Introduction

Coercive control is increasingly recognised as a public health and safety issue in urgent need of response in Australia (Australia's National Research Organisation for Women's Safety [ANROWS], 2021) and globally (Nevala, 2017). As the concept of coercive control, and its relationship to Intimate Partner Violence (IPV), has become better known and understood, policy and legal reforms related to coercive control response and prevention have attracted increasing debate and rapid change. Some states and territories across Australia have taken steps towards criminalising coercive control.⁴ Although, at the time of writing, there is not a consistent legislative response in Australia, the Commonwealth government has published the draft National Principles to Address Coercive Control.

Effective policy and practice efforts to respond to and prevent coercive control should be informed by high-quality research evidence. However, coercive control is a complex issue and, because the concept is relatively recent, the research evidence in this space is still emerging. AIFS' consultations with stakeholders in the child and family sector identified a desire for more information and evidence about coercive control (outlined further in the Method section). This paper aims to synthesise what is currently known about coercive control victimisation to inform policy and practice. Specifically, this paper aims to synthesise the evidence about how common coercive control victimisation is, as well as the risk factors and impacts of coercive control victimisation.

Synthesising research evidence involves bringing together the findings of multiple studies to describe the most reliable information and to consider what is known and unknown about a topic (Gough et al., 2020). Evidence syntheses aim to provide a holistic overview of the research literature, to inform practice, policy decision making (Gough et al., 2020) and future research. At the time of writing, the peer-reviewed quantitative research evidence on coercive control victimisation and how it applies to the Australian context had not previously been synthesised.

A note on terminology: in this paper, we use the term 'victim-survivor'. A victim-survivor is someone who has experienced coercive control victimisation (i.e. been the target of coercive control behaviours by a current or ex intimate partner). The term victim-survivor is used to acknowledge 'the ongoing effects and harm caused by abuse and violence as well as honouring the strength and resilience of people with lived experience' (Victorian Government, 2022).

Structure of this paper

This paper has 8 sections:

• Glossary: provides definitions of key terms and acronyms used in this paper.
• Introduction: explains the aims and focus of this paper, defines coercive control and summarises qualitative insights from Australian literature on what coercive control looks like.
• Why it is important to understand coercive control victimisation: provides a rationale for why it is important to understand coercive control victimisation and the consequences for victim-survivors of a lack of understanding.
• The nature of the evidence base: summarises the characteristics of the studies included in the rapid review and key reflections on the quality of research conducted to date.
• How common is coercive control victimisation? provides a summary of the main insights about the frequency of coercive control victimisation in the general population and clinical samples, as well as implications for practice, research and policy, together with detail on the findings of studies assessing prevalence.
• Risk factors associated with coercive control victimisation: provides a summary of the main risk factors for coercive control victimisation, and implications for practice, research and policy. The rest of the chapter provides more detail on the findings of studies assessing risk factors.
• Impacts associated with coercive control victimisation: provides a summary of the main impacts of coercive control victimisation and implications for practice, research and policy. The rest of the chapter provides more detail on the findings of studies assessing impacts.
• Method: How the rapid literature review was conducted: describes the method used to conduct the rapid literature review. It also describes the stakeholder engagement that informed the rapid literature review as well as potential limitations to be considered in future literature reviews on coercive control victimisation.

Definition of coercive control used in this paper

Even though the need to address coercive control is becoming more widely recognised, there is no single, consistently used definition of coercive control. A previous literature review found that there were 22 different measures and definitions of coercive control used by researchers (Hamberger et al., 2017). A particular point of contention within the literature is whether coercive control is a subtype of IPV or if it is the broader context within which IPV occurs.

Early conceptualisations of coercive control have emphasised the distinction between ongoing patterns of violence and situational violence. Of particular relevance here is the framework developed by Johnson (2008) outlining 4 distinct types of IPV violence use and the extent to which control is a motivating factor. In this framework, 'intimate terrorism' can be considered one of the early conceptual origins of coercive control. It is a relationship where one individual uses violence to control their partner. The type of control motivation here is not a situational one-off, goal-oriented violence. Rather:

the control sought in intimate terrorism is general and long term. Although each particular act of intimate violence may appear to have any number of short-term, specific goals, it is embedded in a larger pattern of power and control that permeates the relationship. (Johnson, 2008, p 13)

'Violent resistance' is when one individual uses violence in response to their partner's intimate terrorism. 'Mutual violent control' is a relationship type where both individuals use violence to control one another. Finally, 'situational couple violence' is where one or both individuals use violence but not to gain long-term and general control over the other. Except for situational couple violence, all the other 3 types of IPV relationships in Johnson's framework include coercive control enacted by at least one of the individuals within the relationship. Since the original conceptualisations by Johnson (2008) and Stark (2007) (described below), the conceptualisation of coercive control has continued to be refined as our understanding of this issue grows (Stark & Hester, 2019).

In this paper, we use the definition of coercive control applied by ANROWS.⁵ This ensures that this paper is consistent with and comparable to other work across Australia on coercive control. ANROWS' definition of coercive control is also consistent with that proposed by Stark (2007; described in the following paragraphs) who originally developed the concept of 'coercive control' and Johnson's (2008) conceptualisation of intimate terrorism.

ANROWS (2021) defines coercive control as behaviours or strategies (including physical and non-physical violence) used to control a victim-survivor and make them feel inferior to, and dependent on, the perpetrator enacting these behaviours or strategies. Coercive control involves ongoing and repetitive strategies that add up over time to impact the victim-survivor's independence, freedom and equality (ANROWS, 2021). The perpetrator's dominance and control over the victim-survivor has the potential to impact 'every aspect of her life, effectively removing her personhood' (ANROWS, 2021, p 1). ANROWS describe coercive control as the broader situation within which ongoing intimate partner violence (IPV) occurs (ANROWS, 2021).

Coercive control is the ongoing and repetitive use of behaviours or strategies (including physical and non-physical violence) to control a current or ex-intimate partner (i.e. victim-survivor) and make them feel inferior to, and dependent on, the perpetrator. (ANROWS, 2021)

Essential to this definition of coercive control are 2 components: (a) it includes physical and/or non-physical violence and (b) the physical and/or non-physical violence is ongoing. Viewing the victim-survivor's abuse as incident specific (i.e. in isolation from other incidences of violence or abuse perpetrated by the same person towards the victim-survivor) or viewing physical injury or physical trauma as a more serious form of abuse 'disaggregates, trivialises, normalises or renders invisible' the ongoing oppression they experience (Stark, 2009, p 1510).

ANROWS (2021) describes coercive control as a kind of power that (almost exclusively) male perpetrators use against (almost exclusively) female victim-survivors. This definition is in part supported by national Australian data suggesting that women have greater likelihood of experiencing family, domestic and sexual violence than men (Australian Institute of Health and Welfare [AIHW], 2018). However, not all organisations or researchers describe coercive control in this way; some examine coercive control victimisation among men as well as women. To account for this broader conceptualisation, this review did not exclude any studies based on participant gender or sexuality.

ANROWS highlight the following list of strategies perpetrators may use to enact coercive control (2022, p 1):

• physical, sexual, verbal and/or emotional abuse
• psychologically controlling acts
• deprivation of resources and other forms of financial abuse
• social isolation
• using systems to cause harm (e.g. the legal system)
• stalking and harassment
• deprivation of liberty
• intimidation.

The list above contains some possible strategies and behaviours used by perpetrators but it is not a comprehensive list. The perpetrator is likely to use any strategies they believe will be effective in controlling the specific victim-survivor. For example, there is emerging research indicating that perpetrators may threaten to or actually harm the family pet in order to control the victim-survivor's behaviour (Collins et al., 2018; Muri et al., 2022).

