Abstract
Background Child maltreatment has received substantial academic attention. However, little remains known about the intersection of different types of child maltreatment and their association with young people's use of violence in the home. Objective To examine the intersection of child maltreatment types and the intergenerational transmission of violence through young people's use of violence in the home. Participants and setting: A total of 5021 young people (16–20 years old) completed an online survey, recruited using non-probability sampling via survey panels. Methods Analyses were carried out using χ2 tests and a series of binary logistic regressions to examine the association between participant characteristics and unique child maltreatment experiences with different forms of maltreatment and young people's use of violence. Results Overall, 29.9% of participants had experienced at least one form of maltreatment and 16.7% reported experiences of multi-type maltreatment. The most common experiences were childhood experiences of domestic violence (CEDV) (27.3%), followed by verbal/emotional abuse (17%), which both increased the likelihood of also experiencing physical abuse (OR = 5.85 and OR = 10.21, respectively). Cisgender females and children living with a disability were more likely to experience all four types of maltreatment. Experiences of verbal/emotional abuse (OR = 4.56), and CEDV (OR = 4.52) increased the risk of young people's use of violence in the home. Conclusions Findings contribute to a growing body of work recognising CEDV as a distinct form of child maltreatment which intersects with other experiences of abuse, including young people's use of violence in the home.
1. Introduction
Child maltreatment is associated with numerous adverse impacts on victim-survivors at a physical, mental, and psychosocial level throughout the lifespan (Carr et al., 2020), and in some cases, it results in fatal outcomes (Angelakis et al., 2019). In addition to the short and long-term negative consequences experienced by victim-survivors, child maltreatment creates a substantial economic burden on societies assoacited with increased medical costs due to poor physical and mental health outcomes and household financial hardship (Harter & Harter, 2021; Kezelman et al., 2015; Liming, 2019). Despite the debate around which experiences constitute child maltreatment, the International Classification of Violence against Children (ICVAC) proposes six categories designed to cover all acts that are considered violence against children, including physical abuse, psychological abuse, sexual abuse, neglect, homicide and other uncategorised acts of violence (Cappa et al., 2023). In the last three decades, children's experiences of domestic violence (CEDV) between parents and/or parent-like caregivers have been increasingly recognised as a type of child maltreatment (Mathews et al., 2020; Radford et al., 2013; World Health Organization, 2020). CEDV includes instances where a young person experiences a parent or other caregiver being subjected to one or more forms of abuse by another parent or caregiver (Hamby et al., 2011). The recognition of these experiences as a type of child maltreatment has led to a surge of scholarly interest in understanding its scope and negative effects on young people (Dodaj, 2020; Kiesel et al., 2016), and to the acknowledgement of children and young people growing up with experiences of domestic violence (DV) as victim-survivors in their own right (Department of Social Services [DSS], 2022).
The importance of accounting for multi-type maltreatment to avoid overestimating the adverse impacts of child maltreatment when examining the associations with single types of abuse has been increasingly recognised in the literature (Fang et al., 2015; Higgins et al., 2023). A review by Debowska et al. (2017) emphasises the prevailing gaps in evidence relating to multi-type child maltreatment, specifically regarding emotional abuse and CEDV. This article aims to address this gap by examining the role of risk factors across unique and multi-type experiences of child maltreatment, including CEDV. To do so, it moves beyond an incident-based model of child maltreatment towards understanding patterns of abuse. Drawing on data from a national survey of over 5000 Australian young people (aged 16–20 years old), this article provides new insights into the intersecting experiences of child maltreatment and the intergenerational transmission of violence by examining self-reported childhood experiences of maltreatment along with later use of violence in the home among study participants. While a large body of evidence alerts to the adverse short- and long-term effects of child maltreatment, including educational disengagement, mental ill health and other life outcomes (Kiesel et al., 2016; Lawrence et al., 2023; Pacella et al., 2023; World Health Organization, 2020), there is a paucity of research examining the self-reported intergenerational transmission of violence. Research has shown that child maltreatment experiences are common among adult DV offender populations (Li et al., 2020; Roberts et al., 2010) but to our knowledge this is the first Australian prevalence study to predict the risk of use of violence among young people in relation to their experiences of different forms of child maltreatment.
2. Background: evidence on child maltreatment prevalence and impacts
Research evidence on child maltreatment prevalence rates including CEDV remains inconsistent, with many studies focusing on the prevalence of types of maltreatment that are directed at children, such as emotional, physical and sexual abuse and neglect. For example, in a recent systematic review, Mathews et al. (2020) identified only 16 out of 23 reviewed studies included measures of CEDV. Failure to include CEDV in explorations of child maltreatment hinders our understanding of the nature and impacts of childhood victimisation experiences given the high prevalence rates of DV in households where children are present (Fitz-Gibbon et al., 2022a). Estimates of the prevalence of child maltreatment vary depending on various factors, including the type of maltreatment that is being measured, variation in definitions and methodological differences between studies. In a series of meta-analyses of specific child maltreatment types, Stoltenborgh et al. (2015) found combined global prevalence rates from self-report studies for both boys and girls of 22.6 per cent for physical abuse and 36.3 per cent for emotional abuse. More recent research evidence suggests higher prevalence rates of physical abuse in sub-Saharan African countries, with 49.7 per cent of boys and 36.5 per cent of girls experiencing this type of violence, whereas the prevalence of emotional abuse in these countries was reported to be lower, with 7.7 per cent of boys and 9.2 per cent of girls identifying experiences of emotional abuse (Amene et al., 2024). While sexual abuse also affects young people of all genders, research shows that girls are more frequently victims of sexual abuse than boys (16% vs 8%) (Amene et al., 2024; Stoltenborgh et al., 2015).
Research exploring CEDV between parents or carers has previously found prevalence estimates around 23.7 per cent (Amene et al., 2024; Radford et al., 2013). However, findings from a recent Australian study using a nationally representative sample revealed a significantly higher prevalence rate of CEDV compared to those reported internationally (Mathews, Pacella, et al., 2023). Specifically, nearly half (43.8%) of the individuals aged 16 to 24 at the time of the study reported CEDV experiences before the age of 18 (Mathews, Pacella, et al., 2023). According to these results, out of the five types of child maltreatment, CEDV was the most frequently reported form, followed by emotional abuse (34.6%), physical abuse (28.2%), sexual abuse (25.7%), and neglect (10.3%) (Mathews, Pacella, et al., 2023).
