Accepted on 16 May 2022            Submitted on 15 Nov 2021

Introduction

Autism is associated with a range of socioeconomic disadvantages across the life course. In the UK, school exclusion rates are three times higher for autistic pupils than in the general population, and autistic children and parents report a lack of understanding and support in educational settings (APPGA 2017). As such, many autistic children do not achieve their full academic potential (Ratto & Mesibov 2015). Upon leaving school, employment opportunities are lacking, even for those with qualifications (Parsons 2015). Consequentially, full time employment rates for autistic adults are low, with only 22% of autistic adults in some form of employment (Office for National Statistics 2021). A small number of follow-up studies have reported additional poor outcomes for autistic adolescents and adults, including co-occurring mental health issues and small social support networks (Eaves & Ho 2008; Howlin & Magiati 2017).

Considering these outcomes, it is important to examine the association between autism and social exclusion. Autism is a condition diagnosed solely through observed or described behaviours, which in clinical practice are benchmarked against constructed standards of normality (Lester, Karim & O’Reilly 2014). Benchmarked diagnostic criteria include persistent deficits in social communication and interaction as well as rigid and repetitive behaviour patterns (American Psychiatric Association 2013). In this sense, autism is associated with social difference, making it a condition of particular relevance for Critical Disability Studies (CDS). CDS explores ‘the complex social, cultural, material, and economic conditions that undergird the exclusion of disabled people’ (Goodley & Runswick-Cole 2016: 2). From this perspective, the marginalisation of disabled people results from the permeation of normativity in contemporary society. As autism is positioned as a deviation from the norm, autistic people may be at risk of exclusion through cultural productions of normalcy and difference (McLaughlin & Coleman-Fountain 2018; Titchkosky 2012).

It is acknowledged there is no single way of describing autism that is universally accepted or preferred (Kenny et al. 2016). Additionally, associating the diagnostic label of autism with social difference should not undermine the positioning of autism as a meaningful form of neurocognitive difference (Waltz 2013). In understanding autism as part of their identities, self-advocates often use identity first language. As such, this paper will refer to ‘autistic people’ as opposed to ‘people with autism’. The usefulness of terms associated with autism, such as ‘high-functioning’ and ‘low functioning’, has also been challenged. This paper uses these terms in quotation marks where necessary.

Homelessness is a form of social exclusion that has not been recorded in follow-up studies of autistic cohorts (Eaves & Ho 2008; Howlin & Magiati 2017). However, many of the aforementioned socioeconomic disadvantages these studies identify are also risk factors to homelessness (Shelter 2018). Additionally, there is a high degree of overlap between homelessness and mental health issues, substance mis/use and factors such as abuse or leaving care (Fitzpatrick et al. 2013; Homeless Link 2014). Exclusion which results from the interplay between different causes and consequences of homelessness, is known as ‘multiple exclusion homelessness’ (MEH) and is more likely to affect single, homeless people (Fitzpatrick, Bramley & Johnsen 2013; Manthorpe et al. 2015). Considering autism is associated with a range of co-occurring conditions and adverse life circumstances, those who experience homelessness may be at increased risk of MEH.

Despite these considerations, the link between autism and homelessness is relatively unexplored. A small number of UK-based studies have substantiated the claim that autism is overrepresented in homeless populations. In 2010, an evaluation of an initiative that encouraged rough sleepers into accommodation recorded that 7 of the 14 participants registered on the autism spectrum (Pritchard 2010). Participants self-reported a pre-existing diagnosis or were categorised as autistic by the lead researcher who had clinical expertise. Autism was also indirectly linked with homelessness in a large survey of autistic adults undertaken by the National Autistic Society. Homelessness was not the focus of this study, but responses revealed that 12% of autistic adults had experienced homelessness since leaving school (Evans 2011). More recently, an epidemiological study found that 12.3% of 106 homeless adults screened positive for autistic criteria as set out in the DSM-V, noting a further 8.5% could be considered as having marginal autistic traits (Churchard et al. 2019). This was the first study in the UK to ascertain prevalence levels of autism in homeless populations in the UK. Subsequent research has used the Autism Quotient-10 questionnaire to record autism prevalence levels of 18.5% in a sample of people accessing homelessness services (Kargas et al. 2019). The authors of both papers suggest future research should focus on the identification and understanding of risk factors for this population (Churchard et al. 2019; Kargas et al. 2019).