Consistent with the strategies outlined by ANROWS listed above, Australian qualitative studies have highlighted some of the common behaviours and strategies used by perpetrators of coercive control as reported by victim-survivors, including:

• monitoring or surveillance of the victim-survivor's location (Moulding et al., 2021)
• stalking the victim-survivor (Eriksson et al., 2022)
• becoming jealous of the victim-survivor spending time with other people (e.g. accusing them of having affairs) (Eriksson et al., 2022)
• restricting the victim-survivor's relationships and social support (Moulding et al., 2021)
• being emotionally, physically or sexually abusive (Eriksson et al., 2022; Pitman, 2017)
• thinking they are superior, entitled, dominant or have ownership over the victim-survivor (Eriksson et al., 2022; Pitman, 2017)
• controlling or criticising the victim-survivor's appearance or weight (Eriksson et al., 2022; Moulding et al., 2021)
• avoiding, obstructing or retaliating in response to the victim-survivor trying to negotiate or maintain boundaries within the relationship (e.g. refusing to communicate, threating to overdose on a substance, or using or threatening physical or sexual violence) (Pitman, 2017)
• establishing double standards in the relationship in terms of rights, reciprocity and accountability (Pitman, 2017)
• placing contradictory expectations on the victim-survivor (e.g. demanding the victim-survivor focus on them and adapt to their needs but not be dependent on them) (Pitman, 2017).

Why it is important to understand coercive control victimisation

This chapter outlines some of the key reasons why it is important to have a greater understanding of common coercive control victimisation and what risk factors and impacts are associated with coercive control victimisation.⁶ In particular, this chapter highlights some of the reported consequences for victim-survivors of:

• a lack of understanding in police and legal responses to coercive control
• community misconceptions of coercive control.

Police and legal responses to coercive control

A lack of understanding of coercive control within the legal system can lead to poor or unjust outcomes for victim-survivors (Boxall et al., 2020; Nancarrow et al., 2020; Ulbrick & Jago, 2018) and impairs the level of protection provided to victim-survivors. Although police are not DFV specialists, they are important frontline workers who can potentially identify patterns of coercive control in callouts related to DFV (Myhill & Hohl, 2019). A study of police reports of domestic violence cases in the United Kingdom found that the reports often referenced coercive control behaviours (Myhill & Hohl, 2019), such as the victim-survivor feeling isolated or frightened and the perpetrator controlling the victim-survivor and being excessively jealous.

If these signs of coercive control are missed or not understood by police, this may result in police viewing the violence on an incident-by-incident basis, as opposed to an ongoing pattern of abuse. This can result in missed opportunities to provide victim-survivors with appropriate referrals and support. It can also mean that: (a) perpetrators are not held accountable for ongoing violent patterns of behaviour, and (b) the cumulative impact on victim-survivors is not adequately acknowledged (Stark & Hester, 2019).

If police view violence on an incident-by-incident basis, this is particularly problematic when victim-survivors use violence to defend themselves from a violent partner (described earlier as violent resistance) and are misidentified as a sole or co-perpetrator (Boxall et al., 2020; Nancarrow et al., 2020; Ulbrick & Jago, 2018). In this kind of situation:

• It may be difficult for police and the broader legal system to identify the person in the relationship most in need of legal protection (Nancarrow et al., 2020).
• Victim-survivors may be unjustly criminalised (Nancarrow et al., 2020).

Victim-survivors can also be harmed by perpetrators using the legal system to enact coercive control. Douglas (2018) interviewed 65 Australian women engaged with the legal system because of intimate partner violence. The women described strategies used by their violent partners to continue controlling and having access to the victim-survivor after separation through legal processes, such as requesting numerous last-minute adjournments of court cases and extending mediation time.

Community misconceptions of coercive control

Challenging community-level misconceptions and increasing awareness of coercive control are important to preventing and responding to coercive control. A nationally representative study of 17,500 Australians aged 16+ years found that approximately 1 in 3 Australians believe that women who do not leave an abusive relationship are partially responsible for the violence they experience (Webster et al., 2018). About 1 in 6 Australians do not believe it is difficult for a woman to leave an abusive relationship (Webster et al., 2018).⁷

These misconceptions do not take into consideration the complexity of coercive control victimisation and place the responsibility of the violence on the victim-survivor rather than the perpetrator. In reality, the periods of time without active violence required to plan how to leave safely or take action on such a plan may not exist (ANROWS, 2021). Perpetrators use a wide variety of strategies to foster a sense of entrapment, erode victim-survivors sense of autonomy and make it very challenging (if not impossible) for victim-survivors to safely leave (ANROWS, 2021). Additionally, research indicates that trying to end a relationship with a violent partner can escalate the violence, including increased risk of intimate partner homicide, as the violent partner attempts to maintain control (Monckton Smith, 2020).

Greater community-level understanding of coercive control could enhance the ability of victim-survivors to seek support (Webster et al., 2018).

The nature of the evidence base

The following 3 chapters of this paper provide a synthesis of research findings about how common coercive control victimisation is, as well as associated risk factors and impacts. The literature synthesised in the following chapters is different to that described in the previous chapter, which served to provide context for the chapters to follow. The 13 research studies included in the following synthesis chapters were identified through a rapid review of the coercive control victimisation literature. The rapid review method is described later in the chapter, see Method: How the rapid literature review was conducted. This chapter explains the nature of the literature on how common coercive control victimisation is, and the risk factors and impacts, as identified through the rapid review method (described below as 'the sample').

Characteristics of studies sampled in the review

Of the 13 studies included in the sample, 7 studies (54%) were published during or after 2017. Roughly half of the studies were from Canada (n = 6, 46%). Four studies (31%) were from the United Kingdom. Only 3 studies (23%) were conducted in Australia.⁸ Appendix A contains a table listing the individual studies and highlighting relevant characteristics of each. This section provides an overview of those relevant characteristics.

Most studies (11, 85%) used the cross-sectional survey method for data collection (data collected at one point in time) and one study had a longitudinal survey methodology (using data gathered over several years). All but one study analysed self-reported data from victim-survivors, with the exception analysing data from police created reports about family and domestic violence incidents where stalking had occurred.

The studies had sample sizes ranging from 40 to 42,000 participants. Eight studies (62%) had samples of over 1,000 participants. Most studies described participants in terms of their gender, parental or marital/cohabitation status and/or service access. Participants were predominantly adult women, with 2 studies also including underage men and women (15+ and 16+ years old) (Myhill, 2015; Pedersen et al., 2013). Only around one-third of the studies (4, 31%) included male participants in addition to women (Frankland & Brown, 2014; Mayshak et al., 2020; Myhill, 2015). In 2 studies, the sample was exclusively mothers (Broughton & Ford-Gilboe, 2017; Skafida et al., 2022). Additionally, 2 studies only included women who were married or cohabitating (Kaukinen & Powers, 2015; Pedersen et al., 2013), and one included only women who had separated from an abusive partner (Broughton & Ford-Gilboe, 2017). Only one study specifically explored the experiences of same-sex attracted adults (Frankland & Brown, 2014). Regarding service access, 2 studies involved current or previous residents of women's or homeless shelters (Ellis et al., 2021; Levine & Fritz, 2016), and one study had recruited clients of domestic violence services (Sharp-Jeffs et al., 2018).

Bendlin and Sheridan (2019) was the only study to analyse police reports. All other studies analysed self-report data from victim-survivors. Bendlin and Sheridan (2019) analysed a dataset of DFV reports completed by police in Perth, Australia, between 2013 and 2017 (n = 9,884 reports). The researchers included only those reports where there had been stalking. The authors of this study did not report participant characteristics beyond 'reports of domestic violence'.

Most studies (70%) adopted customised measurements of coercive control victimisation (i.e. measures constructed by the authors for the purpose of the study). Standardised measures were used by only 4 studies. These included the Women's Experiences of Battering (WEB; Broughton & Ford-Gilboe, 2017; Ford-Gilboe et al., 2016), the Revised Controlling Behaviours Scale (CBS-R; Levine & Fritz, 2016) and the Coercive Control UK Scale (CCUK; Sharp-Jeffs et al., 2018).

