‘“She Tells Me I'm Pushy” is More Likely than the Man Directly Admitting to Being Pushy’: Practitioners’ Views on Screening and Assessing Risk of Intimate Partner Sexual Violence

Abstract


Introduction
Domestic and family violence (DFV) and sexual violence response pathways are often divided, with sex offender programs used in response to sexual offending, and DFV perpetrator interventions-such as Men's Behaviour Change Programs (MBCPs)-used in response to DFV perpetration (Mackay et al. 2015).Despite these siloed pathways, sexual violence and DFV frequently intersect.Most perpetrators of sexual assault against women are current or former partners (Australian Bureau of Statistics [ABS] 2017), and this violence often occurs together with other forms of DFV (Bagwell-Gray et al. 2015;Cox 2015;Sullivan et al. 2012;Tarzia 2021b).
Internationally, little research has examined how DFV perpetrator interventions address sexual violence (Westmarland and Kelly 2017).This study directly addresses this knowledge gap, examining the views of DFV perpetrator intervention practitioners on screening and risk assessment for intimate partner sexual violence (IPSV) perpetration.The findings presented in the paper demonstrate that IPSV is not receiving the same level of attention as other forms of DFV.Key barriers described by practitioners relate to assessment tools, practitioner confidence, service coordination and referrals, nondisclosure, and limitations associated with language and communication.While this study draws on the professional views of practitioners working in Australia, findings hold relevance for understandings of DFV perpetrator intervention work internationally and highlight the need for focused consideration of IPSV.Domestic and family violence (DFV) and sexual violence intersect, with sexual violence often perpetrated by an intimate partner alongside other forms of DFV.While DFV perpetrator interventions are commonly used in response to DFV perpetration, scant research has considered how these interventions identify and address sexual violence, including intimate partner sexual violence (IPSV).Drawing on the findings from an Australian study which involved a survey of 97 practitioners, this paper explores screening and risk assessment of IPSV within the DFV perpetrator intervention context.The research findings demonstrate limited screening and risk assessment of IPSV, particularly when compared to other forms of DFV.This demonstrates a clear need for focused attention on IPSV as part of broader efforts to hold perpetrators accountable for all forms of DFV.

Background
One manifestation of the DFV and sexual violence intersection is IPSV (other examples include sexual violence perpetrated by siblings, in-laws, carers, parents, and others where this co-occurs with non-sexual forms of DFV).Relegating IPSV to the domain of sexual violence or DFV fails to sufficiently capture the unique harms of this violence (Barker et al. 2019;Tarzia 2021a).IPSV is a sign of escalating frequency and severity of DFV, including a higher risk of lethality (Bagwell-Gray 2021; Dobash et al. 2007; Royal Commission into Family Violence [RCFV] 2016).Victim-survivors of IPSV report a range of intersecting harms, including psychological, emotional, and physical conditions.This includes greater levels of shame, and more clinically significant distress than people who experience non-sexual intimate partner violence (IPV) only (Barker et al. 2019;Cox 2015). 1 Issues related to self-esteem and body image, self-loathing, shame, and lack of confidence are also frequently reported by victim-survivors of IPSV (Bagwell-Gray 2021; Easteal and McOrmond-Plummer 2006;Tarzia 2021a).Women who have experienced IPSV report long-term harms and 'invisible impacts' associated with the deprivation of agency, objectification, dehumanisation, and violation of trust experienced (Cox 2015;Tarzia 2021a).Victimsurvivors report fear and difficulty in trusting future partners (Tarzia 2021a), although the impact on relationships is not limited to partners.Rather, the trauma of IPSV 'sits in the body' and is carried by victim-survivors, impacting relationships and interactions with family, friends, caseworkers, and others (Hamilton et al. 2023: 24).Victim-survivors of IPSV are more likely than people who experience non-sexual IPV only to experience certain health conditions, including gynaecological conditions, central nervous system conditions, chronic stress disorders, urinary tract infections, gastrointestinal disorders, headaches, asthma, hypertension, seizures, unintended or unwanted pregnancies, increased likelihood of miscarriage, and sexually transmitted infections (Cox 2015;Palmer and Parekh 2013).
Responses to sexual violence perpetrators typically focus on sexual offending through sexual offender programs (Mackay et al. 2015).These generally operate in custodial settings, recognising that reporting and conviction rates for sexual violence are low, particularly for IPSV (RCFV 2016).Camilleri and Miele (2013) suggest it may be more effective to design sexual offender treatment programs tailored to IPSV offenders.However, current models mean that these programs are limited in reach (to perpetrators who encounter the criminal justice system), and in focus (to convicted offence(s)).
Perpetrators of DFV are often directed to MBCPs.There is variation in the design, theoretical perspective, content, and delivery of these programs; however, in the Australian context, they are largely informed by the Duluth model (Pence and Paymar 1993).Sexual respect is a core module in some programs; however, in Australia and internationally, few studies have examined the nature and extent to which DFV perpetrator interventions attend to sexual violence or the intersections of these violences (Heenan 2004;Westmarland and Kelly 2017;Yllö 1999).Additionally, few studies have examined screening and risk assessment of IPSV for DFV perpetrators.Research in this space has largely focused on the predictive validity of assessment tools to identify DFV recidivism (including, but not limited to, sexual violence) among IPSV perpetrators (see for example, Grann and Wedin 2002;Thomas et al. 2022;Rettenberger and Eher 2013).However, these studies have examined DFV risk among IPSV offenders, rather than IPSV risk specifically.No studies were identified that considered screening and risk assessment practice related to IPSV beyond a focus on assessment tools.
Australia has an overarching and coordinated policy framework: National Plan to End Violence Against Women and Children 2022-2032(henceforth, the National Plan 2022-2032 (Department of Social Services (DSS) 2022)) which was preceded by the National Plan to Reduce Violence Against Women and Children 2010-2022(henceforth, the National Plan 2010-2022(Council of Australian Governments 2011).These frameworks outline the actions needed to end violence against women and children, with specific steps and strategies detailed in accompanying action plans.In line with the Third Action Plan under the National Plan 2010-2022, the National Risk Assessment Principles for Domestic and Family Violence Framework was developed.The National Principles provide an overarching understanding of risk and risk management intended to aid more consistent practice across all Australian jurisdictions.Principle 8 states 'intimate partner sexual violence must be specifically considered in all risk assessment processes' (Toivonen and Backhouse 2018: 10).However, little is known about the assessment of IPSV in practice.

