The Extent and Nature of Autistic People’s Violence Experiences During Adulthood: A Cross-sectional Study of Victimisation

Disabled people are at significantly higher risk of experiencing sexual and physical violence during their lifetime than people without disabilities (Codina et al., 2022; Krnjacki et al., 2016; Mailhot Amborski et al., 2021). The degree of elevated risk varies across disability type, with those who have an intellectual disability or mental illness experiencing the highest rates of violence and abuse (Byrne, 2018; Fisher et al., 2016; Nixon et al., 2017). There is now a substantial research base investigating risk factors, sequelae and prevention strategies for these vulnerable populations (e.g., Byrne, 2018; de Vries et al., 2019; McGilloway et al., 2020; Morgan et al., 2016). Fewer studies have examined violence exposure for autistic people, particularly those without co-occurring intellectual disability.

To date, most research investigating violence amongst autistic people has been based on parent report and largely focused on peer victimisation and abuse during childhood—consistently showing that autistic children experience higher rates of bullying, assault and child maltreatment compared to typical children (Hellström, 2019; Mandell et al., 2005; McDonnell et al., 2019; Pfeffer, 2016). There is emerging evidence that this increased risk continues well into adulthood and extends to criminal victimisation such as physical and sexual assault. Studies that have gathered first-hand information about sexuality, adverse experiences and trauma over the lifespan from autistic adults have found high-rates of sexual and physical violence (Brown-Lavoie et al., 2014; Hartmann et al., 2019; Pecora et al., 2019, 2020; Reuben et al., 2021; Rumball et al., 2020), although it is unclear from these studies whether the violence occurred during childhood, adulthood or both.

Only three studies have included measures with questions specifically pertaining to violence experienced during adulthood. In one large international study, autistic adults reported significantly higher rates of coerced and forced sexual activity, were more likely to have been sexually abused or threatened by a relationship partner and were more likely to be physically assaulted during their adult years compared to non-autistic adults (Griffiths et al., 2019). Similarly, Weiss and Fardella (2018) found that rates of sexual assault and physical assault with a weapon during adulthood were higher for autistic compared to non-autistic adults. In contrast, Gibbs et al. (2021) found no difference in reported physical and sexual violence during adulthood for autistic and non-autistic adults, although this information pertained to the previous two years only.

Bronfenbrenner’s ecological model (1977) has been proposed as a model for understanding factors that may contribute towards interpersonal violence amongst autistic people (Fardella et al., 2018). This model proposes that victimisation is the result of a complex interplay of four levels of factors i.e. ontogenic development (individual factors or experiences), the microsystem (the immediate context in which the victimization occurs, intimate and family relationships etc.), the exosystem (such as school, workplace, community supports) and the macrosystem (broader societal structures/cultural norms). At the ontogenic and microsystem levels, individual characteristics associated with autism such as social-communication differences may make it difficult to recognise dangerous situations or manipulation and coercion by others and lead to increased isolation and fewer protective relationships, all of which may contribute to disproportionately high rates of violence for autistic people (Kerns et al., 2015; Paul et al., 2018; Roberts et al., 2015). At the exosystem and macrosystem levels, socio-cultural factors such as stigma and discrimination often associated with autism (Botha & Frost, 2020; Shtayermman, 2009), inaccurate perceptions of autistic people as inferior or incapable (Heasman & Gillespie, 2018) and negative evaluations of autistic people by others (Sasson et al., 2017) may further increase the likelihood of an autistic person being targeted by perpetrators, who often choose victims perceived as weaker, less able to defend themselves and/or less likely to make accusations against them (Nettleback & Wilson, 2002).

In the neurotypical population, men report more physical violence and women more sexual assaults, intimate partner violence and stalking (Lauritsen & Carbone-Lopez, 2011; US Department of Justice, 2019). These gender differences have been found to be narrowed amongst people with intellectual disability (Platt et al., 2017) and mental illness (de Mooji et al., 2015). Transgender people and those with non-conforming gender identities (TGNC) have been found to have elevated rates of sexual victimisation, harassment and stalking compared to cisgender people (Fedina et al., 2020; Newcomb et al., 2020). Most studies investigating violence in autistic adults have focused on sexual victimisation of women or have consisted of small sample sizes which have not allowed for gender comparisons. One study found higher rates of both sexual and physical violence for autistic women compared to autistic men (Gibbs et al., 2021). Pecora et al. (2020) found no evidence of increased unwanted sexual behaviour or advances for transgender autistic adults, however this was limited to people whose sex at birth was female and was compared to cisgender females only. Further research is needed to understand the association between gender and violence amongst autistic adults.