The scant research on autism and homelessness illustrates how little is known about why autistic people are at increased risk of homelessness. This paper attempts to bridge this gap, presenting a narrative enquiry into the link between autism and homelessness. This study aimed to ascertain why autistic people are at increased risk of homelessness and to facilitate autistic voice through answering the following research questions:

In what way, if any, do behaviours associated with autism lead to homelessness for autistic adults? What do autistic adults who have experienced homelessness consider to be the reasons for their homelessness? Methods Epistemological framework and justification of methods

Risk factors to homelessness are multifaceted and overlapping. However, this complexity should not lead to the conclusion that causes of homelessness are unfathomable (Bramley & Fitzpatrick 2018). Narrative was a suitable choice of methodology as it acknowledges the embodiment of a participant in different social contexts at different stages in their lives (Clandinin 2006). This broad scope allowed for the identification of complex contributors to homelessness, which may not have been captured in a traditional semi-structured interview. Additionally, narratives have the potential to actualise social change and can be beneficial to participants in terms of fostering introspection and self-reflection (Riessman 2008). This is important given that, traditionally, autism research has not included autistic voices or concerned the improvement of autistic people’s lives (Chown et al. 2017).

Narrative is an effective and well-used methodology within homelessness research (see Toolis & Hammack 2015, for example). However, research on autism and narrative production or story telling has tended to focus on linguistic deficits (Rollins 2014). This study was preceded by a pilot study that examined how to engage homeless, autistic people in narrative methods (Stone 2018). As in other populations, the narratives produced by participants were rich and compelling. The present study was therefore approached with a presumption of competence, though consideration was given to interview environment and researcher rapport.

Narratives can also be used to examine ‘macro-contexts’ where connections are made between ‘real life worlds depicted in personal narratives and larger social structures’ (Riessman 2008: 76). This study is framed by a Critical Realist perspective, which posits that there are discoverable underlying social mechanisms that drive social trends but our knowledge of these mechanisms is ‘conceptually mediated’ (Danermark 2002: 15; Fletcher 2017). For this study, Critical Realism was used as a framework to identify driving forces of homelessness for autistic adults. It was considered that autistic, homeless people were best placed to identify factors that increased risk of homelessness, but sometimes these factors may be outside of their understanding or experience. A revised application of thematic analysis was key to understanding how participant experiences related to wider context and theory. This process is explained in more detail below.

Recruitment and data collection

Participants were recruited from five different sites in two cities in the South West of England. Sites included three longer-term hostels, one drug and alcohol hostel, and an autism residence for homeless or vulnerably housed autistic adults. Approached services helped to access potential participants who met the following criteria:

Has a clinical diagnosis of autism or self-identifies as autistic1 Is currently homeless or has previously experienced homelessness

Information about the study was made available in standard and easy-read format. Ten participants consented to take part in at least two interviews. Conducting multiple interviews with each participant aided the development of trust and rapport, resulting in the collection of rich data, as well as allowing for shorter interviews in consideration of participant need. Interviews were held in rooms participants were familiar with at recruitment sites. Interviews lasted between 30 to 140 minutes each.

Narrative interviews can be conducted in different ways but for the purpose of this study involved the researcher working with each interviewee to select, organise and connect meaningful events in their lives (Riessman 2008). Whilst narrative interviews traditionally follow an open format, a topic guide was developed to provide structure where necessary. Development of the topic guide was informed by existing research on risk factors to homelessness and the socioeconomic disadvantages known to affect autistic people. However, to reduce confirmation bias and ensure collected data captured a wide range of contributors to homelessness, the topic guide was loosely based around significant life stages, which included childhood, education, employment, health, relationships and pathways through homelessness.

Given the potential discussion of sensitive issues during interviews, particular consideration was given to participant safety and ethics. Consent was re-checked, and limits to confidentiality were explained, prior to each interview. Relevant support services were signposted where appropriate. A risk assessment focusing on participant and researcher safety was developed prior to ethical approval being obtained from SPS Research Ethics Committee, University of Bristol (reference: SPSREC17-18.C41).

A form of thematic analysis (Braun & Clarke 2006), adapted according to Critical Realism (Fletcher 2017), was applied to the data set. Thematic analysis was in line with the epistemological framework of this study due to its focus on narrative content. Initially a process of data familiarisation involved the personal transcription and re-reading of transcripts. Narrative summaries were constructed that organised key events in participants’ lives chronologically. Themes, or tendencies, that emerged from the data were classed as subordinate or superordinate and organised in coding trees. Coding was continually revised and refined. As themes emerged, they were considered in the context of existing theory. From a Critical Realist perspective, consideration of wider theory can help identify causal mechanisms beyond participants’ immediate experiences. For example, in Fletcher’s (2017) study,2 examining the social conditions behind feeling a ‘loss of control’ helped identify ‘corporatisation’ as a driving causal mechanism. For the present study, redescribing themes using theoretical concepts allowed for consideration of how autism intersects with social exclusion to increase risk of homelessness. It also enabled the identification of ways participants responded to social exclusion as described in the Findings section.

