Family violenceFootnote 1 is prevalent (Finneran & Stephenson, 2013; Leonard et al., 2008; Szalacha et al., 2017; Victorian Agency for Health Information 2020; Walters et al. 2013) yet under-reported (Workman & Dune, 2019) within lesbian, gay, bisexual+, trans and queer (LGBTQ) communities. Family violence is used in this paper as an umbrella term encompassing both violence from a family member and violence from an intimate partner (Department of Communities, 2020; Victorian Governement 2022). Findings from previous international research suggest that LGBTQFootnote 2 people experience similar, if not higher, rates of family violence than heterosexual and/or cisgender people (Barnes & Donovan, 2018; Donovan & Barnes, 2019; Donovan & Hester, 2014; Finneran & Stephenson, 2013). There is also some evidence both nationally and internationally to indicate that trans and gender diverse people may experience even higher rates of family violence than cisgender people from sexual minority populations (Calton et al., 2016; Langenderfer-Magruder et al., 2016; Safe Steps 2015). Despite this elevated nature of need, access to family violence services by LGBTQ people is limited. A previous Australian study of people in LGBT relationships found that 53.5% of female respondents and 67.1% of male respondents who had experienced intimate partner violence did not seek any form of support (Farrell & Cerise, 2007). Of those who did seek support, few sought this from family violence services, opting instead for counsellors or friends and family (Farrell & Cerise, 2007). A scoping review of existing family violence literature has identified three main barriers to accessing family violence support among LGBTQ people (Calton et al., 2016). These include a lack of evidenced-based knowledge on the family violence experiences and support needs of LGBTQ people; stigma regarding LGBTQ identities within the general population and among family violence responders and service providers; and systemic inequities within the family violence sector. Consequently, LGBTQ people are less likely than non-LGBTQ peers to seek support from mainstream agencies, such as the police and family violence services (Barnes & Donovan, 2018; Donovan & Barnes, 2020; Donovan & Hester, 2014). A qualitative study involving interviews of LGBTQ victim-survivors of intimate partner violence found that participants frequently relied on self-care and maintained low-expectations of mainstream services for family violence support (Donovan & Barnes, 2020).

Historically, family violence policy and services have been grounded in heteronormative assumptions of the relationship and gender of both perpetrator and survivor (Rainbow Health Victoria, 2020). These assumptions have led to a focus on violence toward cisgender (i.e., gender identity aligns with that assigned at birth) women as perpetrated by cisgender men, and exclude the experiences of people who identify as LGBTQ or are in non-heterosexual relationships (Seymour, 2019). Family violence responders may perceive reports from LGBTQ people as less serious or not requiring the same support as reports from heterosexual cisgender women. Previous research has found both laypersons (Hamby & Jackson, 2010; Russell et al., 2012; Seelau & Seelau, 2005) and police officers (Cormier & Woodworth, 2008; Russell, 2018) identify male-to-female intimate partner violence as more serious and threatening than female-to-male intimate partner violence or intimate partner violence in same-sex relationships. Additionally, victims of non-male perpetrators can be seen as less credible in their claims of violence (Russell, 2018). These findings illustrate that the gender of the victim-survivor and that of the perpetrator is likely to impact the quality of the support received, suggesting that some subsections of the LGBTQ community may experience greater barriers and experiences of discrimination in the family violence service response setting than others.

Efforts are underway to improve family violence support for LGBTQ populations in Australia (Rainbow Health Victoria, 2020), however the coverage of LGBTQ-inclusive practices currently remains limited in most states and territories. While there has been recognition of LGBTQ communities in the state of Victoria’s family violence frameworks and strategic plans (Victorian Government and Family Safety Victoria 2017), most states and the commonwealth government have insufficiently engaged with the needs of LGBTQ communities. A distinct lack of attention to LGBTIQ experiences of family violence was recognised by the Royal Commission into Family Violence in Victoria, conducted in 2016, which highlighted the paucity of services available to these communities (State of Victoria 2016). While little research has explored the specific role of culturally safe family violence services among LGBTQ people, research focused on the role of Aboriginal Community Controlled Health Organisations (ACCHOs) as well as HIV community-controlled organisations (i.e., AIDS Councils) suggests the importance of ensuring the availability of culturally safe services. For example, Australian based research has shown ACCHOs are frequently reported to be the most preferred, and effective providers of health services for First Nation’s People (Aspin et al., 2012; Nguyen et al., 2016). Similarly, AIDS councils, which play an important role in the healthcare setting in Australia and represent the needs of LGBTIQ communities’ across community-based, government, and clinical spheres (Griew & Griew, 2008) have been shown to have a significant positive impact on the wellbeing of people living with HIV through their delivery of trusted, knowledgeable and community-led healthcare (Australian Federation of AIDS Organisations (AFAO), Australia’s State and Territory AIDS Councils 2016).