The quality of evidence on coercive control victimisation

Research on coercive control victimisation is still at a relatively early stage of development and this limits what can be said about its frequency or impact, especially in Australia. This section of the paper outlines some considerations relating to the nature and quality of the research to date on coercive control victimisation. This section covers: (a) issues with study populations, (b) the way coercive control victimisation is defined and measured, and (c) research design.

Participants in the sampled studies

There is at present little research specific to the Australian context. Only a small proportion of the studies explored for this paper considered Australian experiences (3 out of 13 studies, 23%). There is a need for foundational knowledge about frequency, risk factors and impacts of coercive control victimisation for various subpopulations in Australia. We currently know little about the unique experiences of:

• Aboriginal and/or Torres Strait Islander peoples
• people with disability
• LGBTQIA+ communities
• culturally and linguistically diverse (CALD) communities
• people in older age groups (65+ years)
• children and young people where there is coercive control between their parents
• intersectional experiences across more than one of the above.

To date, data have typically been collected from either clinical samples, where individuals are targeted for recruitment because they have experienced IPV, or from the general population. There are strengths and weaknesses associated with both kinds of studies.

Findings from clinical samples are useful because they can tell us about coercive control victimisation among those who experience an increased risk, such as individuals experiencing IPV. This can aid our understanding of different forms of abuse. However, we can expect some of the findings among this group to differ from that of the general population. For example, the prevalence rate of coercive control victimisation is likely to be higher in clinical samples than in the general population.

Myhill (2015, p 360) notes that clinical samples 'have been useful in exploring the dynamics of abuse and prompting the development of typologies to differentiate between different forms of abuse,' but to establish population-level estimates of coercive control we need to use representative national samples (Myhill, 2015). Additionally, sample sizes in clinical sample studies tend to be small, making it difficult to know how generalisable the findings are to other people accessing support services as a result of IPV (the clinical population). Because the clinical samples tend to be convenience or snowball samples, it is also difficult to know whether there is systematic bias whereby certain service users were less likely to choose or have the opportunity to participate in the studies.

Conversely, general population studies can provide insights into how widespread coercive control is. The general population studies covered in this review did recruit larger numbers of participants but often provided little information about the representativeness of the sample or how participants were selected. Only 4 of the 13 studies provided information about participant response rates (Ellis et al., 2021; Kaukinen & Powers, 2015; Myhill, 2015; Skafida et al., 2022). Response rates are important because they indicate what proportion of people who were given the chance to participate decided not to. Response rate information for the sampled studies can be found in Appendix A.

Victim-survivors of IPV may be under-represented in general population samples for a range of reasons including lack of independence from the perpetrator and social isolation (Kaukinen & Powers, 2015). Participants in general population studies may also be less likely than those in clinical samples to disclose violence (Kaukinen & Powers, 2015). For these reasons, the reported prevalence rates may underrepresent the actual prevalence of coercive control within the broader community (Kaukinen & Powers, 2015).

Defining and measuring coercive control

Summarising the research evidence on coercive control victimisation is also hampered by the debate among researchers and experts on what exactly coercive control victimisation is. A literature review by Hamberger and colleagues (2017) demonstrated this point well when it identified 22 different measures and definitions of coercive control victimisation but none that had been frequently used. This lack of consistency was also evident in the studies sampled in the present review:

• Some emphasised the gendered nature of coercive control by only sampling women victim-survivors who had been in IPV relationships with men, whereas others sampled men and women in opposite- and same-sex relationships.
• Some used archive data from many years earlier that focused on broader conceptualisations of IPV or DFV and selected items to best fit the existing data to a study exploring coercive control. In these cases, coercive control victimisation tended to be conceptualised as the experience of psychological or emotional abuse behaviours in broader scales of DFV (i.e. not physical or sexual abuse). However, in contrast, the foundational work by Stark and colleagues (Stark, 2007; Stark & Hester, 2019) and later by ANROWS (2021) conceptualises coercive control as the context in which IPV occurs, meaning that physical and sexual violence are considered behaviours that can be used to enact coercive control (when not occurring in isolated incidence based abuse).
• Most studies (70%) used customised measurements of coercive control victimisation (i.e. measures constructed by the authors for the purpose of the study). Standardised and psychometrically validated measures were used by only 4 of the 13 sampled studies.

The significant differences between definitions and study populations means that findings across studies are not easily comparable. Among other consequences, this means it is uncertain which risk factors and impacts are most robust and deserving of further investigation. The lack of consistent definitions may also make it more challenging to develop practice responses.

Research design

The studies sampled in this review provide important foundational insights about coercive control victimisation. This section highlights some limitations with the research designs adopted in the studies. It is important to note that all research studies have limitations and many of these are the result of the practical barriers that come with studying a complex area of human experience.

Most of the reviewed studies used a type of data collection method called a cross-sectional survey. In these studies, participants were asked about their experiences and attitudes at a single point in time, with the analysis inferring the relationship between groups of variables. For example, higher levels of coercive control victimisation can be associated with specific outcome variables such as higher levels of negative mental health symptoms. Cross-sectional research cannot establish the direction of this relationship - that is, whether coercive control causes mental health problems or whether mental health problems create a susceptibility to coercive control - or demonstrate change over time. Longitudinal methods are required for those insights.

Some other areas for consideration include:

• the large differences in sample sizes, ranging from 40 to 42,000 participants
• the absence of demographic and other relevant social variables in analyses that might assist in greater understanding of how findings may differ between various subpopulations
• a broad range of risk factors and impact measures across the various studies with little consistency of measures or replication of findings
• a tendency to focus on individual-level factors associated with victimisation, without equal consideration of broader institutional or societal factors (such as gendered drivers of inequality that reinforce violence against women).⁹

How common is coercive control victimisation?

The main insights about the frequency of coercive control victimisation and implications for practice, research and policy are set out in the key messages box. The rest of the chapter provides more detail on the findings of studies assessing frequency.

What the research tells us about the frequency of coercive control victimisation

This section summarises the frequency rates of coercive control victimisation reported across studies. The focus of this paper is on the experiences of victim-survivors, not on the frequency of perpetration behaviours. However, the limitations described in The nature of the evidence base chapter makes it challenging to accurately estimate the prevalence of coercive control victimisation.

The frequency findings for clinical and general population samples covered in this review are summarised separately because the findings are not directly comparable, and each comes with its own strengths and weaknesses. As noted above, the ways in which coercive control is conceptualised and measured in different studies affect reported frequency rates. Because of these challenges in measuring coercive control victimisation, the frequency figures discussed in the following discussion are best interpreted as a rough estimate of the possible range of frequency within clinical and general population samples.

Frequency in general population samples

Findings about coercive control victimisation from general population studies are mixed and not based on representative samples. This makes them difficult to interpret. The reviewed studies reported frequency rates between 7.5% and 28% in samples taken from the general population in the countries where studies were undertaken. The frequency rates for studies looking at international women and mothers ranged from 7.5% to 15% (Kaukinen & Powers, 2015; Nevala, 2017; Skafida et al., 2022). US data were excluded from the sample of this review but Kaukinen and Powers (2015) reported 17% frequency rate in a US general population sample of married or cohabiting women.

In 1 of the 2 Australian studies, Mayshak and colleagues (2020) conducted an online survey of 1,009 Australian adults aged 18-79 years, including male and female participants. Of the 989 participants who responded to the coercive control questions in the survey, 28% (n = 276) of all participants (male and female combined) had experienced coercive control victimisation over the past 12 months from a current or most recent partner.

