Demographic predictors of experiences of homelessness among lesbian, gay, bisexual, trans, gender-diverse and queer-identifying (LGBTIQ) young people in Australia

Homelessness among young lesbian, gay, bisexual, trans, gender-diverse and queer-identifying (LGBTQ+) persons is highly prevalent and constitutes a structural risk to health and future life chances. However, the distribution of homelessness burden is among di ﬀ erent LGBTQ+ subgroups is poorly understood. An Australia-wide cross-sectional online survey was conducted involving 6,481 LGBTQ+ participants aged 14 – 21 years during 2019. Single-predictor logistic regression analyses identi ﬁ ed factors associated with both lifetime and recent experiences of homelessness. Analyses also explored associations between recent (<12 months) experiences of homelessness, experiences of harassment, alcohol consumption, and psychological distress. Higher odds of experiencing homelessness were observed for trans and gender-diverse young people, individuals who identi ﬁ ed with sexual identity labels other than lesbian, gay or bisexual, racially-minoritized persons, disabled persons and individuals from a religious family or household, compared to their respective counterparts. Experiencing homelessness was associated with higher levels of alcohol consumption and higher prevalence of experiencing verbal, physical and sexual harassment, but only modestly associated with higher levels of psychological distress. Homelessness risk and burden is unevenly distributed among LGBTQ+ youth and is linked to outcomes which may potentiate future homelessness. Interventions addressing homelessness among this group must be optimized for those subgroups most vulnerable to experiencing homelessness.


Introduction
As defined by the Australian Bureau of Statistics, an individual is considered homeless if they: do not have suitable accommodation alternatives and their current living arrangement is in a dwelling that is inadequate, has no tenure, or if their initial tenure is short and not extendable; or does not allow them to have control of, and access to, space for social relation.(Australian Bureau of Statistics 2018) Globally, homelessness is a preventable yet persistent and growing public health concern (Bassuk, Hart, and Donovan 2020;Fransham and Dorling 2018) that has considerable immediate and long-term consequences for the mental (Babulal et al. 2022;Schreiter et al. 2021;Vallesi et al. 2021) and physical (Johnson, Drew, and Auerswald 2020;Seastres et al. 2020) health of affected individuals (Narendorf et al. 2022), families (Sylvestre et al. 2018) and communities (Fowler et al. 2019).Crucially, homelessness potentially undermines the future life chances and long-term socioeconomic opportunities of affected persons (Nilsson, Nordentoft, and Hjorthøj 2019), in turn reducing their likelihood of exiting homelessness in the future (Dobson 2022).A crucial determinant of the success of social and policy interventions to reduce and/or prevent homelessness is the accurate identification of groups and individuals that are vulnerable to experiencing housing insecurity.
One group of persistent and particular concern is lesbian, gay, bisexual, transgender, and queer (LGBTQ+) youth; despite a lack of international population-representative data homelessness (McCann and Brown 2021) on the prevalence of homelessness experiences among this group, estimates from some jurisdictions (e.g. the U.S.) suggest that LGBTQ+ youth are 120% more likely than their cisgender and/or heterosexual counterparts to experience (Ormiston 2022).In Australia, the 2014 General Social Survey indicated that 33.7% of respondents who identified as lesbian or gay, and 20.8% of bisexuals, reported that they had ever experienced homelessness, compared with 13.4% of those identifying as heterosexual (Australian Bureau of Statistics 2014).Based on findings observed in other contexts, the risk of homelessness may be particularly high for LGBTQ+ young people.A survey of homelessness-related experiences among more than 1,100 young people in Canada observed that 29.5% identified as LGBTQ+ (Gaetz et al. 2016).In a US study of more than 26,000 young people, the relative risk of experiencing homelessness was significantly greater for those who identified as lesbian, gay, bisexual, or transgender (risk ratio 2.20) (Morton et al. 2018).In Australia, an assessment of N = 2,159 intake records from a youth housing program found that 23% of the sample identified as LGBTQA+ (Hail-Jare and Vichta-Ohlsen 2023)a considerably higher proportion than would be expected based on population estimates (Carman et al. 2020).
Despite this, both policy and service responses to homelessness within Australia have been slow to recognize LGBTQ+ individuals as a vulnerable subgroup (Côté and Blais 2019;McNair et al. 2022), available research on LGBTQ+ homelessness in Australia is presently largely qualitative (Côté and Blais 2019) and quantitative research on this topic is nascent (McNair et al. 2022).These gaps in current understandings of the epidemiological patterns of homelessness limit both service provider and policymaker action.Accordingly, the current study addresses this gap by examining the demographic and biographical factors most closely associated with experiencing homelessness for LGBTQ+ youth and offers insight on how experiencing homelessness impacts the mental health and victimization likelihood of this group.