The already limited family violence services available in Australia predominantly cater to heterosexual cisgender women (hereafter referred to as ‘mainstream’ services), with a limited number of services catering specifically to LGBTIQ people (hereafter referred to as ‘LGBTIQ-specific’ services) or providing LGBTIQ-inclusive care (i.e., those that have had specific training on LGBTQ issues or those that may be recommended by LGBTQ users to their peers; hereafter referred to as ‘LGBTIQ-inclusive’ services). These services are further inadequately funded and may not exist outside of capital cities (State of Victoria 2016), further hindering their accessibility and the opportunity for LGBTQ victim-survivors of family violence to choose a service best suited to them. Moreover, preferences for the type of family violence service available are likely to differ across the LGBTQ community. These preferences may be shaped by individual sociodemographic characteristics, sociocultural experiences or prior experiences with family violence or other services, such as healthcare, welfare, or homelessness services.

To improve access to family violence services for LGBTQ people, it is essential that these services are safe, accessible, and appropriate to LGBTQ needs and, therefore, must be grounded in preferences for service provision expressed by LGBTQ people. The present study explores the family violence service preferences of LGBTQ adults residing in Australia. It focuses on comparison between preferences for mainstream family violence services, mainstream LGBTQ-inclusive services, and LGBTIQ-specific services. Specifically, the study aims to detail the subsections of the LGBTQ community that hold particular preferences, exploring factors associated with service preferences, including demographics, experiences of discrimination and harassment, experiences of homelessness, access to a regular general practitioner (GP) and whether participants felt supported when previously reporting family violence. The outcomes of this study are expected to have important implications for informing family violence policy and guiding the development and provision of services that appropriately cater to the needs of LGBTQ communities.

Methods

Sample and Procedure

Participants were from the [redacted for peer review] survey, which was conducted by authors of this paper and designed in consultation with an Expert Advisory Group and Gender Advisory Board. This cross-sectional Australia-wide survey examined the health and wellbeing of 6,835 LGBTIQ + Australians aged 18 years or older. The survey was completed by participants from all states and territories and was approved by the [redacted for peer review] Human Research Ethics Committee. The survey ran from July to October 2019. Participants were recruited through promotion by LGBTIQ community organisations and paid targeted advertising on Facebook and Instagram.

Measures

Demographics

Demographic characteristics include gender, age, area of residence (inner suburban, outer suburban, regional, and rural or remote), country of birth (Australia, other English-speaking country, other non-English speaking country), level of education, weekly net income and employment status (currently employed or not). To examine sexual orientation, participants were asked, ‘Which best describes your sexual orientation?’ Twelve options were provided, including the option to choose not to answer, and participants were ultimately categorised as either ‘gay or lesbian’, ‘bisexual’, ‘pansexual’, ‘queer’, ‘asexual’, and ‘something else.’ Given that participants who identified as gay were predominantly men and those who identified as lesbian were predominantly women, gay and lesbian were combined into a single mono-sexual orientation category for the purpose of the multivariable logistic regression analyses (as described below). This allowed for the simultaneous examination of gender and sexual orientation in the regression model, without any potential for confounding or bias related to gender.

For gender identity, participants selected a gender term that best described them from a list of options in addition to indicating the sex on their original birth certificate. Based on these responses, gender was then categorised as: cisgender woman (assigned female at birth and identified only ‘female’ as their gender identity); cisgender man (assigned male at birth and identified only ‘male’ as their gender identity); trans woman (assigned male at birth and identified only ‘female’, ‘trans woman’ or ‘sistergirl’ as their gender identity); trans man (assigned female at birth and identified only ‘male’, ‘trans man’ or ‘brotherboy’ as their gender identity); or non-binary (participants identified as only a gender that was not a binary identity or indicated that they could not choose a single gender identity).