Although some studies have focused on a single type of child maltreatment (e.g., Annerbäck et al., 2010), different forms of maltreatment rarely occur in isolation. Multi-type child maltreatment involves the co-occurrence of one or more types of child maltreatment (Higgins & McCabe, 2000). Research evidence indicates that different types of child maltreatment are often correlated and multi-type victimisation has been associated with the most adverse effects (Debowska et al., 2017). For example, using a nationally representative sample, Higgins et al. (2023) examined the prevalence of multi-type child maltreatment by exploring five types of child maltreatment in Australia. Findings indicated that two in five participants (39.4%) reported experiencing any multi-type child maltreatment (i.e., 2–5 types), while a smaller proportion (22.8%) reported experiencing only one type of maltreatment (Higgins et al., 2023). Although less common, 3.5 per cent of participants reported experiencing all five types of maltreatment (Higgins et al., 2023).
Adopting a socio-ecological approach (Bronfenbrenner, 1977), risk factors for child maltreatment have been studied at various levels. For example, at the microsystem and ecosystem levels, stressors such as high family conflict, housing stress, financial stress, childcare burden, harmful substance use or poor mental health of the caregivers have been identified as factors increasing the risk of abusive childhood experiences (Conrad-Hiebner & Byram, 2020; Freisthler et al., 2006). Some of the factors identified in the literature, such as adverse economic conditions, can be understood as structural drivers and are likely to have a greater impact in low- and middle-income countries, where economies are considerably less stable (Maternowska & Fry, 2018).
At an individual level, child characteristics have also been linked to a higher likelihood of experiencing specific types of child maltreatment (Clements-Nolle et al., 2018; Craig et al., 2020; Mathews, Pacella, et al., 2023; Spencer et al., 2005; Sullivan & Knutson, 2000). For example, children and young people with disabilities have been identified as particularly vulnerable to maltreatment. Findings from a United States population-based epidemiological study revealed that individuals with disabilities were 3.4 times more likely to report child maltreatment than those without a disability (Sullivan & Knutson, 2000). Concerning specific types of child maltreatment, children with disabilities were 3.79 times more likely to be physically abused, 3.88 times more likely to be emotionally abused, 3.14 times more likely to experience sexual abuse, and 3.76 times more likely to be victims of neglect compared to children with no disability (Sullivan & Knutson, 2000). Similarly, Spencer et al. (2005) found that children with disabilities had a 2.7 to 11.6 times greater risk of experiencing child maltreatment than children without disabilities. Nevertheless, as Leeb et al. (2012) note, researchers have not always controlled for variables that may confound the relationship between disability and child maltreatment, such as demographic characteristics, timing and severity of maltreatment, or the inclusion of other maltreatment types.
Other factors associated with an increased risk of experiencing child maltreatment include gender and cultural identity. As noted earlier, an extensive body of work highlights the gendered risk of child sexual abuse experiences, with girls experiencing double the rate of sexual abuse compared to boys (Barth et al., 2013; Mathews, Pacella, et al., 2023; Stoltenborgh et al., 2015). While some studies have identified similar prevalence rates of physical and emotional abuse for girls and boys (Stoltenborgh et al., 2015), others have identified a higher risk of emotional abuse for girls (Mathews, Pacella, et al., 2023). More recent research further differentiates between gender, diverse sexual and gender identities as a risk factor, alerting to an increased risk of child maltreatment experiences among lesbian, gay, bisexual, transgender, intersex, and queer (LGBTIQ+) children and young people (Craig et al., 2020). For example, latest nationally representative Australian research shows that gender diverse and sexuality diverse individuals are significantly more likely to experience each type of child maltreatment (i.e., physical abuse, sexual abuse, emotional abuse, neglect and CEDV) as well as multi-type maltreatment, compared to women and men (Higgins et al., 2024).
Further, research on child maltreatment and the experiences of cultural minority groups has alerted to an increased risk for certain populations of children, specifically First Nations children. For example, research on child maltreatment prevalence rates in Canada has identified an increased risk of child maltreatment for First Nations children compared to non-First Nations children (Brownridge et al., 2017). similarly, a higher risk of child maltreatment experiences has been identified for First Nations children in the Australian context (Segal et al., 2019). Additionally, research suggests that culturally and linguistically diverse (CALD) children are more likely to experience child maltreatment than non-CALD children, especially physical abuse (Abdul Rahim et al., 2023) and emotional abuse (Schick et al., 2015). Importantly, disparities in the prevalence of child maltreatment among different ethno-racial groups reflect systemic and structural drivers that disproportionally affect First Nations and CALD children, including ongoing racism and discrimination, and economic hardship (Austin et al., 2020; Australian Institute of Family Studies [AIFS], 2020; Ma et al., 2019).
Recognising existing work, this article further contributes timely evidence about young people's experiences of child maltreatment in Australia. While many studies explore retrospective experiences of child maltreatment in adult samples, often reflecting on experiences that date back multiple decades (Mathews, Pacella, et al., 2023; Nagy et al., 2019), the current study explores retrospective experiences of child maltreatment up to the age of 18 years within a sample of 16- to 20-year-olds who participated in an Australian-based online survey. As a result, experiences sit within the policy and legislative landscape of the last two decades, with many experiences being recent and, at times, ongoing.
3. Methods
3.1. Sample
Young people who participated in this cross-sectional Australian survey study were recruited through research panels managed by Open Research Unit (ORU). The survey was sent to participants living in Australia aged 16–20 years old. A total of 5021 young people participated in the survey. The average age of the sample was 18.16 years (SD = 1.3). Around two-thirds of the sample identified as female assigned at birth (n = 3348) and around one-third identified as male assigned at birth (n = 1623). The overrepresentation of female participants in the sample might reflect gender differences in interest and willingness to participate, also observed in other research (Nuzzo & Deaner, 2023). Further, four per cent of participants identified as having a diverse gender identity. Here, we refer to participants with “diverse gender identities” as those who are not cisgender, including trans and non-binary young people (ACON, 2019). The majority of participants (n = 3275; 65.2%) identified as heterosexual or straight, with 29.2% identifying as sexuality diverse (n = 1468). This group included participants who reported a sexual orientation other than heterosexual. Ten per cent identified as being from a non-English speaking background (NESB) and 5 per cent identified as First Nations Australians. Over a third of survey participants (36%) identified as living with at least one form of disability and/or neurodiversity, which is significantly higher than rates identified in the general population, where 9% identified as living with a disability in Australia's latest youth survey (Tiller et al., 2021). Most of the sample (83%) resided in a major Australian city, followed by 15 per cent of young people residing in regional areas and 2 per cent residing in remote areas. This constitutes a slight overrepresentation of metropolitan participants and underrepresentation of regional participants, with 72 per cent of Australians residing in major cities and 26 per cent residing in regional settings (Australian Bureau of Statistics, 2022).
Participants were recruited using a non-probability sampling method, which did not aim to reflect recent trends in characteristics of young people living in Australia. The survey was only offered in English. The sample is not representative of the wider Australian population, especially with regards to culturally and linguistically diverse young people and young people with limited or no access to the internet and online devices. Findings reported here are therefore not generalisable to the whole Australian population of 16- to 20-year-olds.