Understanding experiences of homelessness among LGBTQ youth
A constellation of structural, institutional, and biographical factors shapes individual trajectories towards experiencing homelessness; however, these factors reliably coincide with extant indicators of social and/or socio-economic disadvantage (Ecker, Aubry, and Sylvestre 2020).Unsurprisingly, research across a variety of societal contexts has consistently identified LGBTQ+ individuals as being overrepresented among homeless populations (McCarthy and Parr 2022).Moreover, LGBTQ+ persons' experiences of homelessness appear qualitatively distinct to those of their cisgender and/or heterosexual counterparts.For instance, various so-called 'pathways' to homelessness have been identified (Côté and Blais 2019) and draw attention to the structural and intrapersonal factors that disproportionately effect LGBTQ+ individuals.These include employment discrimination (Fric 2019;Perales 2022), violence (Cusack, Montgomery, and Byrne 2022;Flatley et al. 2022), negative health impacts of minority stress (Romero, Goldberg, and Vasquez 2020), family instability (Giano 2020;Robinson 2021), challenges with substance use (Prock 2019) and family rejection (Rosenkrantz et al. 2020)s.Additionally, identity-related stigma and discrimination are also thought to directly inform unique pathways to homelessness for LGBTQ+ persons while constituting an additional dimension of victimization experiences whilst they are homeless (Robinson 2021).
Because LGBTQ+ persons do not constitute a monolithic group, certain subgroups of individuals are thought to be at a greater risk of homelessness, and experience greater vulnerability while experiencing homelessness than others.One such group are LGBTQ + young persons, who are assumed to experience significant vulnerability when 'coming out' or disclosing their minority sexual and/or gender identities to parents or caregivers (Robinson 2020).Queer scholarship has classically understood this through the lens of the families-of-choice thesis; this purports that these instances of disclosure may foreground the revelation of one's familial ties as conditional and temporary, as opposed to stable sources of support (Robinson 2018).
Given the relatively early initiation of sexualand/or gender-identity formation processes within contemporary youth cohorts (Bishop et al. 2020;Oakley and Bletsas 2018), it is perhaps unsurprising that familial rejection is cited as a causal factor for homelessness among youth persons across many societal contexts .LGBTQ+ youth and adolescents may be ejected from their homes by families of origin or escape from hostile environments where they experience abuse due to their sexual and/or gender non-conformity (Matthews, Poyner, and Kjellgren 2019).This may be particularly the case for trans and gender-diverse individuals, whom research show be at greater risk of family rejection and ejection, and therefore report greater prevalence of homelessness experiences (Strauss et al. 2020).Regardless of sexual and/or gender identity, these early-life experiences of homelessness are especially deleterious to future life chances, insofar as they significantly obstruct educational engagement and attainment (Forge et al. 2018;Tyler and Schmitz 2018) and therefore, confer future vulnerability to homelessness and housing insecurity (Heerde, Pallotta-Chiarolli, and Parolini 2020).This article will examine the prevalence and predictors of LGBTQ+ homelessness using data from a large online survey of LGBTQA+ young people in Australia, Writing Themselves in 4.An analysis of the predictors of homelessness will allow for a better understanding of how different LGBTQ+ communities experience homelessness and key risk factors for this experience.Findings will assist those involved in homelessness service delivery and policy development to better identify LGBTQA+ young people at greatest risk of homelessness and assist health services to better support this population through targeted health promotion and resource allocation.

Other demographic predictors of homelessness
LGBTQ+ youth and adolescents are at considerable risk of experiencing homelessness relative to their non-LGBTQ+ counterparts.However, scholars have increasingly cautioned against over-emphasizing sexual and/or gender minority identity as the causal driver of experiences of homelessness among these groups (Wheeler, Price, and Ellasante 2017).Researchers instead advocate for a holistic, socio-ecological view of LGBTQ+ homelessness that accounts for concomitant structural drivers such as discrimination and oppression that pertain variously to race, class, ability and age (Castellanos 2016;Shelton and Bond 2017), as well as comparatively distal factors like parents' and caregivers' own vulnerabilities, family violence and abuse-related dynamics, prior criminal justice involvement and limited housing market opportunities (Grooms 2020).Moreover, more recent scholarship has drawn attention to how homelessness pathways and experiences for LGBTQ+ persons are highly variegated, with minoritized experiences and identities other than sexual and/or gender minority informing one's vulnerability to experiencing homelessness.
A review of the existing literature suggests the following characteristics to be individually salient to LGBTQ+ persons' pathways to, and experiences of homelessness.While a small but growing body of evidence convincingly points to the compounding effects of multiple marginalized identities and/or experiences within this context (McCann and Brown 2021), much of this research pertains to the intersection of LGBTQ+ and minoritized racial identities.In comparison, how other social variations intersect with these identities to increase risk of homelessness is presently unclear.