Regular GP

Participants indicated if they have a regular GP, by selecting ‘Yes’, ‘No I don’t have a regular GP, but I attend the same health centre’ or ‘No I don’t have a regular GP, and I attend different health centres’. Responses were coded into a dichotomous ‘Yes’ or ‘No’ variable.

Homelessness

Participants indicated if they have ever or are currently experiencing homelessness. Response options included ‘No’, ‘Yes – once and I am not currently experiencing homelessness’, ‘Yes – more than one, and I am not currently experiencing homelessness’, ‘Yes – I am currently experiencing homelessness for the first time’ and ‘Yes – I am currently experiencing homelessness and have also previously experienced homelessness’. Responses were categorised into a dichotomous variable to indicate whether or not participants had ever experienced homelessness.

Stigma, discrimination, and abuse

Overall experiences of discrimination were assessed by asking, ‘In the past 12 months, to what extent do you feel you have been treated unfairly because of your gender identity?’ and ‘In the past 12 months, to what extent do you feel you have been treated unfairly because of your sexual orientation?’ Participants responded on a five-point scale from ‘not at all’ to ‘very often.’ To facilitate analysis, responses of ‘not at all’ were coded as ‘No’ and all other responses as ‘Yes.’ Responses to the two questions were then combined to indicate whether participants had experienced any unfair treatment based on their sexual or gender identity in the past 12 months.

Additionally, participants were asked if they had experienced any social exclusion or verbal abuse (including hateful or obscene phone calls) based on their sexual orientation or gender identity in the past 12 months. These were coded into dichotomous ‘yes’ or ‘no’ variables.

Felt supported when reporting

To identify those participants who had experienced family violence, participants were asked ‘Have you experienced any of the following from family members? (Choose as many as apply)’ and ‘Have you experienced any of the following from intimate partner(s)? (choose as many as apply)’ Participants chose from 10 forms of violence, including ‘Physical violence’, ‘Verbal abuse’, ‘Sexual assault’, ‘Financial abuse’, ‘Emotional abuse’, ‘Harassment or stalking’, ‘Property damage’, ‘Social isolation’, ‘Threats of self-harm or suicide’ and ‘LGBTIQ related abuse’, with the option to indicate that they had not experienced any of these forms of violence. Participants who selected at least one of the 10 forms of violence were considered to have experienced family violence.

Those participants who reported any experience of violence from an intimate partner or family member were further asked to respond to the question: “The most recent time you experienced abusive behaviour from a family member or intimate partner, did you report it to any of the following? (Choose as many as apply)”. Those who indicated that they had reported their most recent experience of family violence were subsequently asked, ‘The most recent time you reported abusive behaviour from a family member or intimate partner to the following, did you feel supported? (Choose as many as apply)’. Participants were provided with a list of 11 authorities, support services and community leaders, such as ‘Doctor or hospital’, ‘Domestic or family violence service’, and ‘Teacher or educational institution’, with the additional option of ‘Other,’ and asked to indicate for each relevant item ‘I felt supported’ or ‘I did not feel supported’. Responses to this item were combined into ‘Yes’ or ‘No’ indicating whether participants had felt supported after reporting their experience. A category of ‘Not applicable’ was included to indicate those who had not previously reported experiences of family violence to anyone.

Preference for family violence services

All participants, regardless of whether or not they reported experiencing any violence from a family member or intimate partner, were asked ‘If you were to ever need help or support in relation to abuse from family member(s) or intimate partner(s), where would you prefer to receive it?’ Response options included ‘From a mainstream domestic violence service’, ‘From a mainstream domestic violence service that is known to be LGBTIQ inclusive’, ‘From a service that only caters to lesbian, gay, bisexual, transgender, and/or intersex people’, ‘I don’t know’ or ‘I have no preference.’ Responses of ‘I don’t know’ and ‘I have no preference’ were coded as missing.