Description of participants

Table 1 describes participant demographics by pseudonym. Eight of the 10 participants identified as male (one assigned female at birth). Two participants identified as female (one assigned male at birth).

Table 1

Participant demographics.

PSEUDONYM GENDER IDENTITY AGE AUTISMDIAGNOSIS AGE EDUCATIONAL ATTAINMENT CO-OCCURRING CONDITIONS Neil Male 29 25 Some secondary3 ADHD (dp) Stephen Male 26 26 Some secondary Anxiety, depression, learning difficulties Ellie Female 28 12 Some secondary ADHD, depression, bi-polar David Male 25 Self-identified College4 Depression, anxiety, chronic pain Alex Male 40 Self-identified (dp) Some secondary ADHD (dp), depression, anxiety Aaron Male 22 10 Some college (ongoing) ADHD, depression, anxiety, bi-polar, learning difficulties, epilepsy Bryan Male 33 Self-identified (dp) Some secondary Depression, anxiety Michael Male 28 10 Some college ADHD, depression, anxiety, schizophrenia Sarah Female 36 33 Higher Education Depression, anxiety, PTSD (dp) Charles Male 25 24 Higher Education Depression, anxiety

Participant age ranged from 22 to 40 at time of interview. The majority of participants were in their twenties. Seven of the 10 participants had received a clinical diagnosis of autism. These participants received diagnoses of autism spectrum disorder, Asperger’s Syndrome and ‘high-functioning autism’. All clinical diagnoses were received in adolescence or adulthood. Two participants were going through the diagnostic process at time of interview (dp = diagnosis pending).

Educational attainment was low for most participants. A minority had achieved GCSEs or equivalent.5 Two participants had obtained qualifications in youth offenders institutes or prison. The two participants who had obtained higher education qualifications came from wealthier socioeconomic backgrounds.

All participants reported one or more co-occurring conditions, with depression and anxiety being the most common. Three participants had an ADHD diagnosis, with a further two going through the diagnostic process at time of interview. No participants had a diagnosed learning disability. Learning difficulties have been recorded where participants reported needing learning support in school and/or learning disorders such as dyslexia.

Narrative Summaries

The following summaries, adapted from Stone (2021), describe participants’ life experiences and pathways through homelessness.

Neil

Neil, 29, grew up in a coastal town. He enjoyed spending time outdoors to escape his chaotic home environment and first became homeless at the age of 15. Since then, Neil experienced episodic homelessness, often choosing homelessness over adherence to societal expectations. Neil had struggled with self-harm and substance use in the past. Neil received an autism diagnosis a couple of years prior to interview. In response to pressure from his family, Neil had recently stopped rough sleeping and moved into a hostel.

Stephen

Stephen, 26, first became homeless at the age of 16 when he was kicked out of the family home because of escalating behavioural difficulties. Since leaving home, Stephen tried to avoid rough sleeping by staying with friends; he had also spent time in foster care and hostels. Stephen had been to prison twice and had a history of substance use. He also struggled with anxiety and panic attacks. Stephen had been diagnosed with autism a few weeks prior to interview. He found the chaotic hostel he was housed in stressful and adhered to strict routines as a coping mechanism.

Ellie

Ellie, 26, was a trans woman who had been in a hostel for two weeks following a 12-year period of entrenched rough sleeping, staying with friends, and living in squats. Coming out to her abusive mother at the age of 14, she was kicked out of her childhood home and was unable to continue her education. Ellie was diagnosed with autism, AHDH and bi-polar disorder in childhood. Despite her traumatic past, she found solace in her partner of five years, David, and was an autism self-advocate.

David

David, 25, became homeless after his father found out about his relationship with Ellie and kicked him out. David spent the next five years living in squats, staying with friends and rough sleeping. Struggling with chronic pain from an old injury, David had been unable to access healthcare whilst homeless and used marijuana to self-medicate. Whilst David did not have an autism diagnosis, he had taken multiple online tests returning results of high presence of autistic traits. David understood autism as part of his identity and discussed it in terms of benefits.