3.2. Measures
The data were collected as part of the Adolescent Family Violence in Australia (AFVA) study (see Fitz-Gibbon et al., 2022a; Fitz-Gibbon et al., 2022b) during September and October 2021 by ORU using Qualtrics Software. The survey included close-ended and open-ended questions about participants’ sociodemographic characteristics, their living arrangements, their experiences of violence prior to the age of 18 years, including experiencing violence between other family members, being subjected to direct forms of child maltreatment perpetrated by other family members, and their own use of violence in the home at any age.
3.2.1. Sociodemographic information
The survey asked participants to provide the following sociodemographic information: age, gender identity, sexuality, First Nations status, and primary language spoken at home.
3.2.2. Disability and/or neurodivergence
This variable captured neurodiversity, cognitive, physical and audio-visual disabilities. Participants who identified with one or more forms were considered to be living with at least one disability and this variable label is used throughout this paper.
3.2.3. Child maltreatment
Survey items measured eight different types of direct experiences of child maltreatment, including physical abuse, property damage, verbal/emotional abuse, threats to harm or hurt someone close, threats to kill, sexual abuse, non-fatal strangulation, and sexuality and/or gender identity-based abuse. To assess the presence of CEDV, participants were also asked whether they had ever experienced any of these behaviours between parents or adult carers. In this article, we focus on young people's experiences of physical abuse (e.g., hitting, slapping, pushing, punching, kicking), verbal/emotional abuse (e.g., being put down, being told you're useless/stupid/ugly), sexual abuse (e.g., your private parts being touched, being made to touch someone else's private parts, being forced to have sex) and CEDV. Follow-up questions in the survey assessed the nature of these experiences, including the age of first experiencing violence, the relationship with the perpetrator/victim-survivor, the frequency of the abusive behaviours, the impacts of their experiences, whether they had reported their experiences and which services and supports they had found most and least helpful, and why. While the wider study examined experiences of violence involving a diverse range of family members (see Fitz-Gibbon et al., 2022a), this article focuses on CEDV between parents or carers along with experiences of child maltreatment by a parent or carer, with the exception of child sexual abuse experiences, which are included regardless of the perpetrator identity. This decision is supported by the extensive research evidence indicating that experiences of child sexual abuse have a significant impact on later life outcomes, irrespective of the victim-offender relationship (Bertele & Talmon, 2023). Other experiences of violence perpetrated by or between siblings were not included in this study as such experiences may include common patterns of sibling fighting that do not form part of the broader definitions of child maltreatment and may lead to an overestimation of victimisation experiences.
Experiences of physical and verbal/emotional abuse are included in the analysis where young people reported frequent experiences of these types of child maltreatment. This is, where these forms of abuse were perpetrated on at least a monthly basis by an adult carer. Similarly, CEDV between adult carers were included if reported as occurring on at least a monthly basis (i.e., frequent CEDV). This approach is in line with our focus on examining experiences and impact of child maltreatment as a pattern of abuse and to avoid overestimating experiences of child maltreatment. It is further consistent with other recent child maltreatment research, which examines prevalence rates and the predicted effect of experiences of different forms of child maltreatment on different life outcomes in the context of patterns of emotional abuse and/or neglect (Lawrence et al., 2023; Mathews, Pacella, et al., 2023). The child sexual abuse variable included experiences of both contact abuse without intercourse (touching) and/or forced intercourse. Conceptually, both contact and non-contact sexual abuse have been described as highly impactful. Further, while research indicates that the majority (78%) of child sexual abuse survivors report more than one victimisation experience (Haslam, et al., 2023, 2023b), a single occurrence has been described as damaging as a pattern of occurrences (Pérez-Fuentes et al., 2013). Therefore, experiences of child sexual abuse were based on positive endorsement of this item, regardless of the frequency of the behaviours or the family relationship between the victim-survivor and the perpetrator (e.g., including sibling relationships).
3.2.4. Use of violence in the home
Participants’ use of violence in the home was measured including instances where young people self-reported using family violence, including physical violence, verbal and/or emotional abuse against an adult carer and occurring at least monthly.
3.3. Statistical analysis
Missing data were handled using listwise deletion and excluded from the analyses. The analysis was conducted in two stages. First, experiences of child maltreatment were explored at a univariate (descriptive) level. All four child maltreatment experiences were coded as dichotomous variables, representing the absence or presence of patterns of abusive experiences. We calculated prevalence rates for the overall sample and by participant characteristics (i.e., sex, gender identity, sexuality, First Nations status, primary language spoken at home, and disability). Associations between each type of maltreatment (as dependent variables) and young people-related characteristics (as independent variables) were described using cross-tabulations with chi-square tests of significance.
The second stage of the analysis involved multivariate analyses. Binary logistic regression models were fitted for experiences of any child maltreatment type and for each of the four child maltreatment types. For experiences of any type of child maltreatment, one model was fitted, including participant characteristics as predictors. For each type of child maltreatment, two separate models were fitted. The first models only included participant characteristics as predictors, while the second models, in addition to the participant characteristics, included the other forms of child maltreatment types as predictors. Similarly, two separate models were fitted to examine the use of violence in the home. The first model included participant characteristics as predictors, while the second model incorporated both participant characteristics and all forms of child maltreatment types as predictors. Logistic regression is reported with 95 per cent confidence intervals (CI) and odds ratios (ORs). Here, ORs indicated an increased risk of each form of child maltreatment and use of family violence in the home. All analyses were conducted using IBM SPSS Statistics (Version 27).
3.4. Ethical considerations
This study received ethical approval from the [Monash University] University Human Research Ethics Committee (MUHREC; project ID: 27269).
4. Results
4.1. Prevalence of frequent experiences of child maltreatment
In total, 5021 participants completed the survey. Prevalence rates of child maltreatment experiences are provided in Table 1, Table 2 for the sample and stratified across demographic characteristics, and disability and neurodiversity status. In this sample, the most reported type of child maltreatment was CEDV between two adult carers (n = 1371; 27.3%), followed by verbal/emotional abuse (n = 851; 17%). Experiencing frequent physical abuse was reported by 296 participants (5.9%) and 112 respondents (2.2%) reported having experienced child sexual abuse. Regarding experiences of multi-type child maltreatment, 1502 participants (29.9%) reported experiencing at least one type of frequent child maltreatment. Experiencing two types was reported by 575 respondents (11.5%), experiencing three types was reported by 233 participants (4.6%), and 29 participants (0.6%) reported experiencing all four types of child maltreatment. This meant that 837 participants (16.7%) reported having experienced two or more forms of child maltreatment. The prevalence of multi-type child maltreatment by demographic characteristics and disability is provided in Table 1.