Method
The purpose of this study was to understand current screening and risk assessment practices related to IPSV within the DFV perpetrator intervention context.The study was guided by two research questions: 1. What is the current state of screening and risk assessment of IPSV within the DFV perpetrator intervention context? 2. What are the barriers and enablers to identifying and assessing risk of IPSV when working with victim-survivors and perpetrators of DFV in the perpetrator intervention context?

Survey
The survey was hosted on Qualtrics online platform and contained both quantitative (asked on a five-point Likert-scale) and qualitative questions.The survey was advertised via the Monash Gender and Family Violence Prevention Centre and No to Violence (NTV, the peak men's behaviour change body in Australia) X (formerly Twitter) accounts and organisational enewsletters.The survey landing page contained a copy of the explanatory statement and a question regarding consent.The study received ethics approval from Monash University (project ID: 32345).
In seeking to capture breadth of participation, the survey targeted four practitioner groups: • practitioners working with DFV perpetrators (for example, case managers or facilitators) • practitioners working with DFV victim-survivors (for example, family safety contact workers) • trainers who deliver training for practitioners • professionals involved in the design, development, or management of interventions.
The survey sought practitioner information, including jurisdiction, years' experience, education and training, and current and past role(s).The survey also asked practitioners about their screening and risk assessment practices, including questions about the assessment tools utilised, frequency of assessing for IPSV victimisation and perpetration, and the barriers and enablers to undertaking these assessments.Participants were invited to complete questions relevant to their role and experience, as identified in the practitioner information section.
Most practitioners who worked with perpetrators (n=69, 76.7%) and/or victim-survivors (n=42, 62.7%) were involved in screening and risk assessment of DFV perpetration and victimisation respectively.Some practitioners identified having lived experience of DFV.While not the focus of the study, and not asked about in the survey, this expertise informs their understanding of, and practice related to, IPSV and is incorporated into the analysis where appropriate.

Data Analysis
Quantitative data was analysed using SPSS Statistics version 26.Items were reverse coded prior to analysis, where required, to ensure a higher score consistently indicated a more positive response (for example, greater frequency of assessing risk of IPSV).For clarity, descriptive statistics were aggregated into three-point scales within tables.Non-parametric t-tests (Wilcoxon Sign and Mann-Whitney U tests) were also conducted.Practitioners could skip questions they did not want to answer, so response rates (n) varied and are reported throughout.Qualitative data were analysed using thematic coding.

Limitations
This study draws on a small sample size (n=97).Paired samples utilised in t-tests further rely on small subsamples and should be interpreted cautiously.This is an Australian study; however, most practitioners worked in Victoria (n=58, 59.8%), with no data from practitioners in the Australian Capital Territory, Northern Territory, or South Australia, and only one practitioner (1.0%) from Tasmania.The survey draws on self-reported data, with potential variation in practitioners' selfassessment of their practices.The scope of this survey was limited to IPSV; further research examining how other forms of sexual violence are addressed in DFV perpetrator interventions would be valuable.These limitations notwithstanding, study findings offer valuable insights into the under-researched area of screening and assessing the risk of IPSV within DFV perpetrator interventions.