Frankland and Brown (2014) sampled same-sex attracted adults (n = 184), presenting them with 15 different types of controlling behaviours and asking how many of these behaviours they engaged in (M = 2.7, SD = 3.2) and how many their partner engaged in (M = 3.3, SD = 3.7).¹⁰ The most frequently reported behaviours related to dominance and emotional control. The sample was then separated in to low and high levels of coercive control. The frequency of high control was approximately 24% in participants who self-reported about their own behaviours and those of their partner.

It is noteworthy that these 2 Australian studies have similar frequency rates, both of which appear elevated compared to the international studies summarised above. However, the small number of studies conducted in Australia, and limitations related to measurement and sampling, make it difficult to assess the true prevalence of coercive control victimisation in the Australian general population. For example, another Australian study reporting women's prevalence of coercive control victimisation was identified in the grey literature. That study reported a much lower 11% prevalence (Boxall & Morgan, 2021a), which is more aligned with the international studies reviewed.

Frankland and Brown's (2014) study was the only one to assess the frequency of Johnson's 4 types of IPV violence (intimate terrorism, violent resistance, mutual violent control and situational couple violence). Frankland and Brown (2014) also identified a fifth type of violence to identify relationships where control is present but physical violence is not: non-violent control. Participants provided information about their own behaviours and that of their partner. This information was then used to group the couples into 1 of these 5 categories or into a no violence and no control group. More than half of the sample was categorised as being in a no violence and no control group relationship (58.2%). The most common type of IPV relationship was situational couple violence (participants 13.0%, partners 14.7%), followed by mutual violent control (12.5% for both). Non-violent control (participants 7.1%, partners 5.4%) and intimate terrorism (participants 4.3%, partners 6.5%) were similar in frequency. Least common was violent resistance (2.7% for both).

Frequency in clinical samples

Four of the six clinical sample studies used a cross-sectional survey design (Broughton & Ford-Gilboe, 2017; Levine & Fritz, 2016; Myhill, 2015) and one used a longitudinal design (Sharp-Jeffs et al., 2018). Participants were women who had accessed domestic violence services, accessed women's shelters, engaged in custody disputes with ex-partners or separated from an abusive partner. The sample sizes ranged from 40 to 2,335 and each study measured coercive control victimisation differently.

Frequency rates of coercive control victimisation among women in clinical samples ranged between 4.4% and 100% (Broughton & Ford-Gilboe, 2017; Ellis et al., 2021; Myhill, 2015; Pedersen et al., 2013; Sharp-Jeffs et al., 2018). Myhill (2015) also provided a frequency rate of 6% for men.¹¹ In the sixth study, Bendlin and Sheridan (2019) analysed a dataset of family violence reports completed by police in Perth, Australia (n = 9,884). The authors reported a frequency rate of 43.5% (n = 4,304) in family violence police reports that included stalking.¹²

Risk factors associated with coercive control victimisation

The main risk factors for coercive control victimisation and implications for practice, research and policy are set out in the key messages box. The rest of the chapter provides more detail on the findings of studies assessing risk factors.

What the research tells us about risk factors associated with coercive control victimisation

The studies reviewed for this paper identified numerous risk factors for, or characteristics associated with, coercive control victimisation. These varied considerably between studies. We have listed all statistically significant (highlighted in blue) and non-significant findings (highlighted in pink) in the reviewed studies in Table 1. The common cut off for determining statistical significance in the reviewed studies was a p value of <0.05; that is, where there was less than 5% chance that a finding of difference between compared groups, or a relationship between coercive control victimisation and a specific risk factor, was the result of chance. We have included non-significant findings in this table because reporting only significant findings makes it difficult for those working in this area to know the full range of variables that previous research has focused on or the extent to which there is inconsistency in findings.

Five of the factors included in Table 1 are theoretical risk factors for perpetration, not victimisation: psychopathy, Machiavellianism, grandiose narcissism, vulnerable narcissism and sadism.¹³ The study that included these factors reported scores for both perpetrators and victim-survivors (Mayshak et al., 2020). Because the research base on coercive control victimisation is currently small, we decided to include findings about these theoretical perpetration risk factors and likelihood of victimisation to present a more complete picture of existing research evidence.

Note: blue = statistically significant, pink = not statistically significant.

The evidence about risk factors for coercive control victimisation is inconclusive. Although several risk factors for coercive control victimisation have been identified, most of these were only reported in one study. Where risk factors were assessed in more than one study, the findings about their significance were inconsistent. The risk factors with inconsistent findings across studies were:

• being female, compared to male (Mayshak et al., 2020; Myhill, 2015).
  • As described earlier, ANROWS and Stark (Stark, 2007) conceptualise coercive control as being something that, almost exclusively, male perpetrators enact on female victim-survivors. In this context, it is notable that the study of Australian adults aged 18+ years did not find a significant difference between men's and women's levels of coercive control victimisation (Mayshak et al., 2020).
  • In contrast, Myhill (2015) found that, among people who had experienced an abusive relationship, a greater proportion of women compared to men had experienced coercive control. This was a significant difference between men and women. The inconsistency in these findings may in part relate to the samples each study examined. Mayshak and colleagues (2020) examined a general population sample, whereas Myhill (2015) examined a clinical sample of those who have been in an abusive relationship and, therefore, may have been more likely to experience coercive control than the general population.
• having higher levels of educational training (Kaukinen & Powers, 2015; Skafida et al., 2022).
• having lower household income (Kaukinen & Powers, 2015; Skafida et al., 2022).

Because the significant findings have not yet been adequately replicated in multiple studies, it is difficult to draw conclusions from the current evidence.

Psychopathy has also been found to be a risk factor associated with coercive control victimisation but not perpetration (Mayshak et al., 2020). Theoretically, the opposite should be true because psychopathy is a socially aversive personality trait characterised by a lack of remorse, thrill-seeking tendencies, manipulation, lying, impulsivity and behaviour generally motivated by callousness (Miller et al., 2010; Paulhus, 2014). It could be that this finding is the result of methodological challenges in distinguishing between mutually violent relationships and other kinds of IPV. Further research is required to understand the positive association between psychopathy and coercive control victimisation.

Three studies considered ethnicity as a possible risk factor but each measured ethnicity in different ways and the findings across these studies were inconsistent with each other. One study found no significant difference in coercive control victimisation between 4 ethnicity categories (white, Black, Asian and other). One reported that identifying as 'white' is a risk factor compared to those identifying as 'other ethnicity'. One reported identifying as an Aboriginal woman is a risk factor (in comparison to identifying as a non-Aboriginal women).

Age was not found to be a risk factor for experiences of coercive control victimisation in either of the 2 studies that considered this risk factor (Mayshak et al., 2020; Sharp-Jeffs et al., 2018). Both studies used age as a continuous variable to see whether being younger or older was significantly associated with higher or lower levels of coercive control experiences. These are useful foundation findings, requiring replication in representative samples to increase our confidence that there is not a relationship between age and victimisation in this way.

A future direction for exploring age as a risk factor is to compare age groups. It may be that specific age groups experience coercive control victimisation differently, either qualitatively or quantitatively (e.g. people aged 65+ years). These group-based differences are not apparent when examining age as a continuous variable. If age group differences exist, this is important to know so that tailored prevention and intervention strategies can be developed. Refer to Appendix B for discussion on future research directions on age as a risk factor for coercive control victimisation.

Because most of the studies we reviewed used a cross-sectional design, the findings are not able to identify risk factors that caused coercive control victimisation. The research findings can only indicate that the risk factor is associated with coercive control victimisation. This is important evidence for building understanding of coercive control victimisation but there may be other, unidentified factors influencing the relationships between risk factors and coercive control victimisation than are reported here. There is a need for longitudinal research that follows people over time to understand more about the nature of the relationships between coercive control victimisation and the risk factors identified in this review to assess how robust the associations are. This kind of information could help inform prevention and early intervention strategies to support specific subpopulations at risk of or experiencing coercive control victimisation.