Table 1. Prevalence of frequent child maltreatment by any type of maltreatment and multi-type maltreatment, and by demographic characteristics and disability.
Any type of maltreatment (%) n% | Any type of maltreatment (%) Significance Cramer's V | One type of maltreatment (%) | Two types of maltreatment (%) | Three types of maltreatment (%) | Four types of maltreatment (%) | |
|---|---|---|---|---|---|---|
Sex assigned at birth a | ||||||
Female | 1220 (36.4) | 0.21∗∗∗ | 523 (15.6) | 472 (14.1) | 198 (5.9) | 27 (0.8) |
Male | 266 (16.4) | 135 (8.3) | 97 (6.0) | 32 (2.0) | 2 (0.1) | |
Gender identityb | ||||||
Gender normative | 1366 (28.6) | 0.13∗∗∗ | 623 (13.1) | 517 (10.8) | 200 (4.2) | 26 (0.5) |
Gender diverse or questioning | 124 (57.7) | 38 (17.7) | 54 (25.1) | 29 (13.5) | 3 (1.4) | |
Sexual identityc | ||||||
Heterosexual | 818 (25.0) | 0.19∗∗∗ | 384 (11.7) | 323 (9.9) | 100 (3.1) | 11 (0.3) |
Diverse sexualities | 639 (43.5) | 250 (17.0) | 245 (16.7) | 127 (8.7) | 17 (1.2) | |
First Nations statusd | ||||||
First Nations | 88 (34.4) | 0.02 | 32 (12.5) | 31 (12.1) | 21 (8.2) | 4 (1.6) |
Not First Nations | 1402 (29.7) | 632 (13.4) | 538 (11.4) | 208 (4.4) | 24 (0.5) | |
Language spoken at homee | ||||||
MESB | 1383 (30.8) | 0.06∗∗∗ | 607 (13.5) | 538 (12.0) | 212 (4.7) | 26 (0.6) |
NESB | 116 (22.1) | 57 (10.9) | 36 (6.9) | 20 (3.8) | 3 (0.6) | |
Disabilityf | ||||||
No disability | 618 (20.3)∗∗∗ | 0.29∗∗∗ | 319 (10.5) | 211 (6.9) | 83 (2.7) | 5 (0.2) |
One or more disability | 836 (47.8) | 316 (18.1) | 351 (20.1) | 146 (8.4) | 32 (1.3) |
MESB = mainly English speaking background; NESB = non-English speaking background.
∗p < .05.
∗∗p < .01
.∗∗∗p < .001.
a
Excludes 50 participants who did not provide this information.
b
Excludes 33 participants who did not provide this information.
c
Excludes 278 participants who did not provide this information.
d
Excludes 42 participants who did not provide this information.
e
Excludes 6 participants who did not provide this information.
f
Excludes 229 participants who did not provide this information.
Table 2. Prevalence of type of child maltreatment by demographic characteristics and disability.
Physical abuse n (%) | Physical abuse Significance Cramer's V | Verbal/emotional abuse n (%) | Verbal/emotional abuse Significance Cramer's V | Sexual abuse n (%) | Sexual abuse Significance Cramer's V | CEDV n (%) | CEDV Significance Cramer's V | |
|---|---|---|---|---|---|---|---|---|
Sex assigned at birth a | ||||||||
Male | 43 (2.6) | 0.10∗∗∗ | 131 (8.1) | 0.17∗∗∗ | 14 (0.9) | 0.07∗∗∗ | 245 (15.1) | 0.19∗∗∗ |
Female | 249 (7.4) | 711 (21.2) | 97 (2.9) | 1112 (33.2) | ||||
Gender identityb | ||||||||
Gender normative | 262 (5.5) | 0.08∗∗∗ | 756 (15.8) | 0.13∗∗∗ | 96 (2.0) | 0.06∗∗∗ | 1247 (26.1) | 0.12∗∗∗ |
Diverse genders | 31 (14.4) | 87 (40.5) | 14 (6.5) | 113 (52.6) | ||||
Sexual identityc | ||||||||
Heterosexual | 143 (4.4) | 0.11∗∗∗ | 439 (13.4) | 0.17∗∗∗ | 44 (1.3) | 0.09∗∗∗ | 748 (22.8) | 0.18∗∗∗ |
Diverse sexualities | 147 (10.0) | 396 (27.0) | 61 (4.2) | 585 (39.9)
| ||||
First Nations statusd | ||||||||
First Nations | 26 (10.2) | 0.04∗∗ | 51 (19.9) | 0.02 | 16 (6.3) | 0.07∗∗∗ | 80 (31.3) | 0.02 |
Not First Nations | 265 (5.6) | 788 (16.7) | 93 (2.0) | 1282 (27.1) | ||||
Language spoken at homee | ||||||||
MESB | 273 (6.1) | 0.02 | 787 (17.5) | 0.05∗∗∗ | 104 (2.3) | 0.02 | 1259 (28.0) | 0.05∗∗∗ |
NESB | 22 (4.2) | 62 (11.8) | 8 (1.5) | 109 (20.8) | ||||
Disabilityf | ||||||||
No disability | 106 (3.5) | 0.14∗∗∗ | 312 (10.2) | 0.25∗∗∗ | 36 (1.2) | 0.09∗∗∗ | 556 (18.2) | 0.28∗∗∗ |
One or more disability | 185 (10.6) | 520 (29.8) | 71 (4.1) | 772 (44.2) |
MESB = mainly English speaking background; NESB = non-English speaking background.
∗p < .05.
∗∗p < .01.
∗∗∗p < .001.
a
Excludes 50 participants who did not provide this information.
b
Excludes 33 participants who did not provide this information.
c
Excludes 278 participants who did not provide this information.
d
Excludes 42 participants who did not provide this information.
e
Excludes 6 participants who did not provide this information.
f
Excludes 229 participants who did not provide this information.
Prevalence rates of experiencing at least one type of child maltreatment frequently (i.e., at least monthly) varied by demographic characteristics, and disability and neurodiversity status (see Table 1). For example, girls reported higher rates of experiencing any type of child maltreatment than boys (36.4% ν 16.4%; χ2 (1, N = 4971) = 209.68, p < 0.001). Higher rates of any form of child maltreatment were also present among young people with gender diverse identities compared to gender normative participants (57.7% ν 28.6%; χ2 (1, N = 4988) = 82.91, p < 0.001), and among young people with a diverse sexuality identity compared to heterosexual participants (43.5% ν 25.0%; χ2 (1, N = 4743) = 163.92, p < 0.001). These differences were statistically significant. Additionally, young people from a mainly English-speaking background (MESB) were statistically more likely to report experiences of any type of child maltreatment compared to those with a NESB (30.8% ν 22.1%; χ2 (1, N = 5015) = 17.00, p < 0.001). However, there was no difference between First Nations and non-First Nations participants regarding their experiences of any form of child maltreatment (34.4% ν 29.7%; χ2(1, N = 4979) = 2.55, p = 0.110). Lastly, the rates of experiences of any form of child maltreatment frequently were statistically significantly higher for young people who reported having one or more disabilities compared to those without a disability (47.8% ν 20.3%; χ2(1, N = 4796) = 399.14, p < 0.001). The prevalence of each specific type of child maltreatment by demographic characteristics and disability are provided in Table 2, along with significant differences between groups.