Assessing IPSV Through DFV Assessment Tools
Most practitioners working with victim-survivors (n=31, 83.8%) and perpetrators (n=34, 57.6%) reported that the assessment tools they used contained item(s) specifically to assess IPSV.
As expected, frequency of assessment was higher among these practitioners (Mdn=4.00,n=32) than those who used tools that did not capture IPSV (Mdn=3.00,n=14, U=346.0,z=3.055, p=.002).Although unsurprising, this finding highlights the importance of including IPSV items within assessment tools, and is particularly important given that some practitioners, especially those working with perpetrators (n=25, 42.4%), reported that the tool(s) they used did not capture IPSV.
Generally, practitioners reported that assessment tools were 'very' or 'somewhat useful' for identifying and/or assessing risk of IPSV perpetration (n=26, 76.5%), and victimisation (n=25, 80.6%).However, qualitative responses illustrate some barriers to identification and assessment associated with assessment tools.Practitioners described the IPSV items within assessments as 'very blunt' (Practitioner 97, worked with perpetrators and victim-survivors) and said the items often focused on sexual assault.Further, some practitioners were reliant on tools not designed specifically for use with perpetrators, as one practitioner explained: This language is not unique to the DVSAT.For example, the question for use with victim-survivors in the MARAM assessment tool(s) is: 'Have they ever forced you to have sex or participate in sexual acts when you did not wish to do so?' (Family Safety Victoria 2021: 214). 5Sexual assault is also listed as a risk factor in the MARAM tools for use with adults using violence (such as the Adult Person Using Violence Intermediate Assessment Tool).The MARAM tools were used by most practitioners working with perpetrators (n=38, 64.4%) and victim-survivors (n=19, 51.4%).While these tools are accompanied by practice guides to support assessment, the specific framing of the prompt remains focused on forced sexual acts and sexual assault.
Practitioner 21 described a disconnect between the 'box to complete' in assessment tools and the broader practice challenge of meaningfully unpacking sexual violence: The Assessment Forms are one thing-in that there may be a question to ask and/or a box to complete-but unpacking the topic of sexual violence in a way that makes sense both to us as interviewers and to the client as a perpetrator is yet to be discovered.(Practitioner 21, worked with perpetrators) While some practitioners reflected that such tick box approaches do not facilitate conversational assessment of IPSV, practitioners also rely on other resources, such as information sharing, professional judgement, and skill, to conduct assessments.This is also recognised, for example, through the development of the MARAM conversational assessment tools and practice guidelines.This point was expressed by some practitioners who reflected that the usefulness of instruments lay in how they were operationalised: Depending on the level of confidence from the assessor do they go into the sexual violence as in explicit acts or do they just mark the assessment with sexual violence?(Practitioner 71, worked with perpetrators and victim-survivors) It depends on how it is used.Most people implementing [it] do not apply it in a [sic] exploratory discussion process.
(Practitioner 43, worked with perpetrators and victim-survivors) Some practitioners spoke about the 'exploratory' and conversational nature of assessment and, in their own assessment practice, may go beyond asking about forced acts and sexual assault.However, other practitioners, such as Practitioner 21 (quoted above), spoke about ongoing challenges in meaningfully assessing for IPSV.The differences in frequency of assessing risk of IPSV victimisation and perpetration were more distinct when asking practitioners how frequently they assessed risk of IPSV compared to other forms of DFV.As shown in Table 2, most practitioners said they assessed risk of IPSV perpetration (n=34, 52.3%) and victimisation (n=26, 65.0%) with approximately the same frequency as other forms of DFV.Notably, 28 practitioners (43.1%) working with perpetrators and nine (22.5%) working with victim-survivors reported assessing risk of IPSV less often than other forms of DFV.Helps et al. (2023: 32) Results for both frequency items-frequency of assessing risk of IPSV and frequency of assessing risk of IPSV compared to other forms of DFV-illustrate greater reported frequency of assessing risk of IPSV victimisation (M=4.00 and M=2.80) compared to perpetration (M=3.80 and M=2.52), as shown in Tables 1 and 2. This distinction was also reflected in the qualitative data, where some practitioners reported being 'more prepared to listen and support after a disclosure of being a victim-survivor then [sic] [a] perpetrator' (Practitioner 97, worked with perpetrators and victim-survivors).Non-parametric t-tests were conducted to further investigate differences in assessing risk of IPSV victimisation and perpetration.