Family religiosity
Religious parents may struggle to accept their LGBTQ+ children due to perceived incongruities between religious teachings and sexual and gender diversity (Jones et al. 2022).Previous findings suggest that while presumed incompatibility between religious faith and sexual and/or gender diversity are often overstated (Acosta 2020;Adelson, Walker-Cornetta, and Kalish 2019;Avishai 2020), parents' religious beliefs nevertheless foreground their rejection of non-heterosexual/non-cisgender children in ways that coalesce into experiences of homeless for the latter (Rosenkrantz et al. 2020;Ryan et al. 2020).Both the gray literature and qualitative evidence affirms the role that family religiosity plays in LGBTQ+ youth and adolescents' pathways into homelessness (Asquith et al. 2019(Asquith et al. , 2021;;DeChants et al. 2022).At the time of writing, however there is a distinct lack of quantitative research exploring associations between family religiosity and experiences of homelessness among LGBTQ+ populations (Morton et al. 2018).

Race & ethnicity
Raciallyand/or ethnically-minoritized persons experience unique factors which shape their likelihood of experiencing homelessness (Kim and Fredriksen-Goldsen 2017).Structural racial inequality compromises the life chances of multicultural LGBTQ+ individuals, such that raciallyand/or ethnically-minoritized persons frequently experience poorer socioeconomic circumstances (e.g.lower educational attainment, income levels, lifetime savings, etc.) than their racial majority counterparts (Kum 2017), predisposing them to housing insecurity and experiences of homelessness.Young LGBTQ+ Persons of Colour (POC) may also be more likely to experience familial rejection (Logie and Rwigema 2014;Smith, Perrin, and Sutter 2020) and hence, homelessness when disclosing their sexual and/or gender identities.Extant qualitative findings suggest that this may be partially attributable to the implicit racialization of sexual and/or gender minority persons as white (Logie and Rwigema 2014;Worthen 2018).This adds an additional layer of estrangement which may increase the likelihood of young LGBTQ+ POC being alienated from and rejected by their family and communities of origin.

Area of residence
While homelessness is predominantly conceptualized as an issue unique to urban environments (MacDonald and Gaulin 2020), evidence from other anglophone societies such as U.S. and Canada demonstrate the understated and understudied prevalence of rural and non-urban homelessness among the general population (Easterday, Driscoll, and Ramaswamy 2019) Within these contexts, the phenomenon of rural and non-urban homelessness is overlayed by structural factors such as low public investment in housing and service infrastructures, as well as relative poverty and fewer employment opportunities.Rural and non-urban homelessness is likely also underreported due to the especial stigmatization and consequent invisibilization of homelessness within rural and non-urban settings (Buck-McFayden 2022).While rural and other non-urban homelessness remains an under-investigated subject within the Australian context (Petersen 2020;Zufferey and Parkes 2019), available evidence indicates that persons experiencing homelessness within these settings face additional obstacles to accessing social supports and services (Petersen 2020;Zufferey and Parkes 2019).

Disability
The relationship between disability and housing precarity is well-established within the literature (Baker Collins and Fudge Schormans 2021;Beer et al. 2019;Durbin et al. 2018), and having a disability is both a risk factor and perpetuator of experiences of homelessness.Extant research suggests that disabled persons experience substantial socio-economic disadvantage (Cortese et al. 2021)having to contend with interpersonal (Temple et al. 2020), employment (Keramat et al. 2021) and housing discrimination (Jarwala and Singh 2019).These factors constrain disabled persons' access to housing, and therefore facilitate housing precarity.Disabled persons experience unique homelessness trajectories (Baker Collins and Fudge Schormans 2021) and may have more enduring needs and require longer-term support than non-disabled homeless persons.Disabled persons' experiences of pathways to homelessness are further shaped by the kinds of disability experienced.While most existing research focuses on 'visible' (e.g.physical) disabilities, youth with 'invisible' (e.g.intellectual, learning, and developmental) disabilities are likely overrepresented and underserved among homeless populations (Beer et al. 2019).Persons with such disabilities experience additional difficulty accessing appropriate supports (McKenzie et al. 2019;Stone, Dowling, and Cameron 2019), and those who have impaired cognitive or adaptive functioning may respond inconsistently to support (Beer et al. 2019).Despite this, there is a paucity of research examining intersections between LGBTQ identity and disability within the context of homelessness, such that it is uncertain how these marginalized identities and experiences intersect to inform homelessness risk.