4.2. Association between participant characteristics and any type of child maltreatment
Results from the binary logistic regression examining the relationship between having experienced any type of the child maltreatment forms explored in this study and participant characteristics are presented in Table 3. After controlling for a range of factors, several variables were independently associated with young people's reports of having experienced any type of child maltreatment. Specifically, girls, sexuality diverse individuals, and young people with diverse genders were more likely to report they had been subjected to any form of child maltreatment on a frequent basis when compared to the rest of the population. Among the variables examined, the strongest predictor of experiencing at least one form of child maltreatment was having a disability. The likelihood of experiencing child maltreatment was almost three times higher among those who reported living with at least one disability than those who did not. However, the results showed that First Nations status and NESB status were not independently associated with higher reports of experiencing any type of child maltreatment.
Table 3. Logistic regression model for at least one type of child maltreatment.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Female | 0.879 | 0.83 | 112.176 | 2.41a | [2.05–2.83] |
Diverse genders | 0.646 | 0.178 | 13.129 | 1.91a | [1.35–2.71] |
Diverse sexualities | 0.406 | 0.077 | 27.675 | 1.50a | [1.29–1.75] |
First Nations status | 0.068 | 0.153 | 0.200 | 1.07 | [0.79–1.45] |
NESB | −0.137 | 0.125 | 1.200 | 0.87 | [0.68–1.11] |
Disability | 1.012 | 0.072 | 199.391 | 2.75a | [2.39–3.17] |
χ2 = (6, n = 4485) = 559.05, p < 0.001, Nagelkerke R2 = 0.17
NESB = non-English speaking background.
Note. OR: Odds ratio; CI: Confidence interval.
∗p < 0.05.
∗∗p < 0.01.
a
p < .001.
4.3. Physical maltreatment across participant characteristics and other types of child maltreatment
To establish baseline data about the relationship between young people's characteristics and their experiences of physical abuse, we performed a first binary logistic regression model including the six respondent characteristics as independent variables and physical abuse as dependent variable (Table 4). Results indicated that girls, sexuality diverse youth, First Nations young people, and participants who reported having a disability were more likely to report experiences of physical abuse during childhood.
Table 4. Logistic regression models for physical abuse.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 1 | |||||
Female | 0.754 | 0.174 | 18.839 | 2.12b | [1.51–2.99] |
Diverse genders | 0.434 | 0.236 | 3.373 | 1.54 | [0.97–2.45] |
Diverse sexualities | 0.464 | 0.138 | 11.359 | 1.59b | [1.22–2.08] |
First Nations status | 0.498 | 0.228 | 4.780 | 1.65a | [1.05–2.57] |
NESB | −0.127 | 0.249 | 0.261 | 0.88 | [0.54–1.43] |
Disability | 0.876 | 0.136 | 41.255 | 2.40b | [1.84–3.14] |
χ2 = (6, n = 4481) = 138.08, p <. 001, Nagelkerke R2 = 0.08
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 2 | |||||
Female | 0.135 | 0.195 | 0.477 | 1.14 | [0.78–1.68] |
Diverse genders | −0.036 | 0.265 | 0.019 | 0.96 | [057-1.62] |
Diverse sexualities | 0.167 | 0.159 | 1.105 | 1.18 | [0.87–1.61] |
First Nations status | 0.579 | 0.270 | 4.594 | 1.78a | [1.05–3.03] |
NESB | −0.112 | 0.278 | 0.162 | 0.89 | [0.52–1.54] |
Disability | 0.139 | 0.156 | 0.798 | 1.15 | [0.85–1.56] |
Verbal/emotional abuse | 2.323 | 0.206 | 127.146 | 10.21b | [6.82–15.28] |
Sexual abuse | 0.898 | 0.270 | 11.054 | 2.46b | [1.45–4.17] |
CEDV | 1.766 | 0.250 | 49.738 | 5.85b | [3.58–9.56] |
χ2 = (9, n = 4481) = 725.99, p < 0.001, Nagelkerke R2 = 0.40
NESB = non-English speaking background.Note. OR: Odds ratio; CI: Confidence interval.
∗∗p < 0.01.
a
p < .05.
b
p < .001.
A second binary logistic regression model was performed to assess the same set of variables while controlling for the effect of having been the victim of the other forms of child maltreatment (i.e., verbal/emotional abuse, child sexual abuse, and CEDV). The results showed that, among the participant characteristics of interest, when the rest of child maltreatment types were included in the model, only the association between First Nations status and physical abuse remained statistically significant (see Table 4). Essentially, the relationship between child physical abuse and gender identity, sexual orientation or disability was no longer statistically significant. However, all three forms of child maltreatment introduced in the second model were associated with experiencing child physical abuse. For example, young people who had experienced verbal/emotional abuse were 10.2 times more likely to also report experiences of child physical abuse than those who did not report experiences of verbal/emotional abuse. Further, the likelihood of reporting experiences of child physical abuse was almost six times higher for young people who reported CEDV compared to those who did not report CEDV. Finally, young people who had experienced child sexual abuse were more than twice as likely to report experiencing physical abuse during childhood.
4.4. Verbal/emotional maltreatment across participant characteristics and other types of maltreatment
The second set of binary logistic regression models examined, first, the relationship between participant characteristics and verbal/emotional abuse, and second, these associations when also including the rest of child maltreatment types (i.e., physical abuse, sexual abuse and CEDV) (see Table 5). The results indicated that girls, youth with diverse sexualities, young people with diverse gender identities, and participants who reported having a disability were more likely to also report experiences of verbal/emotional abuse. However, when the rest of child maltreatment types were included as predictors in the model, only the association between having a disability and verbal/emotional abuse remained statistically significant, while the odds ratio for gender, sexual orientation and gender identity were no longer statistically significant. Importantly, while reporting disability remained a significant predictor of experiencing verbal/emotional abuse when controlling for other abusive experiences, results from the second model indicated that young people with a disability are 1.5 times more likely to have experienced verbal/emotional abuse, almost halving the odds ratio of 2.7 from the previous model once other experiences of child maltreatment are introduced. In addition, all the other types of child maltreatment had an independent effect on being a victim of verbal/emotional abuse. In other words, experiencing child physical abuse, sexual abuse and CEDV all increased the likelihood of experiencing verbal/emotional abuse (see Table 5).