Assessing Risk of IPSV
The reported frequency of assessing risk of IPSV perpetration (n=32, M=4.03, SD=1.15) was higher than the frequency of assessing risk of IPSV victimisation (M=4.00,SD=1.08), although the mean difference was small (.03) and the result was not statistically significant (z=0.38,p=.707).A second non-parametric t-test (n=31) found that, compared to other forms of DFV, the frequency of assessing risk of victimisation (M=2.74,SD=0.86) was higher than the frequency of assessing risk of perpetration (M=2.58,SD=0.72,MD=0.16).This result was not statistically significant (z=-1.39,p=.166).
Frequency of assessing risk of IPSV perpetration was higher among practitioners working with both perpetrators and victimsurvivors (Mdn=4.00,n=42) compared to practitioners working with perpetrators only (Mdn=3.00,n=23), although the result was not statistically significant (U=590.0,z=1.533, p=.125).Frequency of assessing risk of IPSV perpetration compared to other forms of DFV was also higher among practitioners working with both perpetrators and victim-survivors (Mdn=3.00,n=42) compared to practitioners working with perpetrators only (Mdn=2.00,n=23), although the result was not statistically significant (U=597.5,z=1.728, p=.084). 6  In addition to questions on frequency of assessing risk of IPSV, practitioners were asked how confident they were that IPSV was identified where present.Almost half of practitioners working with perpetrators (n=31, 48.4%) were 'somewhat unconfident' or 'not at all confident' that a history of perpetrating IPSV was identified where the behaviour was present.Comparatively, 22 practitioners (55.0%) working with victim-survivors were 'somewhat unconfident' or 'not at all confident' that IPSV victimisation was always identified where the behaviour was present. 7The low confidence rates expressed by practitioners suggest that more could be done to consistently assess for IPSV in current screening practices.

The Influence of Sexual Assault Training
Mann-Whitney U tests were conducted to examine whether there were differences in screening and risk assessment practice between those who had received sexual assault training (n=25, 28.1%) and those who had not (n=64, 71.9%, missing=8).The results are shown in Table 3. Statistically significant differences between practitioners with sexual assault training and those without were observed across several screening and risk assessment items.Practitioners with sexual assault training reported greater frequency of assessing risk of IPSV perpetration, greater frequency of assessing risk of IPSV perpetration compared to other forms of DFV, and greater confidence that IPSV perpetration is identified where present.These findings indicate that, among participants, sexual assault training enhanced screening and risk assessment with perpetrators.Median scores for frequency of assessing risk of IPSV victimisation and confidence that IPSV victimisation is identified where present were higher among practitioners with sexual assault training, although the differences were not statistically significant.