The current study
Using a series of logistic regression models (binomial and poisson models), we explored the association between homelessness and demographic characteristics such as: (i) age; (ii) sexual identity; (iii) gender identity; (iv) race and ethnicity; (v) family religiosity; (vi) residential location and (vii) disability against lifetime, recent (<12 months).We further investigated associations between experiences of homelessness, experiences of physical and sexual harassment, and psychological distress and problematic alcohol use.

Sample and procedure
The current study reports on a subset of findings from Writing Themselves In 4 (Hill et al. 2021) (WTI4), a cross-sectional, anonymous online survey investigating self-reported health and wellbeing among LGBTIQA+ Australians aged 14-21 years old.Data collection occurred in 2019 using an online instrument.Participation eligibility was contingent on respondents: (i) identifying as LGBTIQA+, (ii) residing in Australia, and (iii) being 14-21 years of age.This survey was promoted through several avenues; chiefly, using targeted, paid advertisements on both Facebook and Instagram and by leveraging online and in-person networks of Australian LGBTQA+ community organizations.To improve the accessibility of the survey for participants experiencing precarious living situations, the survey was also optimized for smartphone compatibility.Previous research suggests has highlighted how individuals experiencing homelessness may be more reliant upon smartphones and other such personalized devices than homed individuals (Humphry 2022), such that the delivery of interventions to homeless populations via these avenues is perceived as being of acceptable viability (Thurman et al. 2021).Moreover, participants were allowed to complete the survey over multiple sessions to account for those who may experience intermittent or unstable internet access (i.e.such as those experiencing homelessness).The present study focuses on a subsection of participants (n = 6,114, 95.2% of the original sample) who reported either lifetime, recent (<12 months), ongoing or no-previous experiences of homelessness.Ethics approval for WTI4 was granted by the La Trobe University Human Research Ethics Committee.

Gender and sexuality
Participants were asked to select the gender that best described their gender identity from a list of 20 options, including the option not to answer.Examples included labels such as Male, Female, Trans Man, Trans Women, Genderqueer and Non-binary.Participants were categorized as cisgender if their assumed gender at birth aligned with their present gender identity.Participants whose assumed gender at birth was different to their indicated identity were categorized as either transgender men, transgender women, or gender-diverse, depending on the selected gender identifier.Sexual identity was captured through a list of options, including 'gay', 'lesbian', 'bisexual', 'pansexual', 'queer', 'asexual', 'homosexual', 'heterosexual', 'prefer not to answer', 'prefer not to have a label', 'don't know', and 'something different'.Due to small of participants identifying as 'homosexual', 'prefer not to have a label', and 'something different' these were collapsed into a 'something different' category.

Experiences of homelessness
In accordance with the expansive definition laid out in the introduction, the current study defined homelessness to our participants as 'not having a stable or safe place to live and can include things like sleeping outside, and living or sleeping in a car, shelter, hostel, or refuge'.Participants were asked if they: (i) had any historic (not including current or recent) experiences of homelessness, (ii) had experienced homelessness within the past 12 months; and (iii) if they were currently experiencing homelessness.They were also asked if they had: (i) run away, (ii) left home because of being asked to leave, (iii) couch surfed or (iv) been homeless.Responses to both items were recorded on a yes/ no binary.

Family or household religiosity
Family religiosity was assessed using a dichotomous yes/no response to the item: 'Is your family or household religious?' Disability Disability status was assessed through an item that asked participants if they had a disability.Disability was defined to participants to exclude either mental health conditions or neurodivergence, Participants were categorized as either 'disabled' and 'non-disabled' depending on their responses.

Kessler psychological distress scale (K10)
The K10 is a self-rated instrument measuring psychological distress which has demonstrated high internal consistencies, high unidimensional factor loadings, and consistent performance across culturally distinct socio-demographic groups (Furukawa et al. 2003).The instrument comprises 10 items capturing psychological distress in the previous 4 weeks, and responses are scored and summed to arrive at a total score ranging from 10-50 points.Australian normative data has been used to establish standard cut-offs which respectively indicate low (score of 10-15), moderate (score of 16-21), high (score of 22-30) and very high (score of 31-50) distress (Andrews and Slade 2001).

Victimization and harassment
Experiences of school-based harassment within the previous 12 months were considered in three types in the present study.These were: physical (e.g.being shoved, punched, or injured with a weapon), sexual (e.g.unwanted touching, sexual remarks, sexual messages, or being forced to perform any unwanted sexual act) and verbal (e.g.been called names or threatened).Responses were recorded on a yes/no binary for each modality.

