**5. Autism spectrum disorders and gender identity**

Among the general population, the prevalence of autism spectrum disorder (ASD) in children is estimated at 1% with a ratio of 1:42 for boys and 1:189 for girls respectively [33]. The evidence that suggests an overrepresentation of ASD in gender diverse samples, particularly in children and adolescents [34, 35], is robust and largely accepted by scientific community. The association between GD and ASD has been of great clinical interest because it has implications for diagnosis and treatment.

### **5.1 Co-occurrence rates**

Trans Pathways study conducted in Australian large sample of trans and genderdiverse young people (n = 859; mean age = 19.4), found that 22.5% of the sample

reported having received a formal ASD diagnosis, while more than one third (35.2%) had highly suspected but undiagnosed ASD [34]. This type of large epidemiological study is difficult to construe in view of diagnostic imprecision of gender dysphoria and heterogeneity of both, GD and ASD constructs and therefore studies in this field significantly vary in methodology and chosen diagnostic constructs. A focus on diagnosis is less sensitive to the presence of subthreshold or mild autistic symptoms, which is why some studies utilised the Broader Autistic Phenotype (BAP) that is defined as a collection of sub-diagnostic autistic traits more common in families of individuals with ASD than in the general population. Evidence of an intermediate phenotype and a latent construct in autism was first reported in the landmark twin study of Folstein and Rutter [36]. Jones et al. [37] used Autism Spectrum Quotient – AQ [38] to measure BAP in a sample of adults with GD, typical adults and adults diagnosed with ASD, and found that 17.5% of the GD sample had a score above the AQ cut-off for BAP. Interestingly, more females with GD scored above the cut-off than males with GD, which is in contrast to the recognised male female distribution in ASD.

The first systematic study into the incidence of autism diagnosis in young people referred to a specialised gender clinic via the use of a diagnostic interview, reported an ASD' higher than expected prevalence rate of 7.8%, and an overrepresentation of ASD diagnoses in boys compared to girls with a ratio of 3:1 [39]. While, overall, this study was methodologically sound, it sadly lacked a clinical control group for comparison. Contrary, the study by Pasterski et al. [40] has shown no difference in relation to ASD' prevalence rate between trans people and the general population by utilising the threshold for a potential diagnosis with an ASD-rate of 5.5%. The disparity of these findings and the difference in prevalence rates could be as a result of the chosen study populations with recognised difference in presentation between children and adults, study design and methodology, utilised diagnostic categories and assessment tools.

Skagerberg et al. [41] reported ASD scores that fell, on average, in the mild/ moderate range in a sample of children and adolescents with GD with no significant difference between boys and girls with GD, and scores that fell in the normal range in a control sample of typically developing young people. Skagerberg et al. measured autistic symptoms using a quantitative measure - the Social Responsiveness Scale [42], that was also used in another controlled study of children with GD [43] with 44.9% of GD' sample scoring within the clinical range for autistic traits with, on average, moderate scores. This Canadian study also examined risk factors for ASD with an overlap of only high birth weight, but not the other risk factors, with both, raised gender nonconformity and autistic traits among children with GD [43].

Glidden et al. [44] systematically appraised 19 out of 58 available articles regarding the co-occurrence of gender dysphoria and ASD from Medline, PubMed, PsycINFO, and Embase databases in the period from 1966 to July 2015. The authors of this systematic literature review concluded that the research in to the co-occurrence between gender dysphoria and ASD is limited, especially for adults. The literature investigating ASD in children and adolescents with gender dysphoria showed a higher prevalence rate of ASD compared with the general population. Since Gidden's systematic review, recent well-designed Dutch study confidently confirmed an over-representation of symptoms of ASD in children and adolescents with GD [45]. Their estimated prevalence of ASD was 14.5%, which is approximately four times higher than the 3.5% in the normative sample and much higher than the prevalence estimate of 1% found in the general population [33]. Their GD sample showed elevated levels of autistic symptomatology on all subdomains, not just on stereotyped behaviour and resistance to change' measures. van der Miesen and colleagues [45] found that young people with GD had more reported autistic

**45**

obsessional interests [52].

**5.3 Developmental hypothesis**

*Autism and Gender Identity*

as non-heterosexual.

*DOI: http://dx.doi.org/10.5772/intechopen.97517*

aspect of personal identity' development [34].

ASD is estimated between 4% and 8% [39, 49–51].