Impacts associated with coercive control victimisation

The main impacts of coercive control victimisation and implications for practice, research and policy are set out in the following box. The rest of the chapter provides more detail on the findings of studies assessing prevalence.

What the research tells us about the impacts of coercive control victimisation

The reviewed studies identified several potential impacts associated with coercive control victimisation, of which several related to mental health. All significant and non-significant findings from the reviewed studies have been summarised in Table 2. Statistically significant findings have been highlighted in blue and non-significant findings have been highlighted in pink.

Direct comparison across studies is challenging because different populations and impacts have been examined. Most of the reviewed studies used cross-sectional study designs (Broughton & Ford-Gilboe, 2017; Ellis et al., 2021; Ford-Gilboe et al., 2016; Levine & Fritz, 2016; Myhill, 2015). As mentioned previously, in this type of study variables are measured at a single point in time. This means we cannot be sure whether a variable (e.g. mental health challenges) is a consequence of coercive control victimisation or a characteristic of people at risk of coercive control victimisation (Setia, 2016).

Note: blue = statistically significant, pink = not statistically significant.

None of the reviewed studies addressing the impact of coercive control victimisation were conducted in Australia. This potentially limits the applicability of these findings to Australia. However, the countries in which studies were undertaken are relatively similar to Australia¹⁴ and so the impacts of coercive control victimisation are likely to be comparable.

This review identified that mental health is the most frequently researched, and so the most well-established statistically significant, impact domain of coercive control victimisation.¹⁵ The more coercive control experiences that participants reported, the more likely they were to also report elevated PTSD symptoms and to be accessing mental health treatment (Levine & Fritz, 2016). Women who had experienced coercive control victimisation had lower levels of self-esteem, confidence, sense of safety and life satisfaction (Sharp-Jeffs et al., 2018). Men and women who had experienced coercive control victimisation were more likely to experience mental health related symptoms such as difficulty sleeping, nightmares and depressive symptoms (Myhill, 2015). However, it is noteworthy that a study specifically measuring depressive symptoms among women did not report a significant association between coercive control victimisation and depressive symptoms (Levine & Fritz, 2016).

As we note above, due to the research designs of the reviewed studies, the relationship between these mental health symptoms and coercive control victimisation is also not fully established. Previous research suggests that experiences of IPV can lead to elevated PTSD symptoms. However, because much of the data on coercive control victimisation and mental health challenges, such as PTSD symptoms, are collected at the same time, we cannot be entirely sure if PTSD symptoms are a consequence of coercive control, or if they instead might place people at greater risk of coercive control victimisation or be associated with some other previous non-IPV related trauma experiences of the victim-survivor.

Women experiencing higher levels of coercive control victimisation also experienced significantly reduced levels of satisfaction and reduced levels of decision-making abilities relating to (Sharp-Jeffs et al., 2018):

• sense of self and satisfaction with their lives
• security and comfort with finances and their physical home
• appearance and overall wellbeing
• having friends, accessing support and having intimate relationships
• connection to their community.

Sharp-Jeffs and colleagues (2018) explored freedom of decision making in a 3-year longitudinal study. They found that after women had left the violent relationship, the association between coercive control and freedom in decision making was no longer significant in any of the above domains (Sharp-Jeffs et al., 2018). Despite these positive changes relating to decision making, qualitative data from the same study indicated that women are likely to experience different kinds of abuse after separation than those they experienced before separation. This is consistent with broader research that has found coercive control and violent behaviours in intimate partner relationships continue and may escalate during and after separation (Bruton & Tyson, 2018; Douglas, 2018; Eriksson et al., 2022; Monckton Smith, 2020).

Another cross-sectional study examined the impact of coercive control victimisation, using a sample containing both women and men, and compared the impact of coercive control with the impact of situational violence (Myhill, 2015). Compared to victim-survivors of situational violence, coercive control victim-survivors were:

• approximately 3 times more likely to sustain a physical injury
• two times more likely to take time away from paid work due to the abuse that they experienced.

Method: How the rapid literature review was conducted

The rapid literature review study followed the method outlined by Fink (2010). This method has 3 main elements: sampling the literature, screening the literature, and extracting data. Figure 1 shows the steps in the procedure. This procedure is consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist for reporting systematic reviews and meta-analyses (Moher et al., 2009).

Figure 1: Flow chart of the review process

How the literature was sampled

Key terms were developed and used to source relevant studies from the electronic databases PubMed and PsycInfo. This search took place between 28 April and 22 June 2022. These databases were chosen because they contain research and practice literature related to coercive control (psychology, social work, sociology, criminology, and health and nursing).

The aim was to sample empirical research papers focusing specifically on coercive control. For this reason, a single search term 'coercive control' was used. The aim was not to sample literature that would come from using other terms that may have been used before the popularisation of the term 'coercive control'. The following terms were not used in the search, even though they relate to the research questions for this review paper: 'prevalence', 'predictors', 'risk factors', 'outcomes', or 'impact'. Keeping the search broad helped the research team find research studies that:

• provided useful background information for this paper
• the research team were not aware of prior to conducting the review
• could be grouped together to create other useful subtopics for practitioner resources on coercive control, beyond the development of this paper.

How the literature was screened

The literature was screened in 3 phases: a practical screen, a methodological screen (Fink, 2010) and a final screen for country or origin (Table 3). Each of these 3 screening phases are described in this section.

Practical screen for the usefulness of the literature

The practical screen is the initial assessment of how useful a piece of literature might be to the overall review (Fink, 2010). This assessment was based on a set of predetermined inclusion and exclusion criteria (Table 3). Stakeholder engagement consultations indicated that the term 'coercive control' has only been adopted in approximately the past 5 years. The search was conservatively narrowed to articles published from 1 January 2012 to June 2022 (10 years) to ensure that research older than 5 years was not missed, if it existed. This time span also means that the literature sampled is recent, while still being able to show some changes in the literature over the past decade.

Different, unrelated definitions for the term 'coercive control' were found in literature outside of DFV contexts. For example, in literature discussing management styles in business, methods of enforcing the law among police officers, and a type of authoritarian parenting that is considered normative. These alternative conceptualisations of coercive control were excluded.

For each study identified through the sampling procedure, the abstract was examined to determine whether the focus was on victim-survivors of coercive control. To be included, each article needed to provide findings and a discussion specifically relevant to one or more of the following:

• how common coercive control victimisation is
• risk factors associated with coercive control victimisation
• impacts associated with coercive control victimisation

Where studies did not include an abstract, an electronic key term search was conducted, and relevant parts of the document were reviewed to establish the context and purpose of the document. Where an electronic key term search was not possible, the entire document was reviewed to assess suitability. One-hundred and forty-seven (147) studies met the practical screen criteria for further review.

Methodological screen for methods used within the literature

The methodological screen was applied to each of the 147 studies (refer to Table 3). The aim was to include studies where primary quantitative data were reported and the study methods were clearly explained. As shown in Figure 1, 68 documents met the methodological screen criteria for further review.

Screen for country of origin

Only 3 peer-reviewed studies were identified from Australia. However, international research can offer additional insights about prevalence, risk factors and impacts associated with coercive control victimisation. Three inclusion requirements were used to limit international literature to countries similar to Australia (Esdaile et al., 2019):

1. high income earning according to the world bank
2. majority English-speaking population
3. population between 4 and 70 million.¹⁷

The final sample consisted of 13 studies from Australia, Canada, Republic of Ireland and the United Kingdom (refer to Table 4). Although having a smaller population than required, Wales was included because it would have otherwise been the only country in the United Kingdom to be excluded. Studies from the USA were excluded because of the large population size of that country. This is noteworthy because the sampling stage of this review identified 40 studies from the USA (refer to Table 4).