Alcohol use disorder identification test (AUDIT-C)
The AUDIT-C (Bradley et al. 2007) is a 3-item measure that screens for potentially problematic alcohol use.Previous studies suggest that the AUDIT-C has both high construct validity (Campbell and Maisto 2018) and high internal reliability (Rumpf et al. 2013), with a Cronbach's Alpha of 0.80.Items in this measure are: (i) How often did you have a drink containing alcohol in the past year (ii) How many drinks containing alcohol did you have on a typical day when you were drinking in the past year?and (iii) How often did you have six or more drinks on one occasion in the past year?Previous validation studies to identify optimal AUDIT-C cut-off scores suggest that a total score of 4 (out of a maximum of 12) is thought to indicate potentially problematic levels of alcohol consumption (102).AUDIT-C scores were dichotomized into problematic (score of 4-12) and non-problematic (score of 0-4) drinking.

Statistical analysis
Statistical analyses were performed in STATA (Version 17 SE; StataCorp, College Station, TX).Firstly, we described all variables.We then estimated a series of forced-entry univariate logistic regressions to examine associations between independent/predictor and dependent/outcome variables.Independent variables for these analyses were: (i) demographic characteristics (k = 7), (ii) family or household religiosity and (iii) disability.Dependent variables were (i) lifetime, (ii) recent (<12 months), (iii) ongoing experiences of homelessness.Single-predictor logistic regressions estimated unadjusted associations between individual independent variables (sexual orientation, gender identity, age, ethnicity category, residential location, household religiosity and disability status) and the dependent variables (homelessness experiences) using forced entry.For recent experiences (<12 months) of homelessness, we also estimated unadjusted associations between the independent variables and recent (<12 months) experiences of physical, verbal, and sexual harassment.Associations between recent experiences of homelessness (<12 months) and psychological distress was investigated using a univariate linear regression, and the relationship between experiencing homelessness and problematic drinking was investigated using a univariate poisson regression.

Experiences of homelessness
Nearly a quarter (23.6%) of all participants reported historic experiences of homelessness, while 11.5% reported only experiencing homelessness in the last 12 months (11.5%) or only an ongoing experience of homelessness (1.9%).

Analytical findings
Previous (>12 months prior) experiences of homelessness.Table 2 displays the regression results for all analyses involving lifetime experiences of homelessness.Compared to cisgender women, we noted significantly higher adjusted odds for experiences of homelessness for trans women, trans men and non-binary participants.Significantly higher odds were also observed for disabled participants compared with non-disabled participants.Likewise, when compared to people who identified as lesbian, higher odds for historic experiences of homelessness were noted for participants who identified as either pansexual or queer.Multicultural participants compared to white participants and participants from religious families or households compared to those from non-religious families or households had higher adjusted odds for historic experiences of homelessness.Similarly, when compared to participants residing in a capital city, higher odds were noted for those living in either regional or rural/remote locations.Lastly, participants in the 18-21 age group had slightly higher odds of experiencing homelessness than their counterparts in the 14-17 age group.
Recent (previous 12 months or less) experiences of homelessness Table 3 displays the regression results for all analyses involving recent experiences of homelessness.From the multivariable regression results, significantly higher odds of reporting recent experiences of homelessness were noted for: trans women, trans men and non-binary participants compared to cisgender women; pansexual participants, compared to lesbians, disabled participants compared to non-disabled, participants from religious families or households, compared to participants from non-religious families/households, participants residing in regional or rural/remote locations compared to participants residing in capital cities and multicultural participants compared to Anglo-Celtic participants.Significantly lower adjusted odds ratios were noted for: gay men compared to lesbians and participants in the 18-21 age group compared to 14-17 age group.

Ongoing experiences of homelessness
Table 4 displays the regression results for all analyses involving current experiences of homelessness.From the univariate regression results, significantly higher odds of reporting recent (<12 months) experiences of homelessness were noted for: trans men compared to cisgender women, disabled participants compared to non-disabled participants and participants residing in rural/remote locations compared to participants residing in capital cities.Psychological distress, physical, sexual and verbal harassment and problematic alcohol consumption Table 5 displays the regression results for all analyses involving K10 and AUDIT-C scores.Significantly higher odds of experiencing physical, sexual, and verbal harassment were noted for participants who reported experiences of homeless in the last 12 months, as compared to those who did not.Higher odds of high or very high psychological distress were noted for participants who reported any experiences of homelessness, as was the likelihood of reporting problematic levels of alcohol consumption.