**5.2 Hypotheses attempting to explain ASD/GD association**

symptoms compared to typically developing children and adolescents, but less reported autistic symptoms compared to children and adolescents with ASD. There seems to be less studies that took an alternative root and investigated GD symptoms within an ASD population. Australian survey by George and Stokes [46] aimed at measuring prevalence of "gender variance" in ASD and found that individuals with ASD of all ages report increased homosexuality, bisexuality, and asexuality, but decreased heterosexuality. Sexual Orientation was surveyed using the Sell Scale of Sexual Orientation in an international online sample of 309 young adults with ASD that were screened with Autism Quotient (M = 90, F = 219, M = 32.30 years, SD = 11.93) which was compared to sexual orientation of 310 controls that were represented by aged-matched neurotypically developing individuals (M = 84, F = 226, M = 29.82 years, SD = 11.85). In the group with ASD, 69.7% identified as non-heterosexual, while in the control group, 30.3% identified

Strang et al. [47] found that children with ASD were 7.59 times more likely to express gender variance by expressing "wishes to be of the other gender" as per Child Behaviour Checklist [48] compared to their neurotypical peers and established equal sex distribution for the gender variance. Similarly, Janssen et al. [49] found that children with ASD were 7.76 times more likely to express gender variance than children from the non-referred comparison group, with no significant difference between boys and girls. There is a consensus that in most cases, gender diverse identities and behaviours are stable and not secondary to ASD but co-occur as an

The exact numbers accounting for the overlap between autism and gender variance has a wide degree of variation, ranging between 6% and 26% of for ASD among gender-variant people, while the rate of gender variance among people with

Gender diverse behaviours, including crossdressing, tomboyism and paraphilias in children and adolescents with ASD may be considered as part of the ASD phenotype and representations of unusual, restricted interests and the development of atypical gender identity in autism could relate to the developmental rigidity that is characteristic of autism. This hypothesis focusing on individual psychological characteristics and obsessional interests suggesting that gender could be among the preoccupations or obsessions often seen in ASD was not fully supported by van der Miesen and colleagues [52]. The study of VanderLaan et al. [53], which suggested that specifically intense obsessional interests are one of the hypothesised mechanisms underlying the possible GD-ASD co-occurrence. Van der Miesen's findings highlight several subdomains of the autistic spectrum that might be involved in this possible association, including social and communication difficulties, but not

The individuals with ASD might not reach normative flexibility in gender development that will equip them with necessary skills to deal with gender variant feelings, which might explain the overrepresentation of ASD in GD. Furthermore, Robinow [54] suggested that neurobiological abnormalities associated with reduced social functioning in ASD, such as those found for frontal and temporal regions, might make it difficult for some children to acquire concepts regarding gender norms. Clinicians have observed that at least some children with GID misclassify

#### *Autism and Gender Identity DOI: http://dx.doi.org/10.5772/intechopen.97517*

*Autism Spectrum Disorder - Profile, Heterogeneity, Neurobiology and Intervention*

female distribution in ASD.

and assessment tools.

reported having received a formal ASD diagnosis, while more than one third (35.2%) had highly suspected but undiagnosed ASD [34]. This type of large epidemiological study is difficult to construe in view of diagnostic imprecision of gender dysphoria and heterogeneity of both, GD and ASD constructs and therefore studies in this field significantly vary in methodology and chosen diagnostic constructs. A focus on diagnosis is less sensitive to the presence of subthreshold or mild autistic symptoms, which is why some studies utilised the Broader Autistic Phenotype (BAP) that is defined as a collection of sub-diagnostic autistic traits more common in families of individuals with ASD than in the general population. Evidence of an intermediate phenotype and a latent construct in autism was first reported in the landmark twin study of Folstein and Rutter [36]. Jones et al. [37] used Autism Spectrum Quotient – AQ [38] to measure BAP in a sample of adults with GD, typical adults and adults diagnosed with ASD, and found that 17.5% of the GD sample had a score above the AQ cut-off for BAP. Interestingly, more females with GD scored above the cut-off than males with GD, which is in contrast to the recognised male -

The first systematic study into the incidence of autism diagnosis in young people referred to a specialised gender clinic via the use of a diagnostic interview, reported an ASD' higher than expected prevalence rate of 7.8%, and an overrepresentation of ASD diagnoses in boys compared to girls with a ratio of 3:1 [39]. While, overall, this study was methodologically sound, it sadly lacked a clinical control group for comparison. Contrary, the study by Pasterski et al. [40] has shown no difference in relation to ASD' prevalence rate between trans people and the general population by utilising the threshold for a potential diagnosis with an ASD-rate of 5.5%. The disparity of these findings and the difference in prevalence rates could be as a result of the chosen study populations with recognised difference in presentation between children and adults, study design and methodology, utilised diagnostic categories