Overall, the limited research to date indicates that autistic people may be at heightened risk for victimisation and violence in adulthood, particularly sexual violence and especially for autistic women. However, the full extent and nature of violence exposure during adulthood is not yet fully understood. Most studies have focused on sexual victimisation, with little known about the rates of physical and other forms of violence (e.g., stalking, sexual harassment). Furthermore, all of these forms of violence rarely occur in isolation, with many individuals experiencing multiple (more than one type) of violence (Fisher et al., 2016). Multiple victimisation has been shown to be common amongst autistic children and adults with intellectual disabilities (Codina et al., 2022; Pfeffer, 2016) and is associated with more detrimental effects on mental health and more severe psychiatric symptoms (Appleyard et al., 2005; Finkelhor et al., 2007; Turner et al., 2016). Similarly, repeated incidents of the same type of violence are commonly reported by people, including disabled people (Classen et al., 2005; Conley et al., 2017; Nixon et al., 2017). Low rates of reporting violence to authorities by disabled people and poor management of these reports by criminal justice professionals (Child et al., 2011; Codina & Pereda, 2021; Fogden et al., 2016) also contribute to increased risk and trauma (McGilloway et al., 2020. To our knowledge, no existing studies have investigated multiple or repeated forms of victimisation in relation to autistic adults, and nor have they examined reporting or conviction rates.

The aim of the current study was to examine the nature and extent of violence exposure among autistic people without intellectual disability during adulthood. To our knowledge, this is the first study to measure both multiple and repeated victimisation across a broad range of violence types. In line with our hypotheses, we found that autistic adults reported higher rates of a range of violence types and were more likely to experience multiple forms of violence and repeated instances of the same type of violence than non-autistic adults. Autistic adults reported higher rates of sexual harassment, stalking and harassment, sexual violence and physical violence than non-autistic adults. In addition, approximately three-quarters of autistic adults had experienced two or more forms of violence and repeated instances of the same violence type since the age of 15 compared to approximately half of the non-autistic adults. These findings are consistent with the limited research that has been conducted in this area (Griffiths et al., 2019; Weiss & Fardella, 2018) and highlight that the risk of increased violence is wide-ranging, extends beyond childhood and is not limited to autistic people with co-occurring intellectual disability.

Our finding that autistic adults are not only more likely to report having experienced various forms of violence, but are also more likely to report multiple and repeated violence is particularly concerning given the elevated risk of significant mental health impacts associated with multiple victimisation (Appleyard et al., 2005; Finkelhor et al., 2007; Turner et al., 2016). The phenomenon of repeat victimisation, that is, that victimisation occurs disproportionately among certain individuals is a consistent finding amongst the general population (Farrell et al., 1995; Tseloni & Pease, 2003; Pease, 1998), with numerous studies finding that prior victimisation strongly predicts future victimisation (Finkelhor et al., 2007; Menard, 2002; Outlaw et al., 2002). A clear theoretical explanation for why this is the case, however, has yet to be determined. One explanation may be that risk factors for violence are relatively stable, for example, lifestyle patterns that may place people in dangerous situations such as a person’s tendency to participate in ‘high risk’ activities or continuing involvement in conflictual relationships (Dean et al., 2007). Alternatively, early experiences of victimisation may affect an individual’s personality, social development and relationship patterns, which may in turn lead to interpersonal problems that increase vulnerability (Classen et al., 2005; Goodman et al., 2001). The limited research with autistic adults has tended to focus on the relationship between individual characteristics and violence exposure with the rationale being that social communication difficulties may increase vulnerability to manipulation or coercion by others and impede an autistic person’s ability to identify and remove themselves from risky situations. Autistic people’s difficulty generalising and learning from prior experiences (Cannon et al., 2021; Pellicano & Burr, 2012) may also be a contributing factor to repeat victimisation among this group. Nevertheless, a clear association between individual characteristics, including social competence, emotion regulation difficulties and autistic traits, has yet to be demonstrated (Gibbs et al., 2021; Roberts et al., 2015; Weiss & Fardella, 2018).

Bronfenbrenner’s (1977) ecological theory when applied to understanding violence victimisation considers individual characteristics, however the emphasis is on the interplay of these characteristics within the social contexts in which people are embedded. The negative attitudes, stigma and stereotyping of autistic people (Botha & Frost, 2020; Cage et al., 2019; Sasson et al., 2017; Shtayermman, 2009) may contribute to potential perpetrators viewing an autistic person as inferior, incapable and compliant and therefore as an “attractive victim” (Meer & Combrinck, 2015). In addition, the social isolation that many autistic adults experience (Howlin, 2013) may limit access to important social supports and networks and increase the likelihood of them remaining in unhealthy or even abusive relationships. Autistic children experience high rates of bullying and peer victimisation (Griffiths et al., 2019; Pfeffer, 2016) and have also been shown to be more likely to experience abuse and maltreatment (Dinkler et al., 2017; Mandell et al., 2005). This early exposure to violence may result in psychological and emotional impacts such as poor self-concept, feelings of powerless and betrayal and post-traumatic symptoms which may in turn exacerbate vulnerability to repeat victimisation. In addition, autistic people have been found to be less likely to confide in others about their experiences of physical and sexual violence (Gibbs et al., 2021). Without support from family, professionals or services autistic people may have difficulty developing protective behaviours and/or extricating themselves from dangerous situations contributing to high rates of multiple and repeat victimisation. Future research must therefore examine individual, contextual and societal aspects that may contribute to elevated risk for autistic people. Understanding characteristics that increase vulnerability may assist clinicians to identify individuals who are at particular risk and inform the development of targeted education strategies and support services. Ultimately, however, research and policymaking should target the broader contextual and societal factors that place autistic people at increased risk for violence so that such risk can be mitigated.