Alex

Alex, 40, was abused in his childhood and introduced to heroin by his father at the age of 11. Struggling with addiction, Alex experienced episodic homelessness throughout his life and had multiple jail terms for drug-related crimes and violence. During a period of stability, his child died, and unable to stay in the family home, he became intentionally homeless. Alex did not have an autism diagnosis but self-identified after the diagnosis of one of his children. Alex struggled with depression and found it hard to think positively about the future.

Aaron

Aaron, 22, was a trans man who had been homeless for around a year. Whilst Aaron grew up in a stable family environment, he felt his autism and epilepsy diagnoses separated him from his peers. In his late teens, Aaron moved across the country for a relationship, which subsequently broke down and severely impacted his mental health. After a succession of short housing rentals, Aaron fell out with his landlord and was made homeless. Aaron had not experienced rough sleeping as he was able to access a youth homelessness support service.

Bryan

Bryan, 33, reflected on a troubled childhood where he witnessed violence from a young age. Bryan often ran away from home, sleeping in unlocked cars or staying with friends. At the age of 10, Bryan petitioned the courts to be placed in a group home. Since leaving home, Bryan began selling drugs and eventually became addicted to heroin. Bryan became street homeless and was in and out of rehab and prison for a number of years. At the time of his interview, Bryan was living in a drug and alcohol hostel, using mindfulness and meditation to aid his recovery.

Michael

Michael, 28, was diagnosed with autism in childhood. He described struggling at school and had an impersonal relationship with his parents. Michael began experimenting with drugs in his teens in an attempt to fit in with his peers. After breaking up with a girlfriend, Michael became homeless, living in his van. Michael’s drug taking escalated, and he had a psychotic episode. Michael eventually found support through a homelessness shelter. At time of interview, he lived in a residence for autistic adults facing homelessness.

Sarah

Sarah, 36, had experienced short periods of homelessness during her life but had stable housing at time of interview. Sarah described a difficult relationship with her parents and would quickly form relationships with men to secure alternative housing. Sarah explained how she had low risk awareness, staying with men she had met at pubs and living out of a backpack. Sarah received an autism diagnosis at the age of 33. She had recently taken time off work for anxiety and stress and felt she was being forced out of employment.

Charles

Charles, 25, described a strict upbringing with an abusive mother where he often felt neglected. Charles did well at school and went on to higher education, achieving a Masters’ degree. Since leaving university, Charles struggled to find stable employment, his parents refused to house him and he spent time sofa surfing. During this time, Charles received an autism diagnosis, which had increased his self-understanding. He eventually found support and moved into a residence for autistic people facing homelessness.

Findings

This paper presents four themes that emerged from data analysis. The first two themes, complex lives and rejection of human difference, consider how complex life factors and social exclusion increased risk of homelessness for participants. The second two themes, concealing difference and renegotiating an autistic identity, examine contrasting ways in which participants responded to social exclusion.

Complex lives Adverse childhood experiences

Participants’ recollections of their childhoods highlighted a number of adverse experiences and hardships. Most participants described growing up in chaotic home environments, with six being subject to physical or emotional abuse from parents or family members. One participant was sexually abused by a family friend. Participants did not disclose the abuse at the time it occurred, and the majority acknowledged the impact these experiences had on their mental health. Ellie described the abuse she experienced from her mother and being unable to deal with this without support:

My mother used to lock me in a room for most of my childhood, with nothing but a mattress, that was my childhood, I had no tools to deal with any of it, I had no direction, what do you do when you are a scared child that has difficulty anyway? (Ellie)

Ellie had come out at the age of 14, which led to her mother kicking her out of the family home. Ellie and Aaron also identified as transgender and reported a lack of understanding from their families, and difficulties accessing medication and support once homeless.

Four participants described growing up in chaotic environments where they witnessed violence, crime and substance use from a young age. They reflected that this instability led to their own involvement in crime during adolescence.

Substance use and co-occurring conditions

All participants reported dependency on drugs or alcohol at some point in their lives. The most common dependencies reported were alcohol and marijuana. Other substances used included heroin, crack cocaine and amphetamines.