Table 5. Logistic regression models for verbal/emotional abuse.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 1 | |||||
Female | 0.851 | 0.107 | 63.581 | 2.34c | [1.90–2.89] |
Diverse genders | 0.703 | 0.176 | 16.010 | 2.02c | [1.43–1.85] |
Diverse sexualities | 0.411 | 0.090 | 20.816 | 1.51c | [1.26–1.80] |
First Nations status | 0.051 | 0.176 | 0.086 | 1.05 | [0.75–1.49] |
NESB | −0.096 | 0.154 | 0.387 | 0.91 | [0.67–1.23] |
Disability | 0.997 | 0.086 | 134.368 | 2.71c | [2.29–3.21] |
χ2 = (6, n = 4484) = 389.23, p <. 001, Nagelkerke R2 = 0.14
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 2 | |||||
Female | 0.384 | 0.138 | 7.781 | 1.47b | [1.12–1.92] |
Diverse genders | 0.487 | 0.237 | 4.218 | 1.63a | [1.02–2.59] |
Diverse sexualities | 0.150 | 0.121 | 1.520 | 1.16 | [0.92–1.47] |
First Nations status | −0.238 | 0.245 | 0.940 | 0.79 | [0.49–1.28] |
NESB | −0.094 | 0.202 | 0.216 | 0.91 | [0.61–1.35] |
Disability | 0.421 | 0.114 | 13.588 | 1.52c | [1.22–1.91] |
Physical abuse | 2.305 | 0.203 | 128.596 | 10.02c | [6.73–14.92] |
Sexual abuse | 1.032 | 0.314 | 10.778 | 2.81b | [1.52–5.20- |
CEDV | 3.395 | 0.130 | 680.131 | 29.83c | [23.11–38.50] |
χ2 = (9, n = 4481) = 1923.26, p < 0.001, Nagelkerke R2 = 0.58 NESB = non-English speaking background.Note. OR: Odds ratio; CI: Confidence interval.
a
p < .05.
b
p < .01.
c
p < .001.
4.5. Sexual abuse across participant characteristics and other types of child maltreatment
Results from the model examining participant characteristics show that several variables were independently associated with experiences of child sexual abuse (see Table 6). For example, girls, sexuality diverse young people, First Nations youth and participants with a disability had higher odds of being victims of child sexual abuse. Nevertheless, again, when assessing the same set of participant characteristics while also controlling for the effect of the rest of child maltreatment types, the only variables that remain independently associated with experiences of child sexual abuse were sexual orientation and First Nation status. Importantly, the odds ratios for these two predictors dropped substantially when compared to the previous model. Regarding the independent effect of the different types of child maltreatment on having experienced child sexual abuse, results indicated that young people who report experiencing physical abuse and verbal/emotional abuse) are more than twice as likely to also experience child sexual abuse. However, CEDV does not increase the odds of also experiencing child sexual abuse (see Table 6).
Table 6. Logistic regression models for sexual abuse.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 1 | |||||
Female | 0.977 | 0.317 | 9.505 | 2.66b | [1.43–4.94] |
Diverse genders | 0.371 | 0.351 | 1.117 | 1.45 | [0.73–2.88] |
Diverse sexualities | 0.818 | 0.226 | 13.161 | 2.27c | [1.46–3.53] |
First Nations status | 0.924 | 0.313 | 8.728 | 2.52b | [1.37–4.65] |
NESB | 0.151 | 0.379 | 0.159 | 1.16 | [0.55–2.48] |
Disability | 0.735 | 0.227 | 10.50 | 2.09b | [1.34–3.25] |
χ2 = (6, n = 4485) = 71.65, p < 0.001, Nagelkerke R2 = 0.08
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 2 | |||||
Female | 0.634 | 0.325 | 3.819 | 1.89 | [0.10–3.56] |
Diverse genders | 0.102 | 0.360 | 0.081 | 1.11 | [0.55–2.24] |
Diverse sexualities | 0.680 | 0.233 | 8.508 | 1.97b | [1.25–3.12] |
First Nations status | 0.820 | 0.333 | 6.049 | 2.27a | [1.18–4.36]
|
NESB | 0.158 | 0.387 | 0.168 | 1.17 | [0.55–2.50] |
Disability | 0.381 | 0.240 | 2.533 | 1.46 | [0.92–2.34] |
Physical abuse | 0.786 | 0.267 | 8.677 | 2.19b | [1.30–3.70] |
Verbal/emotional abuse | 1.009 | 0.311 | 10.526 | 2.74b | [1.49–5.05] |
CEDV | 0.496 | 0.311 | 2.549 | 1.64 | [0.89–3.02] |
χ2 = (9, n = 4481) = 134.73, p < 0.001, Nagelkerke R2 = 0.16
NESB = non-English speaking background.Note. OR: Odds ratio; CI: Confidence interval.
a
p < .05.
b
p < .01.
c
p < .001.
4.6. CEDV across participant characteristics and other types of child maltreatment
As with the rest of the child maltreatment types, a set of logistic regression models first examined the relationship between participant characteristics and CEDV, and second, these associations when also including the rest of child maltreatment types (i.e., physical abuse, verbal/emotional abuse, and sexual abuse). Results for both models are provided in Table 7. Regarding the first model, results showed that girls, young people with diverse sexualities and gender identities, and participants with a disability were more likely to report CEDV between adult carers. When the rest of the child maltreatment types were included in the second model, the association between being gender non-conforming was no longer statistically significant. Importantly, although the relationship between CEDV and the rest of the participant variables remained statistically significant, the odds ratios for these predictors dropped when compared to the previous model. For example, baseline data indicated that, compared to boys, girls were 2.3 times more likely to report CEDV, while this odds ratio decreased to 1.9 when controlling for other forms of child maltreatment. Although with a smaller decrease, the odds ratio for sexuality diverse individuals also dropped when accounting for the effect of other child maltreatment types (1.5 vs 1.3). Lastly, results from the second model showed that the effect of having a disability dropped from an odds ratio of 2.8 to an odds ratio of 2.2. Regarding the associations between CEDV and the other child maltreatment forms, participants reporting experiences of physical abuse and verbal/emotional abuse were more likely to also report CEDV. As in the model using sexual abuse as the dependent variable, the association between reporting experiences of sexual abuse and CEDV was not statistically significant.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 1 | |||||
Female | 0.840 | 0.086 | 96.550 | 2.32b | [1.96–2.74] |
Diverse genders | 0.615 | 0.175 | 12.308 | 1.85b | [1.31–2.61] |
Diverse sexualities | 0.377 | 0.079 | 22.910 | 1.46b | [1.25–1.70] |
First Nations status | 0.081 | 0.155 | 0.274 | 1.08 | [0.80–1.47] |
NESB | −0.099 | 0.127 | 0.610 | 0.91 | [0.71–1.16] |
Disability | 1.017 | 0.073 | 193.030 | 2.76b | [2.40–3.19] |
χ2 = (6, n = 4485) = 512.56, p < 0.001, Nagelkerke R2 = 0.15
Model 2 | |||||
|---|---|---|---|---|---|
Female | |||||
Diverse genders | |||||
Diverse sexualities | |||||
First Nations status | |||||
NESB | |||||
Disability | |||||
Physical abuse | |||||
Verbal/emotional abuse | |||||
Sexual abuse |
χ2 = (9, n = 4481) = 1905.66, p < 0.001, Nagelkerke R2 = 0.50
NESB = non-English speaking background.