Barriers to Screening and Risk Assessment
Thirteen practitioners raised issues with service coordination and assessments conducted by referring agencies as a barrier to IPSV identification.For example, one practitioner reflected that 'no referral agency we deal with (The Orange Door [government delivered DFV service in Victoria, Australia], Corrections, Child Protection, The Magistrates Court) ever identifies sexual violence in IPDFV [Intimate Partner Domestic and Family Violence]' (Practitioner 21, worked with perpetrators).This practitioner explained that sexual violence would only be identified in the referral if there was a specific charge related to sexual offending.They reiterated the siloing of sexual violence and DFV, stating: 'if there is, it isn't a charge connected with the IPDFV'.Further illustrating issues with service coordination, this practitioner continued: Perhaps this is a barrier in our MBCP Assessments, in that, because this form of violence is unaddressed, we don't raise it either.We work (perhaps unconsciously) with the notion that if the above agencies have done their screening and referrals and sexual violence isn't mention [sic] then it must not have happened.(Practitioner 21, worked with perpetrators) This example highlights how assumptions about who is responsible for screening and risk assessment can result in missed opportunities to assess risk.The siloing of work between agencies was similarly raised by other practitioners: In many cases there is limited involvement from other agencies so the entire picture is often limited.(Practitioner 18, worked with perpetrators) Not all clinicians/practitioners screen for it and this can cause a division between services as not everyone may be on the same page.(Practitioner 55, worked with perpetrators and victim-survivors) It has been argued that greater DFV-informed perpetrator screening and risk assessment is needed across referring agencies (Domestic and Family Violence Death Review and Advisory Board 2021; Meyer et al., 2023).Findings presented here reiterate the importance of screening for IPSV (and DFV) across services, including those in contact with identified DFV perpetrators.Practitioners in the present study reflected that disclosures of IPSV 'may not come up in initial screening' (Practitioner 39, worked with perpetrators and victim-survivors) or 'early in the intervention' (Practitioner 10, worked with perpetrators and victim-survivors).Rather, practitioners reflected that disclosures were 'more likely to come up once a relationship has been established with the client' (Practitioner 39, worked with perpetrators and victim-survivors).Therefore, screening and risk assessment should not be restricted to intake or referral assessments but treated as a continual process.
A second barrier raised by practitioners was non-disclosure.Practitioners predominantly spoke about non-disclosure in relation to feelings of shame.Practitioners described how men are more likely to disclose perceived 'less severe' behaviours and hide those 'they feel shame around or that they think they will be judged for disclosing' (Practitioner 20, worked with perpetrators).Shame was identified as a barrier to disclosure from both victim-survivors and perpetrators.For example, one practitioner commented: The unique challenge is often shame, a woman or PUV [person using violence] will talk about other behaviours unless you expressly ask about sexual violence and even then they may not disclose.(Practitioner 71, worked with perpetrators and victim-survivors) Practitioners' experiences of shame as a barrier to disclosure aligns with existing research that suggests both victim-survivors and perpetrators associate IPSV with shame (Cox 2015;Wright et al. 2022).Other factors influencing non-disclosure included perpetrators' fear of legal consequences and practitioners' own hesitation to raise IPSV out of fear this may escalate the DFV risk for victim-survivors.
A third barrier to screening and risk assessment of IPSV is perpetrators' and victim-survivors' understandings of consent and of what is 'normal' in intimate relationships.As one practitioner explained, 'sexual violence may not be identified by the people perpetrating and experiencing it' (Practitioner 20, worked with perpetrators).Other practitioners commented: These framings were linked to the 'gendered societal parameters of what is acceptable and "normal"' (Practitioner 65, worked with perpetrators and victim-survivors) and which influence understandings of consent and IPSV.For instance, the belief that being in a relationship creates an entitlement to, or a duty for, sex (Our Watch, 2021).
Practitioners often framed the issue as a 'lack of understanding' of consent, as one practitioner explained: People say they consent to sex whenever their partner wants it so they don't fight or get angry not understanding this consent is coerced and therefore not consent.(Practitioner 45, worked with victim-survivors) Other practitioners discussed the need to create space for victim-survivors' varied understandings of their experiences and 'leaned in' to the inconsistencies in how consent and IPSV were understood.These practitioners spoke about the ways in which victim-survivors' own contexts and framings shape: … how they're making sense of something and how they communicate that and how processes that do not allow for women's own narratives can contribute to the erasure of women's voice, choice, and agency to define her own reality.
(Practitioner 47, worked with perpetrators and victim-survivors) Relatedly, practitioners described the importance of positioning victim-survivors 'as the best judge of personal safety and need' (Practitioner 64, worked with victim-survivors) and communicating in a way that aligns with victim-survivors' understandings of their experiences.Importantly, practitioners found a focus on shared understandings and language similarly useful for enabling IPSV disclosures from perpetrators: I found the language used it [sic] especially important and the examples given, men will admit to "being pushy" for sex but will deny using coercion or being sexually abusive.(Practitioner 62, worked with perpetrators and victim-survivors) In the context of perpetrator disclosures, this practitioner described how an admission of IPSV was more likely when perpetrators could externalise this behaviour; for example, '"she tells me I'm pushy" is more likely than the man directly admitting to being pushy' (Practitioner 62, worked with perpetrators and victim-survivors).While the externalisation of 'the problem' also needs to be addressed, this example highlights how men's own framings of sexual encounters may act as an entry point to identifying IPSV.
In thinking about the importance of language, Practitioner 47, who had lived experience of IPSV, reflected on how the 'forced' terminology (common in assessments) could be 'too big a jump': In the landscape of disorientation, confusion, self-alienation, isolation-the barometer for even knowing what I wanted or didn't want was smashed-"forced" was an unintelligible construct.(Survivor/Practitioner 47, worked with perpetrators and victim-survivors) Reflecting on their experiences, they continued: I was often the one initiating sex.I wanted the sex but not the pain, injury, and pregnancy.I would repeatedly ask for him to use a condom and lubricant but my voice/requests were ignored.The power dynamic in the relationship erased my standing ground and voice.I wanted the connection and touch but not the sex.(Survivor/Practitioner 47, worked with perpetrators and victim-survivors) This example raises important questions about how we conceptualise sexual violence and how we create space for the messiness and the fluidity with which victim-survivors practice agency, autonomy, and sexual desire in a context of power, control, and violences.