Interestingly, substance dependency developed either as a form of escapism from chaotic home environments and abuse or as a coping mechanism. In the following extract, Stephen explained how he began using substances as a way of coping with anxiety:

I struggle to sleep every night which is part of the other reason why I started misusing substances like alcohol and weed, and I have had a history of using crack as well… the best way I can describe it is I have got so much things in my head and it seems like they are going around at a hundred miles an hour… it’s so hard to slow my mind down. (Stephen)

Participants reported high rates of mental health issues and other developmental and physical conditions. Understanding co-occurring conditions was difficult for participants due to limited access to services and conflicting information from professionals. Ellie, who had received multiple diagnoses in her childhood, explained how she found her conditions overwhelming:

I have got Asperger’s, ADHD and Bipolar disorder, which make my head a complete fucking mess … I struggle to sleep, not because I am not tired but because my brain is *makes whirring noise* something is always going on. (Ellie)

An additional concern was the lack of clinical diagnosis of autism and co-occurring conditions. Participants who were not well supported in childhood tended to find it difficult to navigate diagnostic services in adulthood. Bryan, who had a long history of substance use, expressed frustration over the attempts of others to diagnose him:

Something has always been wrong, they don’t know what it is, people have pointed to other things… I think the drugs had fucked me up a little bit, there’s lots of mitigating circumstances and I don’t think you can pinpoint it to one thing… I am what I am and I don’t think putting a label on it will make a bit of difference. (Bryan)

Rejection of human difference

Participants experienced socioeconomic disadvantages throughout their lives. This impacted their ability to achieve qualifications, find and maintain employment and develop social support networks. These disadvantages both led to, and made it difficult to exit, homelessness.

School and education

Participants who were diagnosed with autism in childhood experienced ostracisation and bullying from peers as well as a lack of understanding and support from family and teachers. Michael explained how his diagnosis led to rejection both at home and school:

I guess that’s why I could never relate to other people because it isolated me and ostracised me within my own family because I was the black sheep and they had the paperwork to prove it…. It was so obvious that I was being taken out of a classroom, it was so obvious that I was the only person with a support worker… I didn’t like the idea that I was different, and all the support really highlighted that I was different. (Michael)

These participants struggled to understand autism as a part of their identity in early years, pushing away from the diagnosis due to its negative connotations. Participants without a diagnosis were also bullied for being different and explained how behaviours that may be associated with autism were interpreted as naughty behaviours. Left unsupported, participants reported high rates of suspensions and expulsions. Alex explained how he was expelled from multiple schools:

Just deemed as a naughty child, they kicked me out of that school, I had to go to another school, they kicked me out of that school, went to another school, kicked me out of that school, went to another school and then in the end they just fucking give [sic] up on me. (Alex)

Employment

After leaving school, 9 of the 10 participants struggled to find employment. Participants reported multiple barriers at the application stage. When an autism diagnosis was disclosed, most employers were unwilling to make reasonable adjustments. If adjustments were made, participants felt they were passed over due to their social differences:

They keep saying we are doing things like changing interviews, the problem is we don’t even get to that stage and you’re not listening… the changes are usually quite pathetic and they are not really done with any sympathy, all candidates are cardboard cut-outs, they fit squares into square holes. (Charles)

If participants were able to secure employment, social interaction was a concern. Some participants reported being let go because they were unable to interact with customers. The pressure to interact with multiple people could also lead to participants leaving employment. This was an issue as participants tended to get entry level jobs, where their roles were customer facing. Within these roles, participants were assumed to have customer service skills and were provided with little or no training. Bryan explained how he left two jobs due to the anxiety he felt interacting with customers:

I had to communicate with customers… that’s where I broke down, so I left there [restaurant], and I’ve worked in sales and marketing for about two weeks and I just felt really awkward, struggling with people. (Bryan)

Social networks and feeling different

Small support networks were also a concern for participants. The majority of participants reported relationship breakdown with family members, and whilst most participants had a few close friends, they found it difficult to form new relationships. Interestingly, participants tended to explain these difficulties in terms of deficits associated with autism:

Autistic people find it harder to engage with others a lot of the time, to network, to be friendly, to form relationships, platonic ones, but also romantic… I have never had a proper girlfriend … I think it would just be too hard for them to adjust. (Charles)

This association of autism with difference was common, with most participants describing their past behaviour as odd or abnormal. In doing so, participants measured their differences against a neurotypical standard. They understood their rejection across the life course as directly resulting from their perceived differences or inability to fit in with neurotypical counterparts. As Neil explained, society discriminates against those who are unable to meet accepted norms:

There is like a million brick walls there, everybody has to jump or go through, and yet they already know [some] of those people aren’t going to be able to do it, they have created a society which doesn’t match everybody. (Neil)

Concealing difference

Participants tended to respond to the exclusion they experienced by concealing social difference. They attempted this in two contrasting ways: masking autistic behaviours and withdrawing from social participation. Masking involved learning and utilising social techniques to try to hide autistic behaviours. The majority of participants felt that masking autism would lead to better social and economic opportunities, as Michael explained:

There is almost this indoctrinated idea that if you are to succeed in the world you will need [to be] less who you are and you will need to be more neurotypical. (Michael)

