Religious Conversion Practices and LGBTQA + Youth

Introduction Multiple jurisdictions are debating responses to United Nations calls for banning attempts at conversion of lesbian, gay, bisexual, transgender, queer and asexual (LGBTQA +) peoples’ identities to fit religious norms. This paper aimed to examine Australian LGBTQA + youths’ experiences and outcomes of religious conversion practices attempting to change or suppress their gender or sexuality. It explored how attending conversion practices related to demographic characteristics and outcomes. Methods A 2019 online health and social well-being survey promoted via diverse social media questioned 6412 LGBTQA + Australians aged 14–21 years on their experiences of sexuality or gender change or suppression practices. Descriptive and inferential analyses were performed to understand relationships between exposure to conversion practices and demographic, socio-behavioural, and health and well-being measures. Results Whilst most participants had never attended counselling, group work, programs or interventions aimed at changing their sexuality or gender identity, 4% had attended such conversion practices. Analyses showed associations between engaging with conversion practices and (1) specific demographics (being cisgender male, multi-gender-attracted, unemployed, affiliated to a religion at the personal or household level); (2) social experiences (increased exposure to social rejection, negative remarks and harassment); (3) socio-behavioural outcomes (decreased education, sport and housing opportunities) and (4) negative health and mental health outcomes (including increased suicidality and self-harm). Conclusions The paper showed that conversion practices are correlated with poor well-being outcomes, providing arguments for expanding inclusive health and mental health services allowing for affirming religious and non-religious identities for LGBTQA + youth. Policy Implications The paper provides evidence supporting bans on conversion practices.


Introduction
Countries are responding to the United Nations' (2020) calls for banning attempts at 'converting' lesbian, gay, bisexual, transgender, queer, asexual and other diverse (LGBTQA +) people to fit religious cisgender heteronormative ideals.
Brazil, Ecuador, Malta and Taiwan banned conversion therapy whilst Canada, Spain and the United States (US) debate localised restrictions (United Nations, 2020;9). Bans are operating in Australia's Capital Territory, Queensland and Victoria ( ACT Government, 2020;QLD Government, 2020;Victorian Government 2021); emerging in Tasmania, South Australia and Western Australia (Hill, 2019;Richards & Skujins, 2020;WA Parliamentary Council, 2020); and debated in New South Wales and the Northern Territory. Grounded in critical emancipatory perspectives with a peerled group of LGBTQA + self-titled 'survivors' of diverse conversion ideology and practices, the Sexual Orientation and Gender Identity Change Efforts Survivor Statement outlines survivor-led definitions of 'conversion ideology' as overt spoken beliefs/teachings, and the underlying culture of a particular community of people that sees LGBTQA + identity as somehow 'broken' or 'sinful', and in need of 'fixing' or suppression to achieve heterosexual reproductive marriage and gender normativity within a largely binary sex/ gender model that aligns with the sex marker assigned at birth (Csabs et al., 2020). Conversion ideology thus goes beyond religion itself, or religious anti-LGBTQA + bias. Within conversion ideology, core religious assertions (like religious rejection threats) are distinct from pseudoscientific false and misleading claims that use psychological language to explain the origins, causes or 'dysfunction' of LGBTQA + identities because they are within democratic governments' intervention scope through laws and ethics codes (United Nations, 2020;Australian Psychological Society, 2007).
Hence 'conversion practices' comprises of processes engaged in towards desired changes in gender and/or sexuality based in conversion ideology. Conversion practices may include (but are not limited to) counselling, group work, programs or interventions (perhaps therapeutic/corrective prayer, performing celibacy or endorsed sexual relationships, personal or group behavioural suppression, etc.). These practices can be formal or informal, and regionally variable-such as witchdoctor exorcisms in Mozambique (United Nations, 2020). Health, well-being and education professionals are hindered in responding to legislative changes due to limited conversion research, especially on LGBTQA + youth (Australian Institute of Family Counselling, 2017;Health Complaints Commissioner, 2019). This paper firstly outlines existing research on conversion ideology and practices, and secondly reports on data drawn from an online LGBTQA + youth survey, specifically designed to explore questions pertaining to engagement with conversion practices.
Insufficient research maps conversion exposure prevalence for LGBTQA + youth, with between 7 and 16 + % of LGBTQ + adults reporting some exposure depending on country-the higher figures were reported in Iran and South Korea (Bishop, 2019;Jones, 2015;UK Government Equalities Office, 2018). UK-based research on 108,100 LGBTQ + people found that 7% were offered conversion practices, 2% undertook them (UK Government Equalities Office, 2018)-including 4% of trans men. US research on LGBTQA + adults estimated that over half of adults exposed to conversion ideology were also exposed as adolescents (Mallory et al., 2019). A study of 3134 Australian LGBTIQA + youth found that 7% were exposed to the conversion ideology messaging 'gay people should become straight' in schools (Jones, 2015;Jones & Hillier, 2012)-more in Catholic (15.4%) and Other Christian (16.4%) schools than government schools (3.6%). Students at schools with anti-homophobia policies had reduced exposure (3.9% vs. 14.3% without policies). Another study showed that 4.9% of 2500 mostly heterosexual cisgender Australian students were exposed to this conversion ideology school-based messaging: over a tenth in ideologically conservative schools (those schools inculcating students with the status quo's traditionalist values within a top-down power structure, Jones, 2020). These students reported greater harms to concentration, grades and attendance due to abuse at school than those not exposed to conversion ideology. They were significantly more likely to consider self-harm (81.8%); engage in self-harming behaviours (61.8%); consider suicide (83.6%); and attempt suicide (29.1%). Conversion ideology is clearly in operation across multiple contexts internationally including schools, and harmful. Quantitative studies of the prevalence and impacts for LGBTQA + youth attending conversion practices however, are lacking.
A survey-based study was developed to explore these research gaps to explore prevalence and outcomes of Australian LGBTQA + youths' exposure to conversion practice. The survey was conducted in advance of conversion bans so these data can be explored in comparison to international data and over time. Using Mallory et al. (2019) estimates, the team hypothesised that Australian youth may report conversion practices to at least half the 7 + % prevalence rates of LGBTQ + adults' exposure to conversion ideology messaging across Western contexts, and may experience harmful effects. The study addressed four research questions: 1. Do Australian LGBTQA + youths' conversion practice attendances differ by demographic factors? 2. What is the relationship between conversion practice attendance and social experiences? 3. What is the relationship between conversion practice attendance and socio-behavioural outcomes? 4. What is the relationship between conversion practice attendance and individual health and well-being?