Discussion
Research findings underline the influence of assessment tool(s) on practice and the need to further develop tools that support meaningful assessment of IPSV.In speaking to the limitations of existing tools, practitioners described the 'blunt' nature of assessment questions, and the focus on 'force' and sexual assault.The focus may reflect that 'forced sex' is a potential indicator for increased risk; however, it does not reflect the breadth of sexual violences that may be experienced in the context of DFV, eclipsing 'routine intimate intrusions' (Kelly 2012: xix) and sexual coercion (Hamilton et al. 2023).Further, such preoccupations with 'force' may reinforce persistent misunderstandings of IPSV as physically forceful or injurious (Waterhouse et al. 2016) and may impede attempts to support victim-survivors and hold perpetrators to account (Hamilton and Tidmarsh 2022;Logan et al. 2015).
Aligning with existing research (Cox 2015), descriptive statistics suggest lower frequency of assessing risk of IPSV compared to other forms of DFV.Descriptive statistics also suggest greater frequency of assessing victimisation compared to perpetration.Although these findings should be interpreted cautiously (and notwithstanding that t-test results were not statistically significant), descriptive statistics align with existing research on DFV screening and risk assessment more broadly (Meyer et al. 2023).Meyer et al. (2023) suggest that the emphasis on screening and assessing risk of victimisation may reflect the greater focus on developing and improving victim-survivor practice over the years, while perpetration is comparatively an emerging priority area.Irrespective of the reasons, the potential emphasis on assessing victimisation risks victim-focused (rather than perpetrator-focused) risk management outcomes (Messing and Thaller 2013;Storey et al. 2014).This may undermine attempts to shift the burden of responsibility from victim-survivors to perpetrators.
Study findings also highlight the potential value of specialist sexual assault training in supporting practitioners to screen and assess risk of IPSV perpetration.Statistically significant differences between practitioners with sexual assault training and those without were observed across the three perpetrator-related screening and risk assessment items.Median scores differed for some victimisation items; however, these differences were not statistically significant.These findings suggest that sexual assault training improves screening and risk assessment of perpetration.The findings appear to indicate that sexual assault training may not have the same effect on screening and risk assessment of victimisation.However, the sample size for the victimisation items was smaller than the perpetration items and may simply not have had the statistical power to detect significant differences.
Qualitative findings highlight the siloing of screening between services, assumptions about who is undertaking this work, and the potential for screening to be missed as a result.This highlights the need for communication, collaboration, and information sharing between agencies and for screening and risk assessment to be conceptualised and practiced as an ongoing process across agencies.This is particularly important in the context of IPSV as, even if screened, these behaviours may not be identified at previous points of service contact.Indeed, practitioners in the present study described the likelihood of disclosures increasing over time once rapport has been established.Depending on the nature and extent of contact with referring services this may not have been possible.
Relatedly, practitioners spoke about non-disclosure as a key barrier to identifying IPSV and this was often linked to shame and discomfort.Some practitioners also reflected on their hesitation to discuss IPSV with perpetrators out of fear of escalating risk to victim-survivors.Victim-survivors' safety should of course be the primary concern.However, practitioner risk aversion may impede the exploration of, and attempts to address, perpetrators' use of IPSV.This is not simply to propose raising IPSV with perpetrators without concern for possible ramifications.Rather, this highlights the need to explore what safe exploration of IPSV with perpetrators can look like.Woodlock et al. (2023) argue that a narrow focus on risk in DFV practice can shift attention away from the pursuit of freedom.That is, from the work that 'create[s] the conditions that enable women and children to be free from male violence, encapsulating the broader feminist project of women's liberation' (Woodlock et al. 2023: 2).Not exploring IPSV perpetration may reflect a prioritisation of safety and a desire to minimise escalating risk; however, it may also result in IPSV continuing (as hidden and unaddressed) and, in this way, impede upon the more aspirational 'freedom work' (Woodlock et al. 2023).In the context of working with victim-survivors, Woodlock et al. (2023: 11) highlight the importance of 'integrated safety-freedom work' and collaborative, survivor-led practice.
There is work to be done in examining how 'safety-freedom work' is, and can be, practiced in work with perpetrators.
Qualitative findings also illustrate consent as a key concept practitioners navigate in screening for, and assessing the risk of, IPSV.Practitioners described their clients' (both victim-survivors and perpetrators) 'lack of understanding' of consent and the psychoeducational role they played as a result.Such accounts were often linked to societal understandings and norms.For example, practitioners are likely to come across perpetrators who may believe they are entitled to sex with a partner, a sense of entitlement they may not assume towards others (Easteal and McOrmond-Plummer 2017;Parkinson 2017).
However, the analysis also reveals some resistance to deficit-focused, 'lack of understanding' framings.Here, practitioners focused instead on navigating the varied understandings of consent, acceptable sexual behaviours, and sexual violence that they encountered in their work with victim-survivors and perpetrators.This approach recognises 'that [victim-survivors] may name the similar experiences differently' (Kelly 2012: xviii).This deficit-resistant framework reflects inconsistencies in definitions and understandings of IPSV and consent found in the wider literature (see for example, Bagwell-Gray et al. 2015).Such inconsistencies have implications for how IPSV is communicated, and the specific language used between practitioners and clients (whether victim-survivors or perpetrators).
Some scholars argue that a shift away from the focus on consent is necessary (Jeffrey 2022;Mason 2023).Jeffrey (2022) argues that 'consent' is co-opted by perpetrators to justify sexual violence and blame victim-survivors for not clearly communicating their lack of consent.There are attempts to address these issues through, for example, affirmative consent reforms.However, Jeffrey (2022: 5) argues that 'a consent focus is insufficient for disrupting deeply entrenched social norms and practices' and should not be the defining feature of ethical sex.Instead of framing the issue in terms of deficit-focused understandings of consent, there is an opportunity to focus on shared and flexible language.Flexibility is essential for recognising that the language that works for one person may be 'unintelligible' to someone else (Survivor/Practitioner 47, worked with perpetrators and victim-survivors).Although not specific to IPSV, one study found 71.9% of IPV and sexual violence victim-survivors surveyed found practitioners' terminology to be a barrier to help-seeking (Hegarty et al. 2022).
The disconnect between victim-survivors' understandings and the language of services necessitates creating space for 'one's own subjective experience and process of negotiation' (Survivor/Practitioner 47, worked with perpetrators and victimsurvivors).This point has been argued in the context of victim-survivors' understandings.For example, Bagwell-Gray et al. (2015: 331) argue that it is critical for '[w]omen's own definitions and conceptualizations of IPSV [to] also inform the terms and definitions' used (Bagwell-Gray et al. 2015: 331).The process of identifying and assessing IPSV should be a collaborative process (Bagwell-Gray et al. 2015).Sex should be a collaborative practice (Jeffrey 2022), and sexual violence involves a denial of that collaboration, so it is important that collaborative practice is not also denied by services working with victim-survivors.
The need to create space for clients' accounts of their experiences was predominantly discussed in the context of working with victim-survivors; however, study findings also illustrate the applicability of some of these practices for working with perpetrators.For example, some practitioners described how allowing men's representations of sexual encounters, including attempts to externalise their actions, could act as an entry point to identifying and addressing IPSV.The use of 'exculpatory discourses' in an attempt to 'mitigate their own culpability' (Cavanagh et al. 2001: 711) through denial, minimisation, and/or externalisation are common in men's accounts of DFV (Seymour et al. 2021).Existing literature has highlighted how this externalising behaviour can act as an 'entry point' for intervention.For example, mental health, problematic alcohol and/ or other drug use (AOD), and couples counselling pathways (and the related potential framing of the 'issue' as one of mental health, AOD, or 'relationship issues' rather than DFV) have been framed as valuable entry points for identifying and responding to DFV (Meyer et al. 2022;Wendt et al. 2020).Leaning in to the 'messiness' of both victim-survivors' and perpetrators' accounts is not to blur the demarcation between sex and violence.Rather, this is about creating space to explore IPSV and its harms in a way that shifts the emphasis away from rigid, 'unintelligible' framings and binaries towards a framework that allows breadth and flexibility.This requires embedding time and space to talk about IPSV, recognising that, for many practitioners working with both victim-survivors and perpetrators, 'a lack of time to explore this risk' and build rapport (Practitioner 65, worked with perpetrators and victim-survivors) is a significant barrier.