Statistical Analyses

Analyses were performed using STATA (Version 16.1, StataCorp, College Station, TX, USA). A series of univariable and multivariable logistic regressions with robust standard errors to account for the variance in sample sizes were used to examine factors associated with preferences for provision of family violence services from one service over another. Multivariable regression analyses were used to explore the factors associated with these preferences while controlling for the confounding impacts of all other predictors variables in the model. Three binary outcomes were explored using the regression models: (1) preference for mainstream or LGBTQ inclusive services; (2) preference for mainstream or LGBTQ-specific services; and (3) preference for LGBTQ inclusive or LGBTQ-specific services. All remaining study variables were included in these analyses as predictor variables. We used the enter method where variables with unadjusted odds ratios with p-values of < 0.25 in the univariable logistic regressions were entered into the multivariable models. Sociodemographic variables were included regardless of the p-values of the unadjusted odds ratios. Tests of collinearity indicated that multicollinearity was not a concern, with all Variance Inflation Factors (VIFs) < 2. Results are reported as adjusted (multivariable) odds ratios (AORs) with 95% confidence intervals (CIs) and P < 0.05 used to assess statistical significance.

Results

Table 1 presents the frequencies and proportions of the demographic characteristics and preferences for family violence services. More than three quarters of the sample identified as cisgender and nearly half as gay or lesbian. The study population ranged in age from 18 to 88 years, with a mean age of 34.1 (SD = 13.8). The majority were aged under 45 years, born in Australia, currently employed and earned an income under $2,000 AUD net/week. The largest proportion of participants lived in inner suburban areas and almost three quarters had completed tertiary education. In all, 4,148 (61.1% of the total sample) participants held a preference for family violence services, while 2,646 (38.9% of the total sample) were unsure or had no preference. Of those who had a preference, 8.8% (n = 363) indicated a preference for mainstream services, 57.5% (n = 2,383) for mainstream services that are known to be LGBTQ-inclusive services and 33.8% (n = 1,402) for LGBTQ-specific services.

Table 1 Sample characteristics (N = 6,835)
Full size table

Mainstream LGBTQ-Inclusive vs. Mainstream Services

Table 2 presents comparisons for preferences between mainstream services and mainstream services that are known to be LGBTQ-inclusive services. Compared to those who identified as gay or lesbian, participants who identified as queer were more likely to prefer mainstream LGBTQ-inclusive services over mainstream services (AOR = 2.27, CI = 1.02–5.08, p = 0.045), while all other sexual orientation groups except for pansexual were less likely to prefer LGBTQ-specific services over mainstream services (Bisexual: AOR = 0.65, CI = 0.43–0.98, p = 0.038; Asexual: AOR = 0.37, CI = 0.16–0.84, p = 0.017; Something else: AOR = 0.53, CI = 0.31–0.91, p = 0.021). Compared to cisgender men, all other gender groups apart from trans women were more likely to prefer LGBTQ- inclusive services over mainstream services (Cisgender women: AOR = 2.53, CI = 1.75–3.66, p < 0.001; Trans men: AOR = 10.21, CI = 2.45–42.56, p = 0.001; Non-binary: AOR = 3.37, CI = 1.68–6.73, p = 0.001). Compared to 18–24-year-olds, all other age groups were more likely to prefer LGBTQ- inclusive services over mainstream services (25–34: AOR = 2.1, CI = 1.34–3.3, p = 0.001; 35–44: AOR = 1.61, CI = 0.94–2.74, p = 0.081; 45–54: AOR = 3.04, CI = 1.64–5.62, p < 0.001; 55+: AOR = 2.5, CI = 1.36–4.58, p = 0.003). Participants with a non-university tertiary/post-secondary education were less likely than those with a secondary education to prefer LGBTQ- inclusive services over mainstream services (AOR = 0.59, CI = 0.38–0.92, p = 0.020). Participants who had experienced unfair treatment based on their LGBTQ identity were more likely to prefer LGBTQ- inclusive services over mainstream services.