Sarah also likened the suppression of autistic behaviours with a better chance of success. She explained how she had been able to progress in her career by using masking techniques. Sarah accepted that she needed to change rather than expecting others to be accommodating of her differences:

I will get in line just because that’s what is expected of me and if I want to get ahead… it’s something I need to do… it’s fall into line with the neurotypicals just because it’s the only way you can actually get anything done. (Sarah)

Whilst masking was considered necessary, participants acknowledged the impact it had on mental health. Sarah was at risk of losing her job following a series of anxiety attacks and time off work. Michael also felt that masking autism had escalated his mental health issues and substance abuse. Interestingly, it was these factors that contributed to his lack of stable employment and housing. Ellie also considered the impact of masking on her autistic identity. Whilst Ellie did not want to disguise who she was, she felt outwardly displaying autistic behaviours was a risk to her safety, particularly when homeless:

I don’t think anybody should change who they are for what the world is around them… but I also understand that my idea of what the world should be and what the world is, is not the same thing, and I have been bullied enough of my life, I have been beaten up enough times, and I have been yelled at and abused enough, to know that sometimes not being yourself is the safest thing to do. (Ellie)

A second way participants responded to social exclusion was to withdraw from society, rejecting societal norms and expectations. For some participants, this began after they were repeatedly refused help by services and they stopped attempting to access support. Other participants began withdrawing from society after they were unable to find or maintain employment or form lasting relationships with others. Michael explained how he grew ‘closed off’ after being unable to secure employment which left him feeling like a ‘failure’. Withdrawing from society had implications for mental health as participants grew despondent and isolated themselves. Additionally, stopping service engagement could both increase risk of, or prolong, homelessness. However, this withdrawal was sometimes seen as preferable to social participation and interaction. Neil described feeling immense pressure to live up to social expectations as the following quote exemplifies:

Everybody is like you shouldn’t do that, you shouldn’t do this…. No leave me alone… I am alright doing what I am doing, but that’s never okay, society wants you to be a part of it and there is not really any option… it’s just not what I dreamed society should be. (Neil)

Renegotiating the autistic identity

Analysis revealed the complicated relationship participants had with the autism label. Upon receiving a diagnosis, participants initially distanced themselves from autism, associating it with perpetuated stereotypes and stigmatisations. However, after connecting with other autistic people and learning about autism online, participants began to look at their diagnosis more favourably. Whilst this was a positive experience, it took time as participants worked to renegotiate their identities:

I am learning not to beat myself up about that, a lot of the guilt and the shame associated with why can’t I function as a normal person… in inverted commas. (Sarah)

Additionally, whilst participants did not blame autism for the hardships they had endured, they began to reframe life events in the context of both their own limitations and societal constraints. A diagnosis in adulthood was often a cathartic experience, encouraging self-reflection. Stephen hoped his diagnosis would help him to understand the difference he had felt throughout his life. In this respect, the diagnosis was comforting:

It was nice to know that I am not just a weird person… like there is a reason for the way that my mind works, to understand certain things, and there is a reason for everything that I have been feeling differently. (Stephen)

Participants also discussed a range of skills and abilities that challenged deficit-based constructions of autism. These skills had not been honed in education, utilised by employers or recognised by support services. Michael explained how social change is needed in multiple aspects of life before autistic people are fully accepted:

If that kind of narrative can be pushed, this is just an important part of who you are… you stop that stigma, you stop that kind of demonising of it, because that was very much my experience, I was being told… how autism would restrict me in life. (Michael)

Participants used their renegotiation of the autistic label as a framework for challenging injustice and discrimination. Examples of this included pushing for adaptations in job interviews and asking for changes in the way welfare and homelessness services worked with autistic people. Some participants also expressed a desire to work in homelessness services. Ellie explained how her lived experience put her in a unique position to understand the challenges that homeless, autistic adults face:

I actually want to become a key worker here, which is odd for someone who has my difficulties, but it is because I get what it feels like, I get where the barriers are… you can’t equip someone with tools if you yourself don’t carry them. (Ellie)