Skagerberg et al. [41] reported ASD scores that fell, on average, in the mild/ moderate range in a sample of children and adolescents with GD with no significant difference between boys and girls with GD, and scores that fell in the normal range in a control sample of typically developing young people. Skagerberg et al. measured autistic symptoms using a quantitative measure - the Social Responsiveness Scale [42], that was also used in another controlled study of children with GD [43] with 44.9% of GD' sample scoring within the clinical range for autistic traits with, on average, moderate scores. This Canadian study also examined risk factors for ASD with an overlap of only high birth weight, but not the other risk factors, with both, raised gender nonconformity and autistic traits among children with GD [43]. Glidden et al. [44] systematically appraised 19 out of 58 available articles regarding the co-occurrence of gender dysphoria and ASD from Medline, PubMed,

PsycINFO, and Embase databases in the period from 1966 to July 2015. The authors of this systematic literature review concluded that the research in to the co-occurrence between gender dysphoria and ASD is limited, especially for adults. The literature investigating ASD in children and adolescents with gender dysphoria showed a higher prevalence rate of ASD compared with the general population. Since Gidden's systematic review, recent well-designed Dutch study confidently confirmed an over-representation of symptoms of ASD in children and adolescents with GD [45]. Their estimated prevalence of ASD was 14.5%, which is approximately four times higher than the 3.5% in the normative sample and much higher than the prevalence estimate of 1% found in the general population [33]. Their GD sample showed elevated levels of autistic symptomatology on all subdomains, not just on stereotyped behaviour and resistance to change' measures. van der Miesen and colleagues [45] found that young people with GD had more reported autistic

**44**

symptoms compared to typically developing children and adolescents, but less reported autistic symptoms compared to children and adolescents with ASD.

There seems to be less studies that took an alternative root and investigated GD symptoms within an ASD population. Australian survey by George and Stokes [46] aimed at measuring prevalence of "gender variance" in ASD and found that individuals with ASD of all ages report increased homosexuality, bisexuality, and asexuality, but decreased heterosexuality. Sexual Orientation was surveyed using the Sell Scale of Sexual Orientation in an international online sample of 309 young adults with ASD that were screened with Autism Quotient (M = 90, F = 219, M = 32.30 years, SD = 11.93) which was compared to sexual orientation of 310 controls that were represented by aged-matched neurotypically developing individuals (M = 84, F = 226, M = 29.82 years, SD = 11.85). In the group with ASD, 69.7% identified as non-heterosexual, while in the control group, 30.3% identified as non-heterosexual.

Strang et al. [47] found that children with ASD were 7.59 times more likely to express gender variance by expressing "wishes to be of the other gender" as per Child Behaviour Checklist [48] compared to their neurotypical peers and established equal sex distribution for the gender variance. Similarly, Janssen et al. [49] found that children with ASD were 7.76 times more likely to express gender variance than children from the non-referred comparison group, with no significant difference between boys and girls. There is a consensus that in most cases, gender diverse identities and behaviours are stable and not secondary to ASD but co-occur as an aspect of personal identity' development [34].

The exact numbers accounting for the overlap between autism and gender variance has a wide degree of variation, ranging between 6% and 26% of for ASD among gender-variant people, while the rate of gender variance among people with ASD is estimated between 4% and 8% [39, 49–51].

#### **5.2 Hypotheses attempting to explain ASD/GD association**

Gender diverse behaviours, including crossdressing, tomboyism and paraphilias in children and adolescents with ASD may be considered as part of the ASD phenotype and representations of unusual, restricted interests and the development of atypical gender identity in autism could relate to the developmental rigidity that is characteristic of autism. This hypothesis focusing on individual psychological characteristics and obsessional interests suggesting that gender could be among the preoccupations or obsessions often seen in ASD was not fully supported by van der Miesen and colleagues [52]. The study of VanderLaan et al. [53], which suggested that specifically intense obsessional interests are one of the hypothesised mechanisms underlying the possible GD-ASD co-occurrence. Van der Miesen's findings highlight several subdomains of the autistic spectrum that might be involved in this possible association, including social and communication difficulties, but not obsessional interests [52].

#### **5.3 Developmental hypothesis**

The individuals with ASD might not reach normative flexibility in gender development that will equip them with necessary skills to deal with gender variant feelings, which might explain the overrepresentation of ASD in GD. Furthermore, Robinow [54] suggested that neurobiological abnormalities associated with reduced social functioning in ASD, such as those found for frontal and temporal regions, might make it difficult for some children to acquire concepts regarding gender norms. Clinicians have observed that at least some children with GID misclassify

their own gender, even at ages beyond those in which correct self-labelling is expected [55]. Social communication deficits might, therefore, underlie the cognitive "lag" that many GD children exhibit in terms of their gender constancy development [55].