Discussion
Our results generally corroborated previous findings from other national contexts which demonstrate that homelessness risk and burden are unevenly distributed among LGBTQ+ subgroups and co-occur alongside other forms of marginality (Siconolfi et al. 2020).Firstly, we noted stark differences in homelessness prevalence rates in the sample, which was notable given the relatively young age of our sample.We found that while 18-21-yearold participants were more likely than their younger counterparts to report lifetime experiences of homelessness, they were less likely to report recent experiences of homelessness.One plausible explanation for these findings is that young LGBTQ+ persons may transiently experience homelessness during adolescence but exit it as and when they gain economic autonomy.Both disability and non-urban residence were significantly associated with heightened odds of experiencing lifetime, recent and current homelessness.
The present study utilized a broad definition of disability which was inclusive of intellectual, learning, and developmental disabilitieswhich are often underrepresented within the existing homelessness literature (Baker Collins and Fudge Schormans 2021).In Australia, disabled persons experience a matrix of structural, institutional, and interpersonal factors that predispose them to housing precarity by constraining the availability and accessibility of housing for non-able-bodied persons (Baker Collins and Fudge Schormans 2021).Recent research suggests that disability type and the disability severity underpin stark differences in relative homelessness risk.as well as the qualitative experiences of homelessness, thought more nuanced understandings of these differences is forthcoming.While these distinctions are not explored within our analyses, our findings nevertheless suggest that disability status compounds upon sexual and/or gender minority status to elevate homelessness risk for youth who hold both identities simultaneously.

Gender and sexuality
Our findings additionally provide some clarity to the differences in homeless vulnerability among the component identity subgroups within the LGBTQ+ umbrella.Where existing research has tended to consider the risk factors for homelessness among LGBTQ+ youth to be largely consistent across identity categories, our findings demonstrate significant variation in the likelihood of homelessness according to sexual identity.Participants identifying as either pansexual or queer appeared to experience the greatest vulnerability to homelessness.Given preestablished associations between experiences of stigma and homelessness risk, bisexual persons may be expected to experience the highest rates of homelessness.Instead, the current analyses raise the possibility that non-gay, lesbian, or bisexual sexual minority persons may experience unique forms of discrimination or rejection which coalesce into homelessness risk.This may be explained by the fact that many forms of bisexual-specific discrimination documented within the literature seem to originate from gay and lesbian individuals (Morandini et al. 2023), and that younger individuals likely have limited, or sporadic contact compared to their adult counterparts (Fish et al. 2019).Likewise, individuals who identify with non-traditional identity labels often report stigma and prejudice stemming from the perceived illegitimacy of their sexual identities (Morandini et al. 2023).While this similarly characterizes the kinds of prejudice bisexual individuals are subjected to, previous findings suggest a greater prevalence of such experiences among individuals who identify with non-traditional identity labels (Morandini et al. 2023).
Research investigating the interface between non-traditional identity label adoption and family rejection and/or acceptance is presently scarce.However, it is possible that the use of such identifiers may be perceived as radical departure from normative understandings of sexuality and sexual identification, and so constitute a source of potential conflict and alienation from one's family and community.Alternatively, past findings suggests that individuals who adopt non-traditional sexual identity labels may demonstrate heightened levels of stigma consciousness than either their bisexual or monosexual counterpart (Morandini et al. 2023), and therefore, poorer psychological well-being and greater psychological distress.Given that both factors are implicated in pathways to homelessness for youths via their contribution to behavioral issues and family alienation (McCarthy and Parr 2022), it is possible that participants who identify with non-traditional sexuality labels are more profoundly impacted by identity-related discrimination, which in turn exerts a detrimental impact on their family relationships in ways that ultimately potentiate homelessness risk.
Our findings are additionally in concurrence with a wealth of existing data that points to consistent experiences of vulnerability and marginality common to trans and gender diverse individuals.Stable patterns of relative vulnerability were furthermore evident across various trans and gender diverse identities, which were largely consistent across lifetime, recent and ongoing experiences of homelessness.That transmen were most likely to report both current and recent homelessness is consistent with some previous findings which place them at the highest risk of homelessness among trans and gender diverse groups (Fraser et al. 2019).While some previous research posits that transmasculine individuals experience elevated rates of victimization and psychological distress, as well as the lower availability of supports comparative to other gender minority groups (e.g.Becerra et al. 2021), much of the available research indicates that transwomen are more likely to experience homelessness (Arayasirikul et al. 2022;Messinger, Guadalupe-Diaz, and Kurdyla 2022;Yarbrough 2023).While our findings could be interpreted as suggesting that the relative profiles of homlessness vulnerability observed between trans men and women may only arise in adulthood or late adolescence, this is more likely attributable to low N among trans women within our sample.