Methods
The findings discussed in this paper are based on a section of the Writing Themselves in 4 survey-the latest in a series of cross-sectional health and social well-being surveys of LGBTQA + young people in Australia (aged 14-21 years) (Hillier, et al., 2010). It was approved by the La Trobe University Human Research Ethics Committee. The survey instrument comprised items pertaining to basic demographic characteristics including age, area of residence, gender, sexuality, education and employment. Questions were included that sought to understand whether survey participants had experience of attending counselling, group work, programs or other interventions aimed at changing their sexuality or gender identity (formal conversion practices). Response options included 'yes', 'no', 'don't know' and 'prefer not to say'. Those indicating 'yes' were subsequently asked to indicate whether they had this experience within the last 12 months or more than 12 months ago. These questions were developed with expert advice from a local organisation that works to support survivors of conversion practices (We The Brave) in collaboration with community advisory boards. Self-rated health was measured using the validated 5-point self-rated health (SRH) scale (Bombak, 2013). Psychological distress was measured using the Kessler Psychological Distress Scale (K10), a ten-item standardised scale developed to measure psychosocial distress in the past 4 weeks validated among young people in Australia (Australian Bureau of Statistics, 2012). Suicidal ideation was defined as 'experiences of thoughts about suicide, wanting to die, or about ending your life', suicide plans as having 'made a plan to attempt suicide or end your own life', suicide attempts as having 'attempted suicide or to end your life', self-harm ideation as 'thoughts about harming yourself on purpose' and self-harm as 'injured or harmed yourself on purpose'. Homelessness was measured as any lifetime experiences of having (1) run away, (2) left home because of being asked to leave, (3) couch surfed or (4) been homeless, based on a previously established measure of homelessness among young people in the US (36). Lifetime experiences of harassment or assault based on participant sexuality or gender identity were measured as follows: verbal (e.g. been called names or threatened); physical (e.g. being shoved, punched or injured with a weapon); and sexual (e.g. unwanted touching, sexual remarks, sexual messages or being forced to perform any unwanted sexual act). Participants were asked if they had disclosed their sexual identity or gender identity to friends, family or classmates. Response options included 'never', 'a few of them', 'some of them', 'most of them' and 'all of them'. Responses were dichotomised to 'any' or 'none' for analyses. Participants who responded they had disclosed their sexual identity or gender identity were asked how supported they feel. Responses were dichotomised from a five-point scale ranging from 'very unsupportive' to 'very supportive' into 'not supported' and 'supported/very supported' for analyses.
Data were collected from 2 September to 28 October 2019, following past social media recruitment strategies (Hillier et al., 2010), including Facebook/Instagram advertising. There were 6418 valid responses from LGBTQA + Australians aged 14-21 years. Data were downloaded into excel and SPSS. Descriptive and inferential statistical analyses examined relationships between conversion practices and: (a) demographics; (b) social experiences; (c) socio-behavioural outcomes and (d) health and well-being.