The exclusion of research from the USA was further supported by research suggesting that prevalence rates of IPV in Australia are equivalent to the average prevalence across Organisation for Economic Co-operation and Development (OECD) countries (approximately 1 in 4 women); whereas prevalence rates for the USA are among the highest for OECD countries (approximately 1 in 3 women) (OEGD, 2020).

In a study of coercive control by Kaukinen and Powers (2015), Canada and the USA were compared specifically because the rate of intimate partner violence was much higher in the USA than in Canada. Kaukinen and Powers (2015, p 240) suggested that differences between the 2 countries may be based on economic factors, namely that women in Canada have greater access to 'income-supplementing programs, such as family leave, extensive employment insurance, and social welfare programming, which likely insulate women from the risk for violence by male intimate partners.' The purpose of screening by country is to identify research findings through the review that may help to understand and address coercive control experiences in Australia.

How data were extracted from the literature

A data extraction template was completed for each article included in the review. The template was developed with an initial series of categories based on: (a) existing literature, (b) standards of research reporting, and (c) standards of research rigour. The data extraction template evolved through the extraction process to reflect the content and specifics of coercive control research. This collection of extracted data documents was used to identify patterns of findings as well as some unexplored areas for future consideration.

How stakeholders informed the review

Coercive control was selected as a priority topic because of feedback provided by the Child Family Community Australia (CFCA) information exchange audience and stakeholders. We then conducted targeted stakeholder engagement with experts within the field of coercive control to inform the development of this rapid literature review study and related practitioner resources. This section briefly outlines how the key stakeholder engagement steps were conducted.

1. CFCA information exchange conducts an annual survey to understand the evidence needs of professionals in the child, family and community welfare sector. The November 2021 survey received 1,233 responses and respondents told us that domestic and family violence (including the subtopic coercive control) was a priority area. Practitioners responding to the survey provided qualitative data relating to practice questions and evidence needs relating to coercive control.
2. Two previous CFCA webinars had a focus on coercive control.¹⁸ Prior to, during and after the live webinars, registrants provided practice questions and feedback relating to coercive control. The qualitative information collected about the CFCA audience in the first 2 steps was collated and questions and insights were grouped into themes.
3. A preliminary search of the coercive control literature was conducted to identify major subtopic areas. The qualitative themes from the first 2 steps were then mapped to the topic areas from the research literature. The result was an initial plan for possible resources to develop. The mapping of the CFCA audience questions to the topics present within the research literature provided a chance to see what answers might be available in the existing literature and where some practice-research gaps may exist.
4. To assess the usefulness and feasibility of the initial plan, we engaged in a second phase of stakeholder engagement. Seven practitioners, service leaders and researchers who are experts in DFV were consulted between February and April 2022. The initial plan was shared with a group of experts for feedback. The experts also suggested research, policy, and practice questions that might be useful to generalist practitioners working with individuals and families impacted by coercive control.

Some potential considerations for future reviews

This rapid review aimed to synthesise contemporary (i.e. published in the last decade) original quantitative research on coercive control victimisation that was conducted in Australia or countries that are generalisable to Australia. This paper does not intend to provide a comprehensive review of all research ever conducted on coercive control frequency, risk factors and impacts globally. Others aiming to extend this review may wish to consider the following, where relevant:

• Search for, and include, grey literature on coercive control.
• Search for and include qualitative studies on coercive control. Such research provides rich and context-specific data that can provide insights for aspects of coercive control victimisation that quantitative research has not yet addressed. This allows for the development of further implications and recommendations for practice, research and policy.
• We conducted a broad search for research that included the phrase 'coercive control' in the title or abstract. Research from the wider family and intimate partner violence field may provide useful information on coercive control by examining related concepts such as stalking or abuse.
• Search for and include research from countries beyond those included in this review (Table 4) or published before 2012.
• Additional searches consistent with systematic review methods that were beyond the scope of this rapid review:
  • searching a larger number of academic databases (i.e. >3 databases)
  • screening the reference lists of included studies
  • screening the studies that have cited the included articles
  • contacting authors of included studies and asking them to provide additional relevant studies or data.

Further reading and related resources

• How to support clients exposed to technology-facilitated coercive control  
  Drawing on the latest research and practitioner insights this AIFS webinar: (a) describes what technology-facilitated coercive control looks like in practice, (b) provides examples of the different ways that victim-survivors might experience technology-facilitated coercive control, and (c) suggests strategies for face-to-face and telehealth practice.
• Extended Q&A for: How to support clients exposed to technology-facilitated coercive control  
  The introduction of this AIFS webinar summarises some key discussion points and highlights from the December 2022 webinar How to support clients exposed to technology-facilitated coercive control. This introduction is followed by an extended Q&A from the December 2022 webinar panel answering more audience questions.
• Reproductive coercion and abuse  
  This AIFS practice guide describes the evidence on reproductive coercion and abuse (RCA). It covers: (a) what RCA is, (b) strategies used by perpetrators, (c) the impacts of experiencing RCA, (d) factors that influence a person's risk of experiencing RCA, and (e) how to ask about RCA victimisation. Finally, some tips are provided for supporting clients who may be experiencing RCA.
• Technology-facilitated coercive control  
  This AIFS practice guide describes the research evidence on technology-facilitated coercive control (TFCC). It covers: (a) what TFCC is, (b) strategies used by perpetrators, (c) the interaction between face-to-face and technology-facilitated strategies of coercive control, (d) client groups who may be at an elevated risk of TFCC victimisation, (e) the impacts of experiencing TFCC, (f) risks of expecting victim-survivors to stop using technology, and (g) tips for supporting clients who may be experiencing TFCC.
• Technology-facilitated coercive control (TFCC): Evidence-based insights for practice  
  This AIFS recording provides an overview of key trends in the research literature relating to technology-facilitated coercive control.
• Intersecting systems and the needs of families: Family law, child protection and domestic violence  
  This AIFS webinar reunited the panellists from AIFS Conference 2022 event 'Intersecting systems and the needs of families: Family law, child protection and domestic violence'. The panellists discussed this topic with a focus on implications for practitioners working in areas related to family and domestic violence, family law and child protection.
• How to ask adult mental health clients about sexual abuse  
  This AIFS short article outlines: (a) why adult survivors of sexual abuse may not disclose their abuse experiences to mental health practitioners, (b) why practitioners might avoid asking about it, and (c) research findings from mental health practitioner samples on how to safely ask clients about sexual abuse histories.
• The power in understanding patterns of coercive control  
  This AIFS webinar explores ways that services can use the language of coercive control to support women to expose patterns of abusive behaviour.
• Sadie's story: Helping women affected by domestic and family violence navigate a fragmented system  
  This AIFS webinar is focused on one woman's story of domestic and family violence. Sadie (not her real name) is an Aboriginal woman and mother of 2. Her story sheds light on how fragmented service systems can often fail to meet women's needs and may even exacerbate the challenges they experience at the intersection of gender and racial inequality.
• Defining and responding to coercive control: Policy brief  
  This ANROWS policy brief is 'designed to assist policy makers developing legal or policy and practice frameworks to prevent or respond to coercive control in relation to domestic and family violence.'
• Australians' attitudes to violence against women and gender equality. Findings from the 2017 National Community Attitudes towards Violence against Women Survey  
  This ANROWS report 'documents findings from the 2017 National Community Attitudes towards Violence against Women Survey (NCAS) and considers them in the context of other related research.'
• Invisible Practices: Intervention with fathers who use violence (Research report)  
  This research report, from ANROWS, documents the project 'Invisible Practices', which sought to develop the workforce capacity and evidence-informed guidelines of practitioners who work with fathers who use DFV. It incorporates the Safe and Together model.
• Invisible Practices: Intervention with fathers who use violence (Practice guide)  
  Supported by the research report above, this ANROWS practice guide, outlines an evidence-based approach for working with fathers who use DFV. It provides practical strategies, reflection questions and techniques to support practitioners in their work.