Since gender is often associated with different types of violence, we compared rates of violence for autistic and non-autistic adults separately for men and women. Autistic women reported higher rates of all forms of violence. In contrast, there was no difference in rates of stalking and harassment, physical violence, multiple violence or repeated violence for autistic and non-autistic men. The number of non-autistic men who reported sexual harassment and sexual violence was too low to conduct statistical comparison. However, the proportions of autistic men who reported sexual harassment and sexual violence are considerably higher than prevalence rates from large population samples of violence during adulthood (56% for sexual harassment and 40% for sexual violence in our sample compared to 25% for sexual harassment and 4.7% for sexual violence from the Australian Personal Safety Survey; ABS, 2016). High rates of sexual victimisation have been found in studies of men with intellectual disability (Platt et al., 2017; Saxton et al., 2006). Disabled men, including autistic men, may be less likely to behave in ways that conform to masculinity stereotypes (Powers et al., 2008) decreasing their power in relationships and leaving them vulnerable to the abuses and subjugation typically experienced by women (Shakespeare, 1999).

The small sample size of TGNC autistic adults and the lack of a comparison group meant that we were unable to conduct any comparative analysis. Nevertheless, rates of reported violence for this group were high across all violence types (ranging from 60% to 86.7%) with an alarmingly high proportion (93.3%) reporting repeat victimisation. This is consistent with research of TGNC people in the general population who have been found to have elevated rates of violence and victimisation compared to cisgender people (Aparicio-Garcia et al., 2018; Grant et al., 2011; James et al., 2016). These gender-related findings can be interpreted within an intersectionality framework (Crenshaw, 1989), which posits that multiply marginalised identities intersect to amplify the risk of violence exposure. Being an autistic person represents one social identity that confers additional risk (a person with a disability). Being an autistic woman or TGNC person combines the added risk of disability with the risk associated with being female or of non-conforming gender.

We also analysed rates of reported violence by gender for both autistic and non-autistic groups separately. As expected, gender differences in the patterns of violence (more physical violence reported by men and more sexual violence reported by women) were apparent in the non-autistic but not the autistic group. This pattern of narrowing of gender differences has also been found amongst people with intellectual disability and those with severe mental illness (Cohen et al., 2006; Harrell, 2012b; Dean et al., 2007; de Mooji et al., 2015; Smith, 2008) and in our study is due to the higher rates of sexual violence reported by autistic men and physical violence reported by autistic women. Overall, these findings underscore the importance of reconsidering issues of gender and violence as they apply to autistic people when undertaking research in this area and also when working with autistic people in clinical settings. Caution is warranted, however, given the small sample sizes, particularly for men and TGNC individuals. Further research with larger samples of men, women and TGNC people will provide important information about the nuanced nature of violence across gender for autistic people. In the meantime, clinicians and other service providers need to be aware of the disproportionate rates of victimisation for this population across all types of violence regardless of gender.

To our knowledge, this is the first time that contextual factors for a range of violence types have been investigated for autistic adults. We gathered information about the age range of the individual at the time of the most recent incident, the gender of the perpetrator and their relationship to them, whether drugs and alcohol played a role and whether they had confided in anyone or reported to police. In the general population, violence is most common during the early adulthood period (Conley et al., 2017; Fedina et al., 2020), however a substantial proportion of the most recent incidents reported by autistic adults occurred after age 25, ranging from approximately 42% of recent sexual harassment incidents for men and stalking/harassment incidents for women, up to 100% of the stalking and harassment incidents reported by autistic men. Just over three quarters of the most recent sexual violence incidents reported by TGNC autistic adults occurred after age 25. Although the mechanisms that result in disproportionately high rates of violence victimisation amongst autistic adults are yet to be understood, these findings indicate that certain contextual risk factors such as risky sexual behaviours and substance use (Duval et al., 2020) that predominate in the early adulthood period and correlate with rates of violence in the general population may not be associated with violence for autistic people. In fact, in our study, 80 to 90% of recent incidents of violence did not involve drugs or alcohol which is consistent with the low rates of drug use and alcohol consumption that has been reported in large studies of autistic adults (Weir et al., 2021).