Erik Erikson described eight stages of psychosocial development through which a neurotypically developing adult should pass from infancy to adulthood [56]. As articulated by Erikson, Identity versus Role Confusion represents the fifth stage of psychosocial development that take place during adolescence between the ages of 12 and 19. It has been hypothesised that individuals with ASD become acutely aware of their uniqueness and differences compared to others during their formative years, and, as a result, may develop confusion of identity and identity crisis which could include gender nonconforming behaviour and GF.

#### **5.4 Social perception and preoccupations hypotheses**

This theory implies that core ASD symptoms of social deficit will likely influence child's ability to interpret social cues when it comes to gender conforming behaviour and navigating nuanced social interactions with same and opposite peer groups. Specific neuropsychological profiles with deficits in "theory of mind," the ability to attribute mental states (beliefs, intents, desires, etc.) to oneself and others and recognise that these are different from one's own, may affect development of the "self" in general. When expressing their gender variance and sexuality young people with ASD may be less inhibited by the social norms or even more oppositional to social restrictions when expressing their gender variance. It could be theorised that excessively rigid cognitive style or dichotomous thinking pattern could predispose a child with ASD to interpret slight gender nonconforming inclination as total and fundamental preference.

#### **5.5 Neurodevelopmental masculinisation**

The theory of the extreme male brain (EMB) stipulates that individuals with autism may develop an extreme variant of the typical male pattern of behaviours and cognitions originating from high levels of foetal testosterone [57]. While prenatal testosterone is linked to the the association between ASD and GD in assigned girls, explaining the male pattern of their identity and behaviour, same theory cannot applied to assigned boys. Adolescent girls with ASD had a significantly higher prevalence of endorsement of item 'the Wish to be of the Opposite Gender' compared to adolescent boys with ASD [52]. Thus, Van der Miesen et al. [52] partly supports Neurodevelopmental Masculinisation hypothesis but found no significant differences in CSBQ total score between boys and girls with GD, and diverging gender differences on the subdomains of ASD, which are not all consistent with the EMB theory, rendering it highly unlikely [52]. In a comparison sample of birth-assigned females diagnosed with GD, Jones and colleagues [37] established increased rates of ASD symptoms, while birth-assigned males diagnosed with GD did not have increased levels of ASD symptoms. Jones and colleagues hypothesised that elevated levels of foetal testosterone may lead not only to reduced empathy, reduced social interest, reduced social skills, and more ASD, but also contribute to developing GD via neurodevelopmental masculinisation pathway [37]. Among adults with ASD, the symptoms of tomboyism and bisexuality were commoner in females with ASD, while assertiveness and leadership, the aspects that are considered to be typically masculine were reportedly weaker in both, females and males with ASD, compared with typically developing controls [57]. This data signifies that an extreme male pattern might not apply to all aspects of gender roles and sexuality.

**47**

*Autism and Gender Identity*

*DOI: http://dx.doi.org/10.5772/intechopen.97517*

**5.6 ADHD comorbidity theories**

A brain MRI study in individuals with ASD also found attenuated typical gender differences in white matter tracts [58], providing support for gender atypicality as

Evidence suggests that core ADHD symptoms and associated externalising disorders are overrepresented in both groups of interest, young people with neurodiversity and young people with gender variance. One large retrospective study had a surprising finding of a significant overrepresentation of gender variance, occurring 6.64 times more frequently among children with ADHD, than among a non-referred comparison group [47]. This study determined that parental report of gender variance was significantly greater in two groups of children, ASD group (5.4%) and ADHD group (4.8%) that collectively represent children with neurodevelopmental disorders, while the proportion of children with gender variance among combined medical group (1.7%) and non-referred comparison group (0–0.7%) were statistically different from ASD and ADHD groups; gender variance occurred equally in girls and boys [47]. In ADHD, impulse control difficulties are essential criteria for diagnosis and could potentially affect gender expression by reducing ability to inhibit primary gender impulses in spite of societal pressure to conform to gender stereotype [47]. Among transgender youth with ASD, children and adolescents may be less aware of the social stereotypes, hence the ASD/ADHD cohort are likely to ignore the societal influences against cross-gender expression and express their gender inclinations more freely or even parade their feelings of gender incongruence as an oppositional response to unaccepted societal rules.