References

Australian Institute of Health and Welfare (AIHW). (2018). Family, domestic and sexual violence in Australia, 2018. Canberra: AIHW. Retrieved from www.aihw.gov.au/reports/domestic-violence/family-domestic-sexual-violence-in-australia-2018

Australia's National Research Organisation for Women's Safety (ANROWS). (2021). Defining and responding to coercive control: Policy brief (ANROWS Insights, 01/2021). Sydney: ANROWS. Retrieved from www.anrows.org.au/publication/defining-and-responding-to-coercive-control

Bendlin, M., & Sheridan, L. (2019). Nonfatal strangulation in a sample of domestically violent stalkers: The importance of recognizing coercively controlling behaviors. Criminal Justice and Behavior, 46(11), 1528-1541. 
dx.doi.org/10.1177/0093854819843973

Boxall, H., Dowling, C., & Morgan, A. (2020). Female perpetrated domestic violence: Prevalence of self-defensive and retaliatory violence. Trends and Issues in Crime and Criminal Justice, 584, 1-17.

Boxall, H., & Morgan, A. (2021a). Experiences of coercive control among Australian women. Canberra: Australian Institute of Criminology.

Boxall, H., & Morgan, A. (2021b). Who is most at risk of physical and sexual partner violence and coercive control during the COVID-19 pandemic? Canberra: Australian Institute of Criminology.

Broughton, S., & Ford-Gilboe, M. (2017). Predicting family health and well-being after separation from an abusive partner: Role of coercive control, mother's depression and social support. Journal of Clinical Nursing, 26(15-16), 2468-2481. 
doi:10.1111/jocn.13458

Bruton, C., & Tyson, D. (2018). Leaving violent men: A study of women's experiences of separation in Victoria, Australia. Australian & New Zealand Journal of Criminology, 51(3), 339-354. doi:10.1177/0004865817746711

Centers for Disease Control and Prevention (CDC). (2022). Fast Facts: Preventing intimate partner violence. Atlanta, GA: CDC. Retrieved from www.cdc.gov/violenceprevention/intimatepartnerviolence/fastfact.html

Collingburn, L., Herd, B., & Arthur, M. (2023). Elder abuse and coercive control reform in Australia: Special report. Brisbane: HopgoodGamin Lawyers. Retrieved from www.hopgoodganim.com.au/page/knowledge-centre/publications/elder-abuse-and-coercive-control-reform-in-australia

Collins, E. A., Cody, A. M., McDonald, S. E., Nicotera, N., Ascione, F. R., & Williams, J. H. (2018). A template analysis of intimate partner violence survivors' experiences of animal maltreatment: Implications for safety planning and intervention. Violence Against Women, 24(4), 452-476. doi:10.1177/1077801217697266

Compass. (2023). Understanding coercive control as elder abuse. Retrieved from www.compass.info/featured-topics/coercive-control/understanding-coercive-control-as-elder-abuse

Coumarelos, C., Weeks, N., Bernstein, S., Roberts, N., Honey, N., Minter, K., & Carlisle, E. (2023). Attitudes matter: The 2021 National Community Attitudes towards Violence against Women Survey (NCAS), Findings for Australia (Research report, 02/2023). Sydney: ANROWS. Retrieved from www.anrows.org.au/publication/attitudes-matter-the-2021-national-community-attitudes-towards-violence-against-women-survey-ncas-findings-for-australia

Douglas, H. (2018). Legal systems abuse and coercive control. Criminology & Criminal Justice: An International Journal, 18(1), 84-99. dx.doi.org/10.1177/1748895817728380

Ellis, D., Lewis, T., & Nepon, T. (2021). Effects of historical coercive control, historical violence, and lawyer representation on post-separation male partner violence against mother litigants who participated in adversarial family court proceedings. Violence Against Women, 27(9), 1191-1210. doi:10.1177/1077801220921939

Eriksson, L., Mazerolle, P., & McPhedran, S. (2022). Giving voice to the silenced victims: A qualitative study of intimate partner femicide. Canberra: Australian Institute of Criminology.

Esdaile, E., Thow, A. M., Gill, T., Sacks, G., Golley, R., Love, P, et al. (2019). National policies to prevent obesity in early childhood: Using policy mapping to compare policy lessons for Australia with six developed countries. Obesity Reviews, 20(11), 1542-1556. doi:10.1111/obr.12925

Federation of Ethnic Communities' Councils of Australia (FECCA). (2020). If we don't count it ... it doesn't count. Canberra: FECCA.

Fink, A. (2010). Conducting research literature reviews: From the internet to paper (3rd ed.). London: SAGE Publications, Inc.

Ford-Gilboe, M., Wathen, C. N., Varcoe, C., MacMillan, H. L., Scott-Storey, K., Mantler, T. et al. (2016). Development of a brief measure of intimate partner violence experiences: The Composite Abuse Scale (Revised)-Short Form (CASR-SF). BMJ Open, 6(12), e012824. doi:10.1136/bmjopen-2016-012824

Frankland, A., & Brown, J. (2014). Coercive control in same-sex intimate partner violence. Journal of Family Violence, 29(1), 15-22. dx.doi.org/10.1007/s10896-013-9558-1

Goodboy, A. K., & Martin, M. M. (2015). The personality profile of a cyberbully: Examining the Dark Triad. Computers in Human Behavior, 49, 1-4. doi.org/10.1016/j.chb.2015.02.052

Gough, D., Davies, P., Jamtvedt, G., Langlois, E., Littell, J., Lotfi, T. et al. (2020). Evidence Synthesis International (ESI): Position statement. Systematic Reviews, 9(1), 155. doi:10.1186/s13643-020-01415-5

Hamberger, L. K., Larsen, S. E., & Lehrner, A. (2017). Coercive control in intimate partner violence. Aggression and Violent Behavior, 37, 1-11. doi:10.1016/j.avb.2017.08.003

Johnson, M. P. (2008). Control and violence in intimate relationships. In A typology of domestic violence: Intimate terrorism, violent resistance, and situational couple violence (pp. 13-28). Hanover: Northeastern University Press.

Jonason, P. K., & Middletone, J. P. (2015). Dark Triad: The 'Dark Side' of human personality. In J. D. Wright (Ed.), International Encyclopedia of the Social & Behavioral Sciences (2nd ed., Vol. 5, pp. 671-675). Oxford: Elsevier.

Kaukinen, C. E., & Powers, R. A. (2015). The role of economic factors on women's risk for intimate partner violence: A cross-national comparison of Canada and the United States. Violence Against Women, 21(2), 229-248. doi:10.1177/1077801214564686

Levine, A. R., & Fritz, P. A. T. (2016). Coercive control, posttraumatic stress disorder, and depression among homeless women. Partner Abuse, 7(1), 26-43. dx.doi.org/10.1891/1946-6560.7.1.26

Mayshak, R., King, R. M., Chandler, B., & Hannah, M. (2020). To swipe or not to swipe: The Dark Tetrad and risks associated with mobile dating app use. Personality and Individual Differences, 163. dx.doi.org/10.1016/j.paid.2020.110099

Miller, J. D., Dir, A., Gentile, B., Wilson, L., Pryor, L. R., & Campbell, W. K. (2010). Searching for a vulnerable dark triad: Comparing Factor 2 psychopathy, vulnerable narcissism, and borderline personality disorder. Journal of Personality, 78(5), 1529-1564. doi:10.1111/j.1467-6494.2010.00660.x

Mission Australia. (2021). Types of domestic violence abuse. Retrieved from www.missionaustralia.com.au/stories/safe-homes/types-of-domestic-violence-abuse

Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The PRISMA Group. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6(6).