Conclusion
Limited research has examined how DFV perpetrator interventions identify, assess, and address sexual violence, including IPSV; as Westmarland and Kelly (2017) note, this area has received considerably less attention than sex offender treatment programs.The present study contributes to our understanding of this knowledge gap through an analysis of screening and risk assessment practices related to IPSV.Drawing on survey data from 97 practitioners working in the DFV perpetrator intervention space in Australia, this paper demonstrates significant gaps in screening and risk assessment of IPSV, particularly compared to other forms of DFV.Such shortcomings contribute to concealing this form of violence.Nationally in Australia, there are calls 'to bring addressing sexual violence out of the shadows' (DSS 2022: 22), yet this paper illustrates the significant work that remains in identifying and addressing IPSV perpetration.
This study demonstrates limits to the utility of DFV assessment tools for identifying and assessing risk of IPSV, highlighting the need for tools that better support exploratory conversations and disclosures.Study findings also illustrate variation in assessment practice, with IPSV assessed less often than other forms of DFV, and IPSV victimisation assessed more frequently than IPSV perpetration.These findings highlight the need for focused attention on IPSV perpetration.Alongside improved assessment tools and guidance, study findings also suggest that further training drawing together, and building upon, existing sexual assault and DFV training could increase screening and risk assessment practice of IPSV perpetration.Understandings of consent were raised as a key barrier to assessing IPSV and demonstrate the challenge practitioners face in navigating the varied understandings held by victim-survivors and perpetrators.There is a critical need to find ways to meaningfully address IPSV in work with perpetrators, without compromising victim-survivor safety, and for further research to examine what safe exploration of IPSV with perpetrators looks like.
Recognition of the intersections of DFV and sexual violence behaviours is critical and has implications for how we understand, identify, and respond to IPSV.Indeed, research with IPSV victim-survivors already establishes the unique harms of this form of abuse, and the inadequacy of responses when IPSV is contained within siloed sexual violence or DFV responses (Cox 2015;Macy et al. 2009;McOrmond-Plummer 2013).This intersection also has implications for responses to perpetrators; the siloing of responses to either sexual offending or DFV pathways can result in many forms of sexual violence, as well as the intersections of sexual violence and DFV, remaining invisible.Research by Carney et al. (2006) suggests that sexual violence perpetrators may be at increased risk of disengagement from DFV interventions, further highlighting the need to identify this violence.Given the co-occurrence of DFV and IPSV, greater attention needs to be paid to how DFV perpetrator interventions identify and address IPSV alongside other forms of DFV.
1 This is not to minimise the harms of non-sexual intimate partner violence (IPV) or to suggest a hierarchy of harm.Rather the purpose here is to highlight the unique harms of IPSV. 4 Participants could select as many role(s) as applicable and, therefore, totals are greater than the full sample (n=97, 100.0%). 5Other assessment tools used by practitioners working with perpetrators included the Common Risk Assessment Framework (CRAF, n=17, 28.8%), internally developed organisational tools (n=10, 16.9%), the Risk, Safety and Support Framework (RSSF, n=4, 6.8%), the Spousal Assault Risk Assessment (SARA, n=2, 3.4%) tool, the Domestic Violence Safety Assessment Tool (DVSAT, n=2, 3.4%), the Safe & Together perpetrator mapping tool (n=2, 3.4%), and the Abusive Behaviour Inventory (n=2, 3.4%).Among practitioners working with victim-survivors, other assessment tools used included the DVSAT (n=6, 16.2), CRAF (n=5, 13.5%), Safe & Together (n=2, 5.4%), and internally developed organisational tools (n=2, 5.4%).The tools in use reflected (in part) jurisdictional differences; for example, that the MARAM is the predominant tool used in Victoria, where most study practitioners were based. 6These analyses were not conducted on the frequency of assessing risk of IPSV victimisation items due to the small sample size of practitioners working with victim-survivors only (n=6).
7 Noting an inconsistency in terminology used between these two items (a history of perpetrating IPSV; IPSV victimisation is always identified).