Race and ethnicity
The results outlined in this paper also add to a mounting body of evidence that implicates structural racial inequality in restricting housing accessibility and availability, whilst corroborating the notion that multicultural LGBTQ+ groups in Australia experience a greater degree of housing precarity than their white counterparts.In Australia, racial and ethnic minority identity reliably denote experiences of racial discrimination (Elias, Mansouri, and Paradies 2021), even for individuals belonging to ethnic groups that are nominally considered 'white' (Marino 2019;Rajkhowa et al. 2019) but are nevertheless considered a racial outgroup by racially hegemonic Anglo-Australians.This observation holds true even for multicultural persons from younger age cohorts (Idriss 2022) for whom racial discrimination is often a nonnegotiable feature of their schooling experiences (Bodkin-Andrews et al., 2021;Nisar et al. 2021;Uptin 2021).That racial discrimination directly constrains the availability of rental housing for some groups of racial minority persons in Australia is well-established (Dunn et al. 2018;MacDonald et al. 2016); our findings additionally suggest that racial discrimination also contributes to housing precarity and experiences of homelessness through more indirect avenues.For young multicultural LGBTQ+ persons specifically, experiences of racial victimization may engender considerable socioemotional difficulty (Priest et al. 2019), which in turn erodes familial relationships and predisposes them to family ejection.
In addition, multicultural LGBTQ+ persons may experience significant difficulty accessing support from the LGBTQ+ community due to racial discrimination or fear of encountering discrimination (Pallotta-Chiarolli, Sweid, and Sudarto 2022; Parmenter, Galliher, and Maughan 2021), and so may only conditionally experience the benefit of connectedness to the LGBTQ+ community (Roberts and Christens 2021).As this pertains to experiences of homelessness, multicultural LGBTQ+ persons may also be underserved by populationspecific services, where support that is sensitive to their cultural needs may be absent (Olivet, Dones, and Richard 2019;Olivet et al. 2021).However, as much of the research with racially-minoritized LGBTQ+ group has been conducted within the U.S. and Canada, it is uncertain which of these findings are applicable to the Australian context.
However, insofar as past research suggests some racial minority LGBTQ+ persons may be more adept at coping with minority stressors than their racial majority LGBTQ+ peers due to their frequent exposure to identity-based discrimination (Moradi et al. 2010), the increased likelihood of experiencing homelessness among our multicultural participants may not be singularly attributable to racial victimization.Rather, this heightened vulnerability likely also reflects multicultural youths' disproportionate reliance on family networks for support comparative to racial majority individuals (Frost, Meyer, and Schwartz 2016;Roberts and Christens 2021).While there is a distinct lack of scholarship examining multicultural LGBTQ+ youths support structures within the Australian context, the prevalence of interpersonal prejudice and discrimination likely mean that both family and community of origin frequently come to comprise the most viable sources of support for multicultural LGBTQ+ youth.That racial minority communities in Australia may be characterized by high levels of social conservatism and heterosexism (Asquith et al. 2019;Perales and Todd 2018;Sullivan and Jackson 2020) suggests that homelessness resulting from family rejectionand subsequently, ejectionis particularly salient to multicultural LGBTQ+ youths.

Family religiosity
The role of family or household religiosity, though not participants' individual religious or spiritual affiliation, was explored vis-a-vis experiences of homelessness and was observed to be modestly associated with both lifetime and recent experiences of homelessness.This association is likely representative of the fact that social conservatism (and hence, hostile attitudes towards sexual and gender minority persons) tends to coincide with religious affiliation and/or religious belief (Baron 2020;Carden 2019).Much of the qualitative evidence suggests that family religiosity potentiates experiences of homelessness via familial/parental rejection (Schmitz and Woodell 2018).However, existing findings also raise the possibility that religious minority groups within Australia may experience minority stressors (Colic-Peisker, Mikola, and Dekker 2019;Pallotta-Chiarolli, Sweid, and Sudarto 2022) that are closely entwined with racism and racial discrimination (Seet and Paradies 2018), and which greatly constrains economic opportunities for individuals of certain religious groups (Hassan 2010;Lippens, Baert, and Vermeiren 2023).As such, associations between family religiosity and experiences of homelessness may not be entirely explained by religious objections to sexual or gender diversity but may denote conditions which potentiate homelessness because of unstable or unpredictable family environments.