Discussion

Reports of abuse in childhood, substance use and co-occurring conditions were notably high in this sample. The level of drug and alcohol use is of note as previous research has reported an inverse relationship between substance use and presence of autistic traits in homeless populations (Churchard et al. 2019). However, participants also reported adverse childhood experiences, such as abuse and introduction to substances at a young age. Additionally, there is evidence that autistic people may misuse drugs or alcohol as a coping strategy for social anxiety or to alleviate the pressure of masking (Adhia et al. 2020; Kunreuther & Palmer 2017). The mental health issues reported by participants are unsurprising given that these are individually associated with childhood trauma, homelessness and autism (Happé et al. 2016; Homeless Link 2014; Torjesen 2019). These findings caution against examining autism in isolation when considering risk of homelessness. Furthermore, the multiple adverse factors reported substantiate the suggestion that autistic people are at risk of multiple exclusionary homelessness (MEH). The presence of these different factors can result in ‘deep exclusion’, requiring broader consideration of complex needs and individualised support (Fitzpatrick, Bramley & Johnsen 2013). For autistic adults experiencing homelessness, it is imperative to consider how the interaction of these different factors work to further exclude an already ostracised population.

Findings highlighted socioeconomic exclusion as the main driver of homelessness for participants. The socioeconomic disadvantages reported reflect the findings of outcome studies, which have found that autistic adults are economically and socially disadvantaged in comparison with both the general population and other neurodevelopmental disability groups (Howlin & Magiati 2017). However, identification of both complex life factors and socioeconomic disadvantage gives only a cursory explanation of why autistic adults are at increased risk of homelessness. In focussing on these elements individually, we are at risk of ignoring ‘the reality of a profoundly unequal set of risks, and potentially disguise deeper structural and other causes that may be identifiable, and possibly also preventable’ (Bramley & Fitzpatrick 2018: 97).

Participants were excluded from normative society due to their perceived inability to fit in and adhere to social norms. For example, whilst an autism diagnosis in childhood should have increased support, participants were separated from their peers and denied the opportunity to take qualifications. Additionally, participants faced multiple barriers to employment as employers refused to adapt their processes to increase accessibility. The exclusions participants described were therefore more than a socioeconomic problem, resulting also from marginalisation or discrimination. Within CDS, marginalisation and discrimination can be understood as relational concepts arising from deep psychological resistance to those who are positioned as ‘other’ (Goodley 2013). Positioning individuals as other calls into question what constitutes appropriate humanness, a consideration that is not restricted to autism but is also concerned with other forms of difference and experiences of marginality (Titchkosky 2012). Throughout their lives, participants had multiple labels assigned to them including autistic, homeless, mentally ill, or transgender. Each of these labels risks denying personhood and can impact on life chances in areas including employment, relationships and housing (Goodley 2020; Link & Phelan 2001).

How labels come to represent otherness can be at least partially explained by expected social, cultural, and political norms. In the Global North, human value is embedded in ableist concepts of individualism and productivism and informed by gendered and racialised attachments to notions of ideality (Taylor 2020; Van Aswegen & Shevlin 2019). As Taylor (2020: 54) explains, ‘Individuals who are perceived to manifest undesirable differences in cognition, behaviour, communication, or performance appear to threaten notions of civic well-being, of nationhood, and of social reciprocity’. Although participants had multiple differences, the autism label has particular implications when considering social exclusion. As discussed above, in its very conception, autism is synonymous with deviation, as its base characteristics are measured against the prevailing norm. In this sense, autism serves as a catchall for behaviours that are labelled non-normative (Fleming 2020: 123). The rejection of autistic people is necessarily individualised and presented as though it is the impairment itself that results in exclusion (Titchkosky 2012). As a result, the deviation of autism becomes the problem of the individual who must change to fit into a society that continually overlooks their needs and difficulties (Howlin 2000).

Whilst participants experienced rejection throughout their lives, the majority were diagnosed, or came to identify, as autistic in adulthood, often after their first experience of homelessness. This emphasises the fact that it is the rejection of behaviours that deviate from accepted social norms that excluded participants from society and increased risk of homelessness. As Liddiard et al. (2019) explain, autistic people are only accepted if they are able to demonstrate ‘normal’ human behaviours. Behaviours that deviated from accepted norms eventually led to association with the autism label. However, all participants were perceived to be more able, at least in terms of their day-to-day functioning. Although the autism label of ‘high functioning’ was originally intended to describe those without a learning disability, it has become synonymous with expectations of increased functionality and better outcomes (Alvares et al. 2020). This misconception had implications for participants whose needs were misunderstood by professionals, resulting in limited access to welfare and appropriate support. Simultaneously, participants were excluded from ‘ordinary orders of daily life’, including education, employment, housing and relationships (Titchkosky 2012: 82). As such, participants represented a doubly disadvantaged group, rejected by society for their perceived inability to fit in whilst being denied the protections granted to those with more obvious disabilities.