Table 2 Factors associated with preference for LGBTQ-inclusive services over mainstream services not known to be inclusive (n = 1,279)
Full size table

LGBTQ-Specific Services vs. Mainstream Services

Table 3 presents comparisons for preferences between mainstream services and LGBTQ-specific services. Compared to those who identified as gay or lesbian, all other sexual orientation groups except queer were significantly less likely to prefer LGBTQ-specific services over mainstream services (Bisexual: AOR = 0.27, CI = 0.17–0.43, p < 0.001; Pansexual: AOR = 0.34, CI = 0.15–0.79, p = 0.012; Asexual: AOR = 0.13, CI = 0.05–0.38, p < 0.001; Something else: AOR = 0.31, CI = 0.17–0.58, p < 0.001). Compared to cisgender men, all other gender identities were significantly more likely to prefer LGBTQ-specific services over mainstream services (Cisgender women: AOR = 2.03, CI = 1.35–3.04, p = 0.001; Trans men: AOR = 30.88, CI = 7.38-129.13, p < 0.001; Trans women: AOR = 5.33, CI = 2.27–12.52, p < 0.001; Non-binary people: AOR = 9.33, CI = 4.01–21.72, p < 0.001). Compared to 18–24-year-olds, all older age groups were significantly more likely to prefer LGBTQ-specific services over mainstream services (25–34: AOR = 2.60, CI = 1.54–4.38, p < 0.001; 35–44: AOR = 2.01, CI = 1.12–3.6, p = 0.019; 45–54: AOR = 2.66, CI = 1.3–5.47, p = 0.008; 55+: AOR = 2.78, CI = 1.32–5.86, p = 0.007). Compared to participants residing in inner suburban areas, participants residing in outer areas (AOR = 0.64, CI = 0.43–0.97, p = 0.035) and in regional cities or towns (AOR = 0.59, CI = 0.37–0.92, p = 0.020) were significantly less likely to prefer LGBTQ-specific services over mainstream services.

Table 3 Factors associated with preference for a LGBTQ only service over a mainstream service not known to be inclusive (n = 1,957)
Full size table

Crucially, participants who had experienced unfair treatment due to their sexual or gender identity (AOR = 2.77, CI = 1.83–4.18, p < 0.001), verbal abuse (AOR = 1.66, CI = 1.07–2.58, p = 0.023) or social isolation (AOR = 1.54, CI = 1.01–2.34, p = 0.044), were significantly more likely to prefer LGBTQ-specific services over mainstream services.

Mainstream LGBTQ-Inclusive Services vs. LGBTQ-Specific Services

Table 4 presents comparisons for preferences between inclusive services and LGBTQ services. Compared to gay and lesbian identified participants, all other sexual orientation groups except queer were less likely to prefer LGBTQ-specific services over mainstream LGBTQ-inclusive services (Bisexual: (AOR = 0.48, CI = 0.38–0.62, p < 0.001; Pansexual: AOR = 0.44, CI = 0.31–0.62, p < 0.001; Asexual: AOR = 0.4, CI = 0.22–0.69, p = 0.001; Something else: AOR = 0.65, CI = 0.45–0.94, p = 0.021). Compared to cisgender men, all gender groups apart from cisgender women were more likely to prefer LGBTQ-specific services over mainstream LGBTQ-inclusive services (Trans men: AOR = 2.55, CI = 1.7–3.81, p < 0.001; Trans women: AOR = 2.77, CI = 1.82–4.22, p < 0.001; Non-binary people: AOR = 2.15, CI = 1.61–2.87, p < 0.001). Compared to those with secondary only education, participants with university-postgraduate education were more likely to prefer LGBTQ-specific services over inclusive services. Participants who had a regular GP were less likely to prefer LGBTQ-specific services over inclusive services (AOR = 0.83, CI = 0.69–0.99, p = 0.034). Finally, participants who had experienced unfair treatment outside of family violence based on their LGBTQ identity were more likely to prefer LGBTQ-specific services (AOR = 1.61, CI = 1.31–1.99, p < 0.001), while those who had felt supported the last time they reported family violence to someone were less likely to prefer LGBTQ-specific services over inclusive services (AOR = 0.6, CI = 0.39–0.93, p = 0.023).