The perpetrators of violence were most commonly known to the autistic person with a substantial proportion of incidents occurring in the context of intimate relationships or friendships. This is consistent with what has been found in the general population for women (ABS, 2016; James et al, 2016). That said, men are most frequently physically assaulted by strangers, which was not the case amongst our sample of autistic men. Similarly, the vast majority of violence perpetrated against men in the general population is by other men, particularly physical violence; however, here, just over 40% of recent physical violence incidents were perpetrated by women. Although we were only able to gather information from a small sample of autistic men, these findings provide further support for the notion that the gender patterns related to violence in the general population may not apply to autistic people and further supports the call for more research in this area. In terms of confiding in others about their experiences of violence, autistics adult reported that they had told someone following the majority of stalking and harassment incidents (75–88.9%). But this was not the case for physical violence, and even less so for sexual violence. These low rates of confiding in others are largely consistent with findings from an earlier study with a different sample (Gibbs et al., 2021) and are deeply concerning. Social support in relation to traumatic experiences and access to appropriate psychological services, both of which have been shown to positively impact well-being after trauma (Dworkin et al., 2018; Rusch et al., 2015) are reliant on the person disclosing violence to their families, networks and/or treating professionals. It is critical that autistic people are able to disclose violence experiences so that appropriate mental health and social supports can be put in place and protective strategies can be identified and implemented to reduce the risk of repeated victimisation.

Approximately one quarter of the recent incidents reported (n = 75) were viewed as a crime at the time by the person involved. This could indicate that the remaining incidents were less serious and did not meet the threshold of criminal violence. The question posed, however, related to the autistic person’s perception of the incident at the time it occurred. It is also possible that even serious incidents of violence were not recognised as such at the time, which is common in the case of intimate partner and dating violence (Mouzos & Makkai, 2004). Regardless, despite considering the incident to be a criminal act only 54% (n = 41) were reported to police, and of those only 5 resulted in a conviction. Low rates of reporting to police and low rates of conviction have also been observed in a number of population groups including college students, individuals with intellectual and other disabilities and TGNC people and is often linked to distrust and fear of the legal system and legal processes (Conley et al., 2017; Donovan & Hester, 2008; McGilloway et al., 2020). The most common reasons for not reporting to police provided by our autistic participants were that police would not be able to assist them, lack of trust in police, fear of legal processes and a perception that they would not be believed. This is consistent with several recent studies investigating interactions between autistic people and police that have described low rates of satisfaction and feelings of fear and mistrust of the justice system and law enforcement (Gibbs & Haas, 2020; Hartmann et al., 2019; Koffer Miller et al., 2021). Efforts to improve reporting and conviction rates should be undertaken so that autistic people feel confident reporting crimes to police, perpetrators are brought to justice and necessary supports and compensation can be provided.

This study has a number of limitations. First, the autistic group was predominantly comprised of those who had completed tertiary study and may not be representative of the broader population of autistic people. We also did not conduct any verification of diagnostic status beyond a self-report screening tool. Second, the sample size of males and TGNC adults was low and there was only one TGNC person in the non-autistic group which meant that no comparative analyses were able to be conducted. Nevertheless, most of the research into autistic people’s experiences of violence in adulthood has been focused on autistic women. This work therefore provides initial insight into the violence experiences of men and TGNC autistic adults. Third, the data were based on retrospective reporting which may have resulted in recall biases. Finally, although we investigated a broader range of violence types than in previous studies, there are other forms of violence and victimisation that were not measured such as emotional and psychological abuse and workplace bullying. Future research aimed at extending knowledge in relation to the full extent and nature of violence experiences should also include these additional forms of violence.

The findings from this study provide further evidence of the extent of violence exposure for autistic people that occurs well into the adult years, with autistic women and those who are transgender or gender nonconforming appearing to be at greatest risk. The study also offers an initial insight into the context of violence experiences among this group with some indication that most violence is perpetrated by men, by people that are known to them, rarely involves drugs or alcohol and often are not reported, even when considered to be a criminal act. These results have several important implications for practice and policy. Health care and allied health practitioners should be made aware of the higher risk of violence for this group so that trauma symptoms are identified and appropriate treatment can be provided. Similarly, specialised training for law enforcement and legal professionals in relation to autism and availability of advocates to support autistic people throughout the legal process may improve criminal justice interactions and outcomes and increase willingness amongst autistic people to report criminal violence to police. The focus of future research should be on gaining a fuller understanding of factors underlying the increased risk of violence amongst autistic people so that targeted prevention strategies can be developed and evaluated. Until such time as any broader societal and systemic factors can be identified and addressed, autistic adolescents and adults should be provided with direct education about their rights, the meaning of consent, what constitutes violence and how to access related information, services and support should they be victimised or have concerns about situations in which they may find themselves.