Results
Analyses involving categorical independent variables were χ 2 tests of independence (bivariate) or log-linear analyses (multivariate). Prior to these analyses being conducted, assumptions of these statistical tests were assessed, and all were upheld for each analysis (i.e. assumed independence of observations, all groups were mutually exclusive, expected count per call > 5, etc.). Analyses using continuous independent variables were between subject t-tests. Again, for each analysis, all relevant assumptions were assessed and upheld (i.e. all data were normally distributed, homogeneity of variance, etc.).
Of 6412 LGBTQA + youth who answered the question 'Have you ever attended counselling, group work, programs or interventions aimed at changing your sexuality or gender identity?', 256 (4.2%) indicated that they had attended conversion practices; of the remaining 5949 who indicated that they had not attended such a practice, 159 indicated that they did not know and 48 indicated that they preferred not to say. For parsimony, the latter two groups were excluded from analyses; thus, all analyses presented hereafter are a comparison of those participants who reported they had attended conversion practices with those who reported that they had not attended such practices. Of the 265 individuals who had attended conversion practices, 144 (56.3%) reported that this occurred in the last 12 months. For prudence, and because we did not ask details about how recent the practices were outside of 12 months, the data supplied from these individuals form the corpus presented in this paper.

Conversion Practices Attendance and Demographic Factors
A series of analyses were conducted to uncover associations between previous conversion practice attendance and LGBTQA + youths' demographic factors (age, gender diversity, gender, sexual attraction, personal and household religious affiliation, country of birth and employment status). Age was the only a demographic factor measured on a continuous scale. As such, a between-subject t-test was used to test if there were differences in the average age of individuals who had versus those who had not attended conversion practices. The analysis revealed that there were no differences in age between those who attended (M = 17.16, SD = 2.11) and those who did not (M = 17.32 , SD = 2.06), t(6203) = 1.185, p = 0.236, Cohen's d = 0.076.
The remaining demographic variables were categorical, and so a series of χ 2 tests of independence were performed to examine the relation between each variable and conversion practice attendance ( Table 1). The analyses revealed there was no association for gender diversity or country of birth-thus, cisgender and gender diverse participants were equally likely to have attended conversion practices, as were individuals born in Australia compared to overseas. Significant associations were revealed for the remaining variables: gender, sexual identity, personal and household religious affiliation and employment status.
Specifically, there was a significant association between conversion practice attendance and being cisgender male, multi-gender-attracted, religion-affiliated (at either the personal or household level) and unemployed (in comparison to being cisgender female, same-gender-only-attracted, without religious affiliation or employed, respectively). A log-linear analysis was conducted to predict conversion practice attendance from each of these variables. For parsimony, only the five variables with a significant bivariate association with attendance were included in the model, and only their main effects were tested (i.e. excluding interactions between variables). A test of the model with the five predictor variables against a model with only the constant was statistically significant, suggesting that the predictors (as a set) can reliably distinguish between individuals who have vs. have not attended conversion practices χ 2 (5) = 53.36, p < 0.001, and that this model is a good fit (Hosmer-Lemeshow statistic = 3.29, p = 0.915). The prediction success of the model was 96.2% (Table 2 shows coefficients). Cisgender males in the sample were almost two-and-a-half times more likely than cisgender females to have attended conversion practices, and individuals who had attended conversion practices were ~ 50% more likely to come from a religious household and to be unemployed.

Conversion Practices Attendance and Social Experiences
A series of between-subject t-tests were used to test if there were differences in social experiences for Australian LGBTQA + youth who had and had not attended conversion practices (Table 3). Those who had attended conversion practices were more likely than those who had not attended to be out to: family, friends, co-workers, classmates, teachers and sports teammates.
Importantly, of those participants who were out, those who had attended a conversion practice were more likely to have non-supportive responses to their disclosure from their family and classmates (although not from their friends, coworkers, teachers or sport teammates). Additionally, those who had attended conversion practices were more likely to have answered 'yes' to questions on whether they had heard negative remarks about sexuality, gender expression, trans people and people with intersex variations in their place of education, and about trans people and people with intersex variations in their places of employment, than those who had not attended.
Chi-square tests of independence were performed to examine the relationships between experiences of harassment and conversion practice attendance (Table 4). There was an association for all types of harassment: those who attended conversion practices were four-and-a-half times more likely to have experienced verbal and physical harassment, and almost three times as likely to have experienced sexual harassment, than those who had not attended.