Table 1 : Frequency of assessing risk of IPSV perpetration and victimisation
Most practitioners reported that they assessed risk of IPSV perpetration (n=43, 66.1%) and IPSV victimisation (n=32, 78.0%) 'always' or 'most of the time', as shown in Table1.Notably, one in five practitioners working with perpetrators (n=13, 20.0%) and one in 10 practitioners working with victim-survivors (n=5, 12.2%) reported 'rarely' or 'never' assessing risk of IPSV perpetration or victimisation, respectively.

Table 2 : Frequency of assessing risk of IPSV perpetration and victimisation compared to other forms of DFV
a missing=1.Table2 reproduced from

Table 3 : Screening, risk assessment, and intervention practice items by sexual assault training Sexual assault training Total Mann-Whitney U Test
a Asymptotic significance is displayed.*Statisticallysignificantatp<0.05.**Statistically significant at p<0.005.Table3adapted fromHelps et al. (2023: 60) Stereotypes and myths as well as rigidly held beliefs by perpetrators, victims, the public and unfortunately, practitioners who work in the field.(Practitioner 34, worked with perpetrators) Some AFMs [affected family members] believe men have a right to sex if married or within a relationships [sic], womens [sic] sexual needs and boundaries are not prioritised by society.(Practitioner 64, worked with victim-survivors) 2The MARAM framework, implemented in Victoria following the Royal Commission into Family Violence (2016), seeks to improve victim-survivor safety via greater collaborative practice identifying, assessing, and managing DFV (for more information, see https://www.vic.gov.au/training-for-information-sharing-and-maram).Developed in the United States, the Safe & Together model frames DFV perpetration as a parenting choice and centres victim-survivor strength and child wellbeing.The model intends to improve intervention in DFV perpetration (for more information, see https://safeandtogetherinstitute.com/the-sti-model/model-overview/). 3 NTV is the peak men's behaviour change body in Australia.For further details on their training programs, see https://ntv.org.au/trainingand-professional-development/.