In masking autism or withdrawing from society, participants accepted the permeation of socially normative ideals. When masking, participants were attempting to achieve a desired norm on the understanding that social spaces are subject to regulations that dictate how people should act (McLaughlin & Coleman Fountain 2018). In withdrawing from society, participants conceded that they were incapable of meeting these regulations. Masking or withdrawing did not ultimately protect participants against homelessness6 and had implications for mental wellbeing. Furthermore, in taking these measures, participants were unable to challenge the stigma associated with autism or perceived abnormal behaviours. However, study findings also highlighted how participants began to reclaim their autistic identities and contest their social exclusion. The complicated negotiation of identity participants embarked on provided a platform from which to begin challenging social norms and expectations. For example, after adopting an autistic identity, some participants challenged exclusionary employment practices with varying degrees of success.

These actions can be understood through the lens of dis/human theory, a resistant alternative that uses disability to disrupt entrenched ideas of what it means to be human (Goodley 2020). Whilst humanity and disability have historically been considered in juxtaposition with one another, Goodley and Runswick-Cole (2016) argue that the friction between these constructs can in fact be productive. Humanity is not a rigid concept, and disability labels, such as ‘autistic’, can inform and broaden what we understand as being human. In this sense, the label is reframed in terms of possibility and disruption (Goodley 2020). Indeed, the recent repositioning of autism as a form of neurodiversity both reclaims the humanity of autistic people whilst seeking acceptance of neurological differences. In considering risk of homelessness, the application of dis/human theory demands a recognition of autistic people’s rights to education, work, welfare and housing whilst simultaneously questioning normative notions of these social practices (Goodley & Runswick-Cole 2016).

Approaching autism from a dis/human perspective recognises that people are more complex than DSM-V criteria allows, placing less emphasis on binary distinctions of ‘autistic’ or ‘not autistic’ (Richards 2016). Regardless of whether or not an individual meets strict diagnostic criteria, the fact they either associate with this label or it has been associated with them has implications for how they are understood and supported by society. This consideration is particularly pertinent given the high prevalence of autistic traits found in homeless populations that would not necessarily warrant a clinical diagnosis (Churchard et al. 2019). Because participants either tended to be diagnosed in adulthood or rejected their childhood diagnoses due to associated stigma and ostracisation, it took time for participants to negotiate an autistic identity. When autism was utilised to challenge exclusionary practices, this was usually done on an individual basis, with limited lasting impact. It could therefore be argued that risk of homelessness for autistic adults will only be mitigated when a collective effort is made to address the exclusions inherent in our social structures and practices. Repositioning autism and associated behaviours in line with dis/human theory is a natural precursor to addressing these exclusions. In becoming dis/human, autistic people reclaim their humanness and associated human rights.

Conclusion

Whilst this paper has focused on a theoretical reinterpretation of findings, it is acknowledged that there are practical adaptations to services that could benefit autistic people experiencing homelessness. The autistic experience of homelessness, limitations of study design and practical implications for services and policy are discussed in previous work by Stone (2021).

The multiple hardships and co-occurring conditions participants reported cautions against examining autism in isolation. Uncovering this complexity has only been possible through the application of narrative methods, which take a whole life approach. The dis/human perspective recognises the intersectional relationship between disability and other identities, which have historically been considered less than human (Liddiard et al. 2019). This is significant for homeless, autistic people who may acquire labels that impact opportunity and social acceptance. Discussion of additional intersectional factors, such as sexuality and gender identity, was beyond the remit of this article but should be considered in the development of further theory.

This paper has associated the socioeconomic exclusion participants experienced with the permeation of normative ideals in society. This reinterpretation of autism exclusion has implications for how society understands and supports those who deviate from accepted norms. Positioning autism within dis/human theory can provide a framework from which to challenge normative assumptions. In practical application, this could improve access to education, employment and support services for autistic adults and mitigate against homelessness. However, the participants of this study are representative of a wider group who face significant daily difficulties resulting from society’s understanding of autism, co-occurring conditions and homelessness. Service providers and policy makers must therefore also take responsibility for repositioning autism and driving change. As such, addressing the risk of homelessness for autistic people requires a dualistic approach: promoting change within existing frameworks whilst challenging these same frameworks that produce inequity.

Notes

1Self-identification allowed for the inclusion of individuals who associated with the autism label. Understanding the autism label as a way of categorising difference does not negate its use in clinical and cultural settings. 

2Fletcher’s study used a critical realist framework to examine the effects of major agricultural policy on farm women in Canada. 

3UK secondary schools accommodate pupils between the ages of 11 and 18. 

4In the UK, college refers to an education setting, usually attended between the ages of 16–18, prior to University.