Table 4 Factors associated with preference for a LGBTQ only service over an inclusive service (n = 2,792)
Full size table

Discussion

More than half (57.5%) of the sample who held a preference for family violence service provision expressed a preference for mainstream services that were known to be LGBTQ-inclusive and approximately one-third (33.8%) expressed a preference for family violence services that were LGBTQ-specific, while less than one-tenth (8.8%) expressed a preference for mainstream services that were not known to be LGBTQ-inclusive. These preferences reflect those previously reported in the extant literature regarding Aboriginal Community Controlled Health Organisations and AIDS Councils, which evidence preferences for and effectiveness of community controlled and culturally competent service provision (Aspin et al., 2012; Griew & Griew, 2008; Nguyen et al., 2016). While the preference outcomes of the present study demonstrate a clear overall preference for mainstream LGBTQ-inclusive or specific services, the findings of the study further illustrate apparent nuances in these preferences, with a number of individual factors found to be associated with preferences for service provision. The results illustrate a greater preference for LGBTQ-specific services among trans and non-binary identified people while bisexual, pansexual and asexual people were more likely to prefer mainstream services that are known to be LGBTQ-inclusive. Additionally, differences in preference for services appear to be dependent on engagement with a regular GP and experiences of support when previously reporting experiences of family violence, with those who had not felt supported when they reported their most recent experience of violence more likely to prefer LGBTQ-specific services.

Mainstream vs. Mainstream LGBTQ-Inclusive or LGBTQ-Specific Services

A small proportion of participants expressed a preference for mainstream family violence services that are not known to be LGBTQ-inclusive. While seemingly counter-intuitive to maintain a preference for these services, we speculate this preference may be driven by a desire for greater access to services, given the greater number of mainstream services available compared to LGBTQ-specific or inclusive services, or a belief that they should be able to access any kind of service and be treated effectively and with respect. Additionally, some participants may not have perceived their LGBTQ identity as relevant to family violence service access. It is possible that some participants were in a relationship that replicates mainstream assumptions, such as a bisexual cisgender woman with a perpetrator who is a cisgender man. Regardless, survey participants expressed an overwhelming preference for either mainstream services that are known to be LGBTQ-inclusive or LGBTQ-specific services over mainstream services. These preferences were particularly evident among lesbian and gay identified participants who were generally more likely than those of other sexual orientations to prefer both mainstream LGBTQ-inclusive and LGBTQ-specific services, as well as queer-identified people who were more likely to prefer mainstream LGBTQ-inclusive services. Compared to cisgender men, cisgender women were more likely to prefer LGBTQ-specific and inclusive services, as were non-binary people and trans men. Trans women were no more likely than cisgender men to prefer inclusive services, but they were more likely to prefer LGBTQ-specific services. Interestingly, preferences for LGBTQ-specific services were much greater than preferences for mainstream LGBTQ-inclusive services among trans men (10x higher odds vs. 30x higher odds, respectively). Additionally, age, residential location and recent experiences of unfair treatment and harassment were all associated with preferences for LGBTQ-specific and inclusive services. These findings highlight the varying needs and experiences of subsections of the LGBTQ community, but most importantly stress the need for the availability of LGBTQ-inclusive and LGBTQ-specific services. Given the clear preference for LGBTQ services over mainstream services that are not known to be LGBTQ-inclusive, we feel it appropriate for this paper to have a greater focus on detailing the factors associated with a preference for LGBTQ-specific or inclusive services. These outcomes, detailed below, can then guide family violence policy and practice initiatives that are most congruent with the preference of LGBTQ communities.

Mainstream LGBTQ-Inclusive vs. LGBTQ-Specific Services

When comparing preferences for inclusive mainstream services or LGBTQ-specific services, more participants expressed a preference for inclusive mainstream services, which may reflect a desire to have the same access to services as the general population and an expectation of receiving safe, affirming, and knowledgeable care within these services. However, preferences between these two types of services differed across subsections of LGBTQ people. Specifically, compared to lesbian and gay identified participants, those who identified as bisexual, pansexual, asexual or something else were more likely to prefer mainstream services that are known to be LGBTQ-inclusive services over LGBTQ-specific services. Participants identifying as lesbian or gay may be more likely than their multi-gender attracted or asexual peers to prefer LGBTQ-specific services, as violence occurring within same-sex intimate relationships may be less likely to be taken seriously by family violence responders who hold heteronormative assumptions about relationship dynamics (Russell, 2018). Previous qualitative research of LGB experiences and preferences in healthcare found explanations given by participants who were comfortable attending non-LGB specific providers included their ability to “pass” as heterosexual, and feeling their sexual orientation was unrelated to the services being sought (Martos et al., 2018). In the context of family violence, these explanations may be more relatable to those who identify as bisexual, pansexual, or asexual.