Monckton Smith, J. (2020). Intimate partner femicide: Using Foucauldian analysis to track an eight stage progression to homicide. Violence Against Women, 26(11), 1267-1285. dx.doi.org/10.1177/1077801219863876

Moulding, N., Franzway, S., Wendt, S., Zufferey, C., & Chung, D. (2021). Rethinking women's mental health after intimate partner violence. Violence Against Women, 27(8), 1064-1090. doi:10.1177/1077801220921937

Muri, K., Augusti, E. M., Bjørnholt, M., & Hafstad, G. S. (2022). Childhood experiences of companion animal abuse and its co-occurrence with domestic abuse: Evidence from a national youth survey in Norway. Journal of Interpersonal Violence, 8862605211072176. doi:10.1177/08862605211072176

Myhill, A. (2015). Measuring coercive control: What can we learn from national population surveys? Violence Against Women, 21(3), 355-375. doi:10.1177/1077801214568032

Myhill, A., & Hohl, K. (2019). The 'Golden Thread': Coercive control and risk assessment for domestic violence. Journal of Interpersonal Violence, 34(21-22), 4477-4497. doi:10.1177/0886260516675464

Nancarrow, H., Thomas, K., Ringland, V., & Modini, T. (2020). Accurately identifying the 'person most in need of protection' in domestic and family violence law. Sydney: ANROWS.

Nevala, S. (2017). Coercive control and its impact on intimate partner violence through the lens of an EU-wide survey on violence against women. Journal of Interpersonal Violence, 32(12), 1792-1820. doi:10.1177/0886260517698950

Organisation for Economic Co-operation and Development (OECD). (2020). Intimate partner violence. OECD family database. Paris: OECD. Retrieved from www.oecd.org/els/soc/SF-3-4-Intimate-Partner-Violence.pdf

Parker, R., & Lamont, A. (2010). Evaluating programs. Communities and Families Clearinghouse Australia. Melbourne: Australian Institute of Family Studies.

Parliament of New South Wales. (2022). Crimes Legislation Amendment (Coercive Control) Bill 2022. Retrieved from www.parliament.nsw.gov.au/bills/Pages/bill-details.aspx?pk=4024

Paulhus, D. L. (2014). Toward a taxonomy of dark personalities. Current Directions in Psychological Science, 23(6), 421-426. doi:10.1177/0963721414547737

Paulhus, D. L., Curtis, S. R., & Jones, D. N. (2018). Aggression as a trait: The Dark Tetrad alternative. Current Opinion in Psychology, 19, 88-92. doi.org/10.1016/j.copsyc.2017.04.007

Paulhus, D. L., & Dutton, D. G. (2016). Everyday sadism. In V. Z.-H. D. K. Marcus (Ed.), The dark side of personality: Science and practice in social, personality, and clinical psychology (pp. 109-120). Washington, DC: American Psychological Association.

Pedersen, J. S., Malcoe, L. H., & Pulkingham, J. (2013). Explaining aboriginal/non-aboriginal inequalities in postseparation violence against Canadian women: Application of a structural violence approach. Violence Against Women, 19(8), 1034-1058. doi:10.1177/1077801213499245

Pitman, T. (2017). Living with coercive control: Trapped within a complex web of double standards, double binds and boundary violations. British Journal of Social Work, 47(1), 143-161. Retrieved from ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=psyc14&NEWS=N&AN=2017-44803-009

Queensland Government. (2023). First stage in legislating against coercive control passes parliament. Retrieved from statements.qld.gov.au/statements/97226

Rogers, P., & Goodrick, D. (2010). Qualitative data analysis. In J. Wholey, H. Hatry, & K. Newcomer (Eds.), Handbook of Practical Program Evaluation (3rd ed., pp. 429-453). San Francisco, CA: Jossey-Bass.

Safe from Violence. (n.d.). Non-physical violence. Retrieved from www.safefromviolence.tas.gov.au/resources-hub/non-physical-violence

Seniors Rights Service. (2021). Response to coercive control discussion paper. Sydney: Senior Rights Service. Retrieved from www.parliament.nsw.gov.au/ladocs/submissions/70614/Submission%20-%20127.pdf

Setia, M. S. (2016). Methodology series module 3: Cross-sectional studies. Indian Journal of Dermatology, 61(3), 261-264. doi:10.4103/0019-5154.182410

Sharp-Jeffs, N., Kelly, L., & Klein, R. (2018). Long journeys toward freedom: The relationship between coercive control and space for action-measurement and emerging evidence. Violence Against Women, 24(2), 163-185. doi:10.1177/1077801216686199

Skafida, V., Morrison, F., & Devaney, J. (2022). Prevalence and social inequality in experiences of domestic abuse among mothers of young children: A study using national survey data from Scotland. Journal of Interpersonal Violence, 37(11-12), NP9811-NP9838.

Smoker, M., & March, E. (2017). Predicting perpetration of intimate partner cyberstalking: Gender and the Dark Tetrad. Computers in Human Behavior, 72, 390-396. doi.org/10.1016/j.chb.2017.03.012

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

Stark, E. (2009). Rethinking Coercive Control. Violence Against Women, 15(12), 1509-1525. doi:10.1177/1077801209347452

Stark, E., & Hester, M. (2019). Coercive control: Update and review. Violence Against Women, 25(1), 81-104. dx.doi.org/10.1177/1077801218816191

Ulbrick, M., & Jago, M. (2018). 'Officer she's psychotic and I need protection': Police misidentification of the 'primary aggressor' in family violence incidents in Victoria. Victoria: Women's Legal Service Victoria and Monash University.

Victorian Government. (2022). Preferred terminology: Definitions of language used in this strategy. Retrieved from www.vic.gov.au/family-violence-lived-experience-strategy/preferred-terminology#victim-survivor

Webster, K., Diemer, K., Honey, N., Mannix, S., Mickle, J., Morgan, J. et al. (2018). Australians' attitudes to violence against women and gender equality. Findings from the 2017 National Community Attitudes towards Violence against Women Survey (NCAS) (Research report, 03/2018). Sydney: ANROWS. Retrieved from www.anrows.org.au/publication/australians-attitudes-to-violence-against-women-and-gender-equality-findings-from-the-2017-national-community-attitudes-towards-violence-against-women-survey

Appendix B: Future research directions on age as a risk factor for coercive control victimisation

In the discussion on Risk factors associated with coercive control victimisation, it was noted that a future direction for exploring age as a risk factor is to compare age groups. Age was not found to be a risk factor for experiences of coercive control victimisation in either of the 2 studies reviewed in this paper that considered this factor (Mayshak et al., 2020; Sharp-Jeffs et al., 2018). Both studies used age as a continuous variable to see whether being younger or older (higher or lower age score) was significantly associated with higher or lower levels of coercive control experiences. These are useful foundation findings, requiring replication in representative samples to increase our confidence that there is not a relationship between age and victimisation in this way.

However, it may be that specific age groups experience coercive control victimisation differently, either qualitatively or quantitatively. These group-based difference are not apparent when examining age as a continuous variable. If age group differences exist, this is important to know so that tailored prevention and intervention strategies can be developed. As an example of possible quantitative difference, an Australian report found that women aged 18-24 were 6 times more likely to experience coercive control during the COVID-19 pandemic than women aged 55+ (Boxall & Morgan, 2021b).

As a qualitative difference, research on elder abuse highlights the unique abuse and IPV experiences for people aged 65+ years. Decline in physical and mental capacities from age 60 and the associated increase in reliance on others, such as intimate partners for care, increase the possibility for perpetrators to exercise coercive controlling behaviours (e.g. restricting movement, communication and interactions, threatening the continuation or quality of care, or inflicting abuse under the guise of care) (Compass, 2023). These circumstances also decrease the likelihood that victim-survivors will disclosure their abuse or access support (Seniors Rights Service, 2021). The intersectionality of coercive control victimisation and being aged 65+ is an important area for further investigation in Australia to inform law and policy reform (Collingburn, Herd, & Arthur, 2023; Seniors Rights Service, 2021).