and FV-related characteristics – including time to recidivism – of youth FV differ according to the developmental periods of early adolescence, late adolescence, and young adulthood. Management and intervention approaches will need to be cognisant of the high levels of mental health issues across all age groups while also considering the age-related variation in sociodemographic, psychosocial, and FV-related characteristics which exist. Additionally, the results show an absence of a statistically significant difference in FV recidivism between those in early adolescence, late adolescence, and young adulthood.