Subsections of the LGBTQ population who experience greater discrimination, harassment, and a lack of knowledge on the part of service providers may be more likely to want to access care that is specific to LGBTQ people. Recent Australian literature suggests that trans and gender diverse people experience greater discrimination and poor treatment in mainstream health care settings than cisgender peers, including provider ignorance of trans issues (Bartholomaeus et al., 2021; Bretherton et al., 2020; Hill et al., 2020). As such, the present study found that trans and non-binary people were considerably more likely to express a preference for LGBTQ-specific services, with no difference found between cisgender men and women.

Holding a postgraduate research degree was associated with a greater likelihood to prefer LGBTQ-specific services. These findings may result from a greater health literacy among people with higher levels of education and increased awareness or previous use of LGBTQ-specific services, which may have informed their preference for LGBTQ-specific family violence services. However, given a lack of literature on preferences for family violence service provision among LGBTQ populations, these explanations are speculative and require further research to understand the potential drivers of these choices.

Additionally, those with a regular GP were more likely to express preference for inclusive services over other options. This finding may reflect an interest by participants with a regular GP to engage with the mainstream service sector or greater trust in mainstream services as cultivated by positive relationships with a GP. Moreover, participants with a regular GP may experience support from their GP to navigate the health system and are consequently more confident that they will be able to access inclusive services. Accordingly, previous Australian research has shown that sexual minority women with a regular GP were more likely to utilise additional healthcare services such as alcohol and mental health support services (McNair et al., 2018).

Conversely, those who had recently felt that they were treated unfairly due to their gender or sexual orientation and those who had not felt supported when they reported their most recent experience of violence, were more likely to prefer LGBTQ-specific services. Previous research suggests that LGBTQ people who have had negative experiences seeking help for family violence through a mainstream setting will in future not seek formal help, and instead rely on informal help-seeking, such as friends or family (Santoniccolo et al., 2021).

These findings highlight the need for service experiences where LGBTQ people feel that they are well supported, respected, and treated fairly. For many, these services would need to be available in the form of LGBTQ-inclusive mainstream services and for others it would appear necessary that these cater solely to LGBTQ people. To meet this need, adequate resourcing of LGBTQ- controlled organisations and a scale up of the cultural-competency training and accreditation of mainstream services to provide inclusive culturally safe care is required. Cultural competency training that is interactive, multidisciplinary, and community-engaged and led is essential to effectively reduce provider knowledge gaps (Leslie et al., 2017; Seay et al., 2019; Taylor et al., 2018), and must be translated into practice-based frameworks that address LGBTQ-specific needs through affirming attitudes, values, and relational approaches (Keuroghlian et al., 2017).

Limitations and Future Research

The present study adds considerably to the limited existing literature of LGBTQ people’s experiences of family violence and is the first of its kind to explore, in detail, preferences for family violence service provision in Australia. Additionally, the findings are based on the largest nationwide survey of LGBTQ adults in Australia. However, given the size of the [redacted for peer review] survey and the many aspects of LGBTQ lives that were covered by this survey, there was a limit to how much information was collected on family violence and family violence service experiences specifically. Consequently, the study did not collect information on recent or current experiences of family violence or details of the family violence services that had been accessed. Specifically targeted family violence surveys and qualitative explorations of experiences with receiving family violence support from services as well as the drivers for service preferences may provide further useful information, especially in helping to ensure services resolve barriers and meet the needs of LGBTQ clients.

Conclusion

It is essential that LGBTQ people have access to services that are safe and supportive of their needs, irrespective of their circumstances or identity. To support all subsections of the LGBTQ community, the present study highlights a necessity for services that specifically cater to the needs of LGBTIQ people (such as LGBTIQ community-controlled organisations) as well as ensuring that mainstream family violence services are attentive to and inclusive of LGBTQ people and can provide a suitable standard of care. This will require a reform in narrative around family violence, which at present excludes LGBTQ people and relationships (ref Donovan and Hester, 2010), training of mainstream services to provide inclusive care and the implementation or scale-up of LGBTQ-specific family violence services, including shelters and safe spaces for survivors to receive support. The findings of this study can be used to support family violence policy, practice, and training initiatives.