Note. OR: Odds ratio; CI: Confidence interval.
∗∗p < 0.01.a p < .05.b p < .001.
4.7. Use of violence in the home across participant characteristics and other types of child maltreatment
The last set of binary logistic regression models examined, first, the relationship between participant characteristics and use of violence against at least one adult carer on a frequent basis, and second, these associations when also including the rest of child maltreatment types (i.e., physical abuse, sexual abuse and CEDV). Results for both models are provided in Table 8. Results from the first model indicated that young people with a disability were more likely to report the use of violence against an adult carer on a frequent basis. However, when the child maltreatment types were included in the second model, the relationship between gender identity and use of family violence was no longer statistically significant. Although the association between disability and use of family violence remained statistically significant, the effect of having a disability dropped from an odds ratio of 2.9 to 1.7 when accounting for the effect of the child maltreatment types. Regarding the associations between using violence at home and the child maltreatment types, participants reporting experiences of physical abuse, verbal/emotional abuse, and CEDV were more likely to also report using family violence in the home. However, the association between experiencing child sexual abuse and the use of family violence was not statistically significant.
Table 8. Logistic regression models for the use of violence in the home.
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 1 | |||||
Female | 0.549 | 0.166 | 10.969 | 1.73 | [1.25–2.40] |
Diverse genders | 0.197 | 0.264 | 0.553 | 1.22 | [0.73–2.04] |
Diverse sexualities | 0.201 | 0.141 | 2.055 | 1.22 | [0.93–1.61] |
First Nations status | 0.517 | 0.227 | 5.172 | 1.68 | [1.07–2.62] |
Non-English speaking background | −0.398 | 0.275 | 2.095 | 0.67 | [0.39–1.15] |
Disability | 1.065 | 0.139 | 58.762 | 2.90 | [2.21–3.81] |
χ2 = (6, n = 4485) = 121.89, p < 0.001, Nagelkerke R2 = 0.07
B | S.E. | Wald | OR | 95% CI | |
|---|---|---|---|---|---|
Model 2 | |||||
Female | 0.042 | 0.180 | 0.054 | 1.04 | [0.73–1.48] |
Diverse genders | −0.207 | 0.281 | 0.543 | 0.81 | [0.47–1.41] |
Diverse sexualities | −0.087 | 0.155 | 0.312 | 0.92 | [0.68–1.24] |
First Nations status | 0.480 | 0.257 | 3.475 | 1.62 | [0.98–2.68] |
Non-English speaking background | −0.366 | 0.289 | 1.605 | 0.69 | [0.39–1.22] |
Disability | −0.366 | 0.289 | 1.605 | 0.69 | [0.39–1.22] |
Physical abuse | 0.517 | 0.152 | 11.572 | 1.68b | [1.25–2.26] |
Verbal/emotional abuse | 0.357 | 0.173 | 4.245 | 1.43a | [1.02–2.01] |
Sexual abuse | 1.518 | 0.185 | 67.281 | 4.56b | [3.17–6.56] |
CEDV | 1.508 | 0.210 | 51.353 | 4.52b | [2.99–6.82] |
χ2 = (10, n = 4485) = 502.57, p < 0.001, Nagelkerke R2 = 0.29 NESB = non-English speaking background.
Note. OR: Odds ratio; CI: Confidence interval.
∗∗p < 0.01.
a
p < .05.
b
p < .001.
5. Discussion
The findings presented here add to the growing body of literature examining child maltreatment prevalence rates along with factors predicting an increased risk of child maltreatment experiences. In line with other recent Australian evidence (Mathews, Pacella, et al., 2023), our findings show that the risk of child maltreatment in Australia is gendered. While higher rates of sexual abuse among girls compared to boys have been widely documented (Barth et al., 2013; Mathews, Pacella, et al., 2023; Stoltenborgh et al., 2015), the gendered nature of other forms of child maltreatment has received less attention. Cisgender females in our study were more likely to experience all four types of child maltreatment and this remained applicable when controlling for other characteristics, including having a diverse sexuality and gender identity, having a disability, and reporting diverse cultural backgrounds. These findings highlight the ongoing need for gender transformative education and holistic primary prevention strategies to address the underlying individual, community and societal risk factors associated with the occurrence of child maltreatment (Austin et al., 2020; Mathews, Thomas, & Scott, 2023) and further reduce the occurrence along with well-established short- and long-term effects of all forms of child maltreatment (Carr et al., 2020; Kiesel et al., 2016; Lawrence et al., 2023).
The findings presented here illuminate the importance of the role of multi-type child maltreatment experiences in understanding the increased risk of specific types of victimisation experiences during childhood. Several participant characteristics predicted a higher risk of experiencing physical forms of frequent child maltreatment. Here, being female, identifying as gender diverse, identifying as First Nations and living with a disability all increased the risk of experiencing physical maltreatment as a pattern of abuse during childhood, which is in line with findings from wider research (Craig et al., 2020; Mathews, Pacella, et al., 2023; Segal et al., 2019; Sullivan & Knutson, 2000).
However, once we controlled for the role of other child maltreatment experiences, only First Nations status and other child maltreatment experiences remained significant, with the latter being the strongest predictor of physical abuse. Specifically, while First Nations young people were 1.7 times more likely to experience physical abuse than those without First Nations status, participants reporting CEDV between parents or carers and those who experienced patterns of verbal/emotional abuse had a higher likelihood of also experiencing physical abuse, 6-fold and 10- fold, respectively. It is important to note that young people's cultural identity should not be viewed as directly increasing their risk for child maltreatment. Rather, these findings highlight how structural and systemic discrimination may amplify the likelihood of interpersonal and community-level risk factors for child maltreatment within First Nations families. These factors include intergenerational trauma, social and economic disadvantage, limited access to social support systems, exposure to substance misuse or inadequate housing (Austin et al., 2020; AIFS, 2020).