Health and wellbeing outcomes associated with homelessness
For psychological distress, assumptions of normality and heteroscedasticity of residuals were not met for the linear model and bootstrapping with 1000 resamples was employed to generate bias-corrected and accelerated confidence interval.For LGBTQ+ youth and adolescents, experiencing homelessness is associated with significantly higher levels of alcohol consumption, but only modestly associated with increased psychological distress.This latter finding may variously reflect the possibility that the psychological distress typically engendered by experiences of homelessness is buffered by the sense of autonomy and self-mastery that some LGBTQ+ young people are able to attain (Stewart andTownley 2019, 2020).However, that these young people report significantly higher alcohol consumption levels may be indicative of the paradoxical role of peer relationships for homeless youth and may concurrently explain why experiences of homelessness were associated with only slightly elevated levels of psychological distress within our sample.
These relationships offer a sense of community, social support, and opportunities for collaboration between homeless youths (Oliver and Cheff 2014;Rice et al. 2005).However, the increased access to recreational substances sometimes facilitated by these networks may encourage disaffiliation with conventional social structures among homeless youth, which may complicate future attempts to exit homelessness (Stewart andTownley 2019, 2020).Alternatively, as our analyses did not capture drinking behaviors prior to experiencing homelessness, these elevated alcohol consumption levels may reflect attempt to cope with the negative subjective experience of homelessness and/ or may be a contributing factor to participants' pathways to homelessness (Carver et al. 2021).
Furthermore, experiencing homelessness appears to be associated with experiences of physical, verbal, sexual harassment, with the present analyses pointing to a general clustering of risk.We were unable to determine whether participants experienced harassment at higher rates whilst homeless, or if prior experiences of harassment shaped their experiences of homelessness (e.g.motivating them to flee from an abusive home environment).Insofar as pre-existing scholarship suggests that victimization contributes to experiences of homelessness, is positively associated with length of time spent homeless, and occurs because of experiencing homelessness (DiGuiseppi et al. 2022;Tong et al. 2021), it is highly probable that experiences of harassment factor into experiences of homelessness at multiple timepoints, and originate from a variety of sources (e.g.parents, peers, other homeless persons, etc.).This is particularly salient to LGBTQ+ youth, who are especially predisposed to experiencing physical and sexual victimization (Hatchel, Merrin, and Espelage 2019).That we noted associations between verbal harassment and experiencing is homelessness somewhat novel, as similar associations are absent from more recent scholarship (Flatley et al. 2022) investigating experiences of victimization while homeless.These findings may point to the role that verbal harassment plays in potentiating experiences of homelessness for LGBTQ+ youths, as suggest by qualitative findings (Ecker, Aubry, and Sylvestre 2022).Alternatively, qualitative findings derived from adult LGBT populations (Robinson 2018) raises the possibility that homeless LGBTQ+ youths may be less adept at strategic identity management and disclosure than LGBTQ+ adults, potentially exposing them to relatively high levels of verbal discrimination.

Limitations
Several limitations circumscribe the findings presented.Firstly, as participants for the current study were self-selected, and may not be representative of Australian LGBTQ+ youth populations.This was further evident from the sample's demographic composition, as trans women were represented at inordinately low rates.Determinations of the extent to which the present sample deviates from national LGBTQ+ populations is hampered by a lack of census data that captures sexual orientation and gender identity (Australian Bureau of Statistics 2022).Additionally, the current analyses relied exclusively on selfreport data; given the stigma attached to homeless persons within Western contexts (Markowitz and Syverson 2021;Weisz and Quinn 2018), it is probable that experiences of homelessness were underreported within the current sample.Further, the small number of non-Anglo-Celtic participants from many ethnic or culture groups within our sample necessitated their collective grouping into a single analytic category to achieve the requisite statistical power for the analyses performed above.This is not optimal, considering that racism and racialized experiences vary markedly between ethnic and racial groups in Australia (Elias, Mansouri, and Paradies 2021), and demonstrably further intersect with other demographic factors such as gender, national origin, linguistic ability, and religious affiliation (Sharif et al. 2022).Lastly, our study did not capture either the duration or frequency of participant experiences of homelessness, which previous findings suggest are crucial determinants of both health (Rew et al. 2019;Siconolfi et al. 2020) and educational outcomes (De Gregorio et al. 2022) for young people experiencing homelessness.This restricts us from being able to examine intergroup differences as they pertain to homelessness severity.

Conclusions
Experiences of homelessness appear to be particularly pervasive among youth who are: (i) trans and gender-diverse, (ii) disabled and (iii) who reside outside of urban centers.Experiences of homelessness furthermore appear to coincide with harassment experiences, problematic alcohol consumption and high psychological distress.Our findings offer valuable insight into to homelessness risk and housing precarity; more crucially, our analyses demonstrate the structural factors which potentiate homelessness are unevenly distributed among sexual and gender minority groups.Additionally, significant variance exists even within individual identity subgroups and categories, suggesting that other demographic and biographical factors are often similarly pertinent as sexual and/or gender minority identity in predicting experiences of homelessness.Homelessness presents a pressing but ultimately preventable threat to both LGBTQ+ youths' immediate health and well-being, but also to their future life chances.Future interventions to address homelessness among this group must take into consideration the unique needs of those LGBTQ+ identity sub-groups most vulnerable to experiencing homelessness.

Table 2 .
Factors associated with lifetime experiences of homelessness (N = 1501).

Table 4 .
Factors associated with ongoing experiences of homelessness (N = 121).

Table 5 .
(a)Relationship between experiences of homelessness and K10 scores (b) Relationship between experiences of homelessness and AUDIT-C scores.