**6. Clinical guidelines for co-occurring autism spectrum disorder and** 

Adolescents with ASD may not embody a binary transgender presentation, while some may conflate sexuality with gender and need affirming education. In the course of psychological therapy, one may wish to explore whether traits of ASD such as intense/obsessional interests or social communication deficits contribute to a child's gender schema (e.g. wishing to be a specific anime character) and, eventually

The initial clinical guidelines for ASD-GD [59] in adolescents have been developed using Delphi method [60] and contain strong advocacy for adolescents with ASD to gain equal access to gender-related services and not be precluded from gender affirming care when diagnostic criteria of GD are met. The guidelines emphasise the need for carrying out gender assessments in tandem with an in-depth consideration and accommodation of ASD-related factors that may impact genderrelated exploration and identify broader needs of neuro-diverse adolescents [59]. Since the initial guidelines' have been publicised, more recent study by Strang et al. [61] reported on a proposed community-driven clinical model to attend to the broader care needs and preferences of adolescents with ASD/neurodiversity and GD. It is also important to engage young people with a lived experience of ASD/GD and their carers into a productive dialogue about further services and interventions as they hold unique insights into how services can best respond to the complex needs of affected individuals and promote many other related domains including education, employment, housing and family services. Certainly, there are reports that the education and community outreach programs as part of gender diversity services are very important to the service users and have grown exponentially in a

**gender dysphoria or incongruence in adolescents**

way that was not initially anticipated [62].

one of the potential underlying mechanisms for co-occurring GD–ASD.

*Autism Spectrum Disorder - Profile, Heterogeneity, Neurobiology and Intervention*

include gender nonconforming behaviour and GF.

total and fundamental preference.

**5.5 Neurodevelopmental masculinisation**

**5.4 Social perception and preoccupations hypotheses**

opment [55].

their own gender, even at ages beyond those in which correct self-labelling is expected [55]. Social communication deficits might, therefore, underlie the cognitive "lag" that many GD children exhibit in terms of their gender constancy devel-

Erik Erikson described eight stages of psychosocial development through which a neurotypically developing adult should pass from infancy to adulthood [56]. As articulated by Erikson, Identity versus Role Confusion represents the fifth stage of psychosocial development that take place during adolescence between the ages of 12 and 19. It has been hypothesised that individuals with ASD become acutely aware of their uniqueness and differences compared to others during their formative years, and, as a result, may develop confusion of identity and identity crisis which could

This theory implies that core ASD symptoms of social deficit will likely influence child's ability to interpret social cues when it comes to gender conforming behaviour and navigating nuanced social interactions with same and opposite peer groups. Specific neuropsychological profiles with deficits in "theory of mind," the ability to attribute mental states (beliefs, intents, desires, etc.) to oneself and others and recognise that these are different from one's own, may affect development of the "self" in general. When expressing their gender variance and sexuality young people with ASD may be less inhibited by the social norms or even more oppositional to social restrictions when expressing their gender variance. It could be theorised that excessively rigid cognitive style or dichotomous thinking pattern could predispose a child with ASD to interpret slight gender nonconforming inclination as

The theory of the extreme male brain (EMB) stipulates that individuals with autism may develop an extreme variant of the typical male pattern of behaviours and cognitions originating from high levels of foetal testosterone [57]. While prenatal testosterone is linked to the the association between ASD and GD in assigned girls, explaining the male pattern of their identity and behaviour, same theory cannot applied to assigned boys. Adolescent girls with ASD had a significantly higher prevalence of endorsement of item 'the Wish to be of the Opposite Gender' compared to adolescent boys with ASD [52]. Thus, Van der Miesen et al. [52] partly supports Neurodevelopmental Masculinisation hypothesis but found no significant differences in CSBQ total score between boys and girls with GD, and diverging gender differences on the subdomains of ASD, which are not all consistent with the EMB theory, rendering it highly unlikely [52]. In a comparison sample of birth-assigned females diagnosed with GD, Jones and colleagues [37] established increased rates of ASD symptoms, while birth-assigned males diagnosed with GD did not have increased levels of ASD symptoms. Jones and colleagues hypothesised that elevated levels of foetal testosterone may lead not only to reduced empathy, reduced social interest, reduced social skills, and more ASD, but also contribute to developing GD via neurodevelopmental masculinisation pathway [37]. Among adults with ASD, the symptoms of tomboyism and bisexuality were commoner in females with ASD, while assertiveness and leadership, the aspects that are considered to be typically masculine were reportedly weaker in both, females and males with ASD, compared with typically developing controls [57]. This data signifies that an extreme male pattern might not apply to all aspects of gender roles and sexuality.

**46**

A brain MRI study in individuals with ASD also found attenuated typical gender differences in white matter tracts [58], providing support for gender atypicality as one of the potential underlying mechanisms for co-occurring GD–ASD.