Reporting experiences of verbal/emotional abuse was also primarily predicted by other forms of child maltreatment, including a 10-fold higher risk among young people who also reported experiences of physical abuse and a nearly 30-fold higher risk among young people having grown up with CEDV. Similarly, experiences of CEDV were strongly predicted by experiences of verbal/emotional abuse, increasing 30-fold when co-occurring. These findings highlight the significant overlap of different forms of child maltreatment experiences, especially with CEDV significantly increasing the risk of children and young people also experiencing verbal/emotional and physical abuse directed at them and vice versa. The high co-occurrence rates of CEDV and verbal/emotional abuse are of particular concern when considering that the recent Australian Child Maltreatment Study (ACMS) alerted to non-physical (emotional) abuse being the second biggest contributor to health risk behaviours (including self-harm, suicidal behaviours and problematic substance use), after child sexual abuse (Haslam, et al., 2023, 2023b).
While the cross-sectional nature of the data does not allow for an identification of whether certain child maltreatment experiences precede others, findings highlight the significant risk of children experiencing multi-type child maltreatment, where more than one form of child maltreatment is present. The exception to this is child sexual abuse, which was not predicted by the presence of CEDV, suggesting it is a distinct form of parental/carer offending not necessarily linked to the presence of other forms of child maltreatment in the home.
Finally, findings from this research regarding the intergenerational transmission of domestic and family violence reveal a strong correlation between young people's experiences and the use of family violence in the home. Specifically, experiences of verbal/emotional abuse and CEDV were significantly associated with a higher risk of young people's use of family violence in the home. While experiences of physical abuse also doubled the risk of using family violence in the home, experiences of verbal/emotional abuse and CEDV increased the risk of using family violence towards other family members more than four-fold. This highlights the detrimental effects of non-physical forms of child maltreatment on young people's social and emotional development and functioning in the context of interpersonal relationships also observed in other child maltreatment studies (Carr et al., 2020; Lawrence et al., 2023). While individual characteristics also predicted an increased risk in young people's use of family violence in the home, with cisgender girls and young people living with a disability more likely to report use of violence in the home, this risk reduced substantially once we controlled for childhood experiences of maltreatment. These findings are not surprising as all other models showed that females and children with a disability were significantly more likely to report all forms of child maltreatment. In other words, the use of family violence in the home is predicted by experiences of CEDV in the home, with females and young people with a disability being significantly more likely to experience childhood victimisation across the different maltreatment categories.
6. Conclusion and implications for practice
One in six children reported at least two forms of child maltreatment in our study. While a number of characteristics increased the risk of individual types of child maltreatment prior to controlling for the risk of multi-type maltreatment, only gender and disability status were consistently significant across all forms of child maltreatment, prior to controlling for multi-type maltreatment. The gendered nature of child maltreatment experiences identified in this study highlights the need for a gendered lens in primary prevention and early intervention efforts. Here, it is critical to recognise and respond to child maltreatment risk across the gender identity spectrum while maintaining an awareness that female children are disproportionately affected by all forms of child maltreatment and related adverse social and emotional outcomes. Further, the observation that young people living with a disability, including cognitive disabilities and presentations of neurodiversity, were significantly more likely to experience all forms of child maltreatment highlights opportunities for disability services to identify and respond more effectively to child maltreatment and related risk factors. This requires a multi-pronged approach across disability services, including the National Disability Insurance Scheme (NDIS), to effectively identify and create suitable referral and support pathways for parents/carers within vulnerable families that present with multiple risk factors associated with child maltreatment. Further, it raises the need for trauma-informed practice, to promote the screening for the occurrence of child maltreatment at the intersection with disability/neurodiversity assessments and diagnoses, to identify opportunities for parent- and child-centred interventions and facilitate referral pathways for child- and youth-centred recovery and healing support.
Beyond individual characteristics, findings regarding the co-occurrence of multiple types of child maltreatment raise important implications for policy and practice in child and family welfare, youth support, specialist DV and healthcare services. While certain individual characteristics may increase children and young people's risk of experiencing child maltreatment overall, our findings suggest that detection and intervention opportunities across different service settings, including health, education, child, youth and family welfare and DV specialist services, should focus on screening for and recognising children's experiences of individual types of maltreatment, including but not limited to physical forms of child maltreatment. Non-physical forms of maltreatment – including coercive and controlling behaviours - can be harder to detect and related risk of harm may be underestimated. However, our findings show that experiencing verbal/emotional abuse significantly increases the risk of experiencing physical abuse and CEDV, warranting a focus on early detection of and responses to child maltreatment beyond a focus on individual incidents of physical harm.
The growing recognition of children experiencing DV between parents or carers as victim-survivors in their own right (DSS, 2022; see also Meyer & Fitz-Gibbon, 2022) and the increased risk of multi-type child maltreatment established in our study, highlights the critical nature of screening for and risk assessment of parental/carer DV and other types of child maltreatment to ensure holistic and trauma-informed responses to affected families, children and young people. This must include routine screening and risk identification for DV and other forms of child maltreatment in child and family support services, child and youth (mental) health services along with consistent professional development in trauma-informed practice to identify early indicators of risk of harm in educational and healthcare settings.
Our study findings around the increased risk of young people's use of violence in the home when affected by CEDV and other forms of child maltreatment also highlight the critical importance of early detection of and response to CEDV and other forms of child maltreatment as a form of prevention and early intervention. This requires moving beyond current crisis intervention models that disrupt immediate exposure to harm and translate into child-centric and youth-centred recovery support that fosters social and emotional wellbeing and reduces the risk of intergenerational transmission of family violence and other adverse outcomes. Our findings around the intersection of children and young people's experiences and use of violence in the home also clearly highlight the need for early interventions and recovery support when child maltreatment is detected to prevent the intergenerational transmission of violence in families.
Beyond the importance of identifying and responding to occurring forms of child maltreatment and related risk, the findings presented here have larger implications for holistic investment in gender transformative reforms, including primary prevention of and early interventions for vulnerable families at an individual, community and societal level to address the multifaceted risk factors associated with child maltreatment and enhance family and community wellbeing (Austin et al., 2020; Mathews, Thomas, & Scott, 2023). This could include further support for, and elevation of gender transformative prevention programs such as those that target healthy relationships, gender equality, healthy masculinities and models of fathering as well as whole of community approaches to progressing gender equality and advancing the elimination of violence against women and children (see, for example, Fergus, 2024).
Finally, this research, including the broader study from which this article draws, was derived through the voices of children and young people. We recognise that this is one of several research studies in Australia which have sought to support greater input of children into the generation of research evidence used to inform child, youth and family welfare policy and practice (see also Fitz-Gibbon et al., 2023). Building on national and international evidence (cf. Case et al., 2020; Grace et al., 2024; Grove et al., 2024), children and young people's voices, including the voices of gender and sexuality diverse young people and young people living with a disability, should form part of future policy and practice design addressing the risk and occurrence of child maltreatment to ensure service delivery reflects the experiences and needs of child victim-survivors and their families.