Conversion Practices Attendance and Socio-Behavioural Outcomes
Chi-square tests of independence were performed to examine the relation between LGBTQA + youths' socio-behavioural outcomes and conversion practice attendance (Table 5). There were associations between having conversion practice attendance and being more likely to have answered in the affirmative to specific questions measuring: increased drug use and smoking, feeling less safe at place of education, higher levels of truancy and unemployment, greater homelessness (and for that homelessness to be specific to their being LGBTQA +) and decreased sport participation (specifically to avoid sport in order to avoid discrimination). This association was not significant for alcohol consumption or being enrolled in education.

Conversion Practice Attendance and Health and Well-being
A series of between-subject t-tests were used to test if there were differences in self-reported current psychological health between LGBTQA + youth who had and had not attended conversion practices (Table 6). Those who had attended reported increased symptomology of anxiety and psychological distress and reported less health and wellbeing than those who had not attended. A series of χ 2 tests of independence were performed to examine the relationships between attendance at a conversion practice and both suicidality and selfharming behaviours ( Table 7). The analyses revealed an association between having attended a conversion practice and suicidality: specifically, individuals who had attended a conversion practice were over two-anda-half times more likely to think about or plan suicide, and almost four times more likely to attempt suicide, than individuals who had not attended a conversion practice. Chi-square tests of independence were conducted to see if previous conversion practice attendance was associated with any previous mental health diagnoses (Table 8). There was an association between having attended conversion practices and being more likely to have had a diagnosis for all 10 mental health conditions considered (listed in Table 8). The odds ratios (range 1. 50-4.89) show that attendance at conversion practices is related to individuals being at least one-and-a-half times as likely to have a mental health diagnosis (including almost three-and-a-half times as likely to have been diagnosed with post-traumatic stress disorder/PTSD, and almost five times as likely to have been diagnosed with schizophrenia). Most LGBTQA + youth conversion attendees and nonattendees now preferred only LGBTQA + inclusive/only counselling services (58.4% vs. 57.6%) above non-inclusive services, and in-person general counselling (71.5% vs. 68%) above text/web chat or telephone general counselling. However, telephone counselling preference increased (to 8.4% from 4.9% for conversion attendees, vs. to 4.9% from 2.1% for non-attendees) for sessions regarding self-harm and/or suicide.

Emphasising Specific Demographic Factors
The portion of Australian LGBTQA + youth who attended 'conversion counselling, group work, programs and interventions' was smaller than the 4.9-7% of Australian students reporting school-based exposure to the 'gay people should become straight' messaging in recent studies (Jones, 2015;. Thus, conversion ideology-even narrowly defined around gay identity-may be more widespread than conversion practices, or easier to capture via surveys. Combining the groups who answered 'Yes', 'Don't know' and 'Prefer not to say' showed a total of 7.3% attended or potentially attended conversion practices, matching Australian and UK conversion ideology figures (Jones, 2015;UK Government Equalities Office, 2018), though falling below Iran and South Korea's 16% + figures (Bishop, 2019). Given over half of LGBTQA + youth attended conversion practices in the past year, Australian conversion practices were 'current'. It is possible that there is a greater number of LGBTQA + youth attending conversion practices than evident in these data, as younger participants perhaps required more nuanced definitions or safer/private spaces for responding, and limitations of the study included its lack of such definitions and spaces. These are recommended for future studies. The findings show a strong comparative representation of cisgender males in those who experienced conversion practices, consistent with the emphasis on white gay males in US adult research and gay males in African and Asian adult research (Bennett, 2003;Bishop, 2019;Erzen, 2006;Gerber, 2012;Waidzunas, 2015;Wolkomir, 2006), and contrasted against UK and Australian adult research-though this may change as new conversion practice movements targeting gender diversity spread (UK Government Equalities Office, 2018). Whilst research has found that trans and gender diverse Australians often became atheist adults, viewing religion as incommensurate with their identity (Jones, 2015), this is not the case for many LGBTQA + youth. Postconversion supports that are broadly religion-friendly are important given participants attending conversion practices often retained religious affiliations.

Negative Social, Socio-Behavioural, Health and Well-being Outcomes
Attending conversion practices was linked to higher identity disclosure to social networks. This may be because conversion practices can result from disclosures to family and community (APA Task Force, 2009). LGBTQA + youth who had attended conversion practices reported significantly less social support and more negative remarks on LGBTQA + topics (likely due to the increased attention on such topics in conversion efforts). In addition, they experienced considerably more verbal and physical harassment both generally and recently-perhaps within conversion practices or the contexts impelling their participation (perhaps both), although our data do not allow us to determine with certainty. Such rejection and harassment may partially explain the increased rates of feeling unsafe, truancy and disengagement in educational institutions and decreased employment and sports participation among young people who had attended conversion practices-however, this requires further research. Given Australian governments legally require school attendance, they hold responsibility to redress schoolbased harassment through exemption-free policy protections for LGBTQA + youth and inclusive curricula.
LGBTQA + youth who attended conversion practices also experienced elevated levels of homelessness compared to those who had not attended. This experience was often directly linked to their LGBTQA + identities including related mental health issues and family rejection. Such impacts were, for many, current and ongoing and potentially cyclical. Engaging in education, sport or employment is difficult without a home address; as is gaining a residence without qualifications, social contacts and/or income.
Research shows that LGBTQA + youth experience greater vulnerability to poor health and well-being (Australian Bureau of Statistics, 2012Statistics, , 2018Jones, et al., 2016;Lawrence, et al., 2015); however, LGBTQA + youth who attended conversion practices reported poorer health, and higher rates of smoking and drug use than those who did not. They were also at considerably higher risk of psychological distress and mental health conditions, self-harm and suicide than other LGBTQA + youth-echoing the higher risks seen in previous studies for Australian students exposed to conversion ideology. These risks may be under-stated, as many participants who attended conversion left well-being survey questions unanswered.
The significant social, cultural, physical and mental harms associated with conversion practices found in this study support the need for legislation banning these practices whilst being mindful of the impacts to young conversion practice attendees who may engage in promoting conversion to peers. However, banning conversion practices is not enough. These data suggest a need for more inclusive LGBTQA + youth services across Australian healthcare, education, housing and employment generally-which can contain hostile environments. For those who attended conversion practices, there is a greater need for in-person (as well as text/web chat and phone-based) crisis care and general counselling services that is both LGBTQA + inclusive and religion-friendly. It is imperative to support LGBTQA + and religious identities coexisting in more affirming frameworks (Jones et al., 2021), and socio-behavioural/institutional re-engagements in efforts to break potential cycles of exclusion.

Strengths and Limitations
Two key strengths of this study are that it provided a large sample, and the first Australian LGBTQA + youth sample, for a study actively exploring conversion practices (not just conversion ideologies). Over a quarter of the cohort were gender diverse-substantially more than the 3.1% of Australian LGBTQA + samples previously (Hillier et al., 2010), reflecting increased structural supports available to Australian gender diverse youth (Smith et al., 2014). Moreover, sexualities in the new cohort show increased trends of more frequent use of bisexual and non-binary sexual identities by same-gender-attracted females and gender minorities (Hillier et al., 2010). Contrastingly, there was greater sexual variability among Australian cisgender male youth, who previously mostly (82%) identified as gay. These diverse participants enabled the study to show the conversion push impacts a greater variety of individuals than shown in previous Australian data.
Limitations included that the survey was a cross-sectional LGBTQA + study, privileging breadth of coverage over detail for any one sub-group, and potentially excluding those claiming conversion practices 'worked' (identifying as cisgender and straight). Further, the study only reported on formal conversion practices; informal conversion practices (potentially delivered via activities such as everyday pastoral care or youth-led prayer groups and including for example casual compliments on gender-conforming outfits) can operate in subtle ways unobvious to participants and (therefore) not picked up in surveys. Thus, the conversion practice attendance figure should be considered a conservative minimum. Future research should define both conversion ideology and conversion practices in expanded questions supporting participants' understanding, provide 'safe spaces' supporting younger participants' ability to safely complete the survey without adult reprisal and explore emerging legislative impacts to understand conversion bans' effectiveness.

Conclusions
This study reflected Mallory et al. (2019) estimates that youth are exposed to conversion practices at around half the rate of their (and international LGBTQ + adults') exposure to conversion ideology messaging in Western contexts. It also supported LGBTQA + adult UK findings showing that conversion practices were experienced by people of diverse gender identities, cultural and religious backgrounds, and US findings emphasising targeting of religious cisgender males. It expands Canadian findings linking conversion practices to significant socio-behavioural and health/well-being harms and to mental health diagnoses. Given the prevalence, severity and range of impacts of conversion practices reported by participants in this study, the findings support calls for conversion practice bans and for inclusive educational, employment, healthcare and mental health services mediating negative impacts for affected youth.