Gender, sexuality, race and culture

This topic covers where autism meets gender, sexuality, race, ethnicity, culture and class. Four entries are here: gender diversity, sexual orientation and sex education, race and ethnicity, and autism around the world. A closing section covers class, money and access: who is counted as autistic, what diagnosis and support cost families, how access to assessment varies, and who can afford to be open about being autistic.

Gender-diverse identities are more common among autistic people than in the general population: in two samples of autistic US adults, 16.1% and 19.8% were gender diverse, against 0.5% to 4.5% reported in general adult populations.¹

Gender-diverse identities are more common among autistic people than in the general population. In two samples of autistic US adults, 16.1% and 19.8% were gender diverse, against 0.5% to 4.5% reported in general adult populations, and a whole-population New Zealand study linking census records for 1,266,453 young people found autism in 3.9% of those whose gender was outside the male/female binary.^1,3 Autistic people are also overrepresented at gender clinics: about 6.3% of 1,843 adults referred to Canada's largest publicly funded adult gender service had a recorded autism diagnosis, and 9.7% of 786 gender-diverse children and adolescents at an Israeli national clinic.^9,8 Gender-diverse autistic people report more mental-health difficulties than other autistic people: among 9,027 autistic adolescents, the 36 identified as transgender or gender diverse had about four times the odds of a depression diagnosis, an effect at least as strong as a family history of depression.¹³ No outcome study of gender-affirming medical care specific to autistic people was found as of October 2026, and a systematic review found no proposed explanation for the overlap with substantial empirical support.²⁸

Autistic adolescents and adults report a wider range of sexual orientations than non-autistic people. In an international sample of 309 autistic and 310 non-autistic adults, 69.7% of the autistic group reported an orientation other than heterosexual, against 30.3% of the comparison group,²⁹ and in a larger online survey of 2,386 adults autistic respondents were less likely to describe themselves as heterosexual and more likely to report asexuality or another orientation.³⁰ In a whole-population New Zealand study, autism was identified in 2.2% of non-heterosexual young people and 1.4% of heterosexual young people, which the authors report as 1.8 times as high.³ A systematic review found asexuality reported more often in autistic samples and concluded that the link is not yet clarified.³³ Sex education is usually written for non-autistic students, and autistic adults describe a poor fit: a systematic review of 56 studies found autistic people share typical sexual desires but that sensory sensitivities and communication difficulties can complicate relationships, and called for tailored sexual-health education.³⁴ No study found as of October 2026 has measured what sex education autistic people receive nationally.

How autism is counted has changed by group. In the 2002 US surveillance data, Black, Hispanic and other children were less likely than White children to have a documented diagnosis, and the shortfall for Black children held across IQ levels,³⁵ while by 2022 Black, Hispanic, Asian and Pacific Islander and American Indian 8-year-olds were counted at similar or higher rates than White children.³⁶ A 2026 systematic review of US studies found delayed diagnosis the most common error, and named provider bias and culturally inequivalent tests among the reasons.⁴⁰ Black children were more often given a different label first: in one city sample they were 2.6 times less likely to receive the autism diagnosis at a first specialty visit, and more often given adjustment-disorder or conduct-disorder diagnoses instead of ADHD.³⁷ Among autistic adults, records studies find Black, Hispanic and Asian people less likely to have a recorded psychiatric diagnosis and lower continuity of depression treatment.⁴¹ Police contact is more common than among non-autistic peers: by age 21, about 20% of one autistic cohort had been stopped and questioned and nearly 5% arrested, in a study with no comparison group.⁴⁴ Most gene discovery has been done in people of European ancestry,⁴² though a study of more than 15,000 Latin American people found the relevant biology shared across ancestries.⁴³

Prevalence estimates are not comparable across countries because each counts people who meet a definition by a particular method: a global review of 99 estimates reported a median of about 1 in 100 and a range from 1.1 to 43.6 per 1,000, and traced the spread to method and to awareness, services and help-seeking.⁴⁵ A meta-analysis of seven African studies pooled a clinically confirmed prevalence of about 1% with wide variation between the studies,⁴⁶ and a Kenyan cohort that screened 11,223 children confirmed neurodevelopmental disorders in 522 of them, with autism at about 1.6%.⁴⁷ The Global Burden of Disease 2023 estimate for autism is modeled.⁴⁸ Knowledge and stigma vary widely: across 6,830 people in 13 countries, autism knowledge scores differed by 17 points between the highest and lowest country.⁴⁹ The one program that travels is the World Health Organization's Caregiver Skills Training, which trains non-specialist providers to coach caregivers; a meta-analysis of 75 trials of caregiver training found moderate effects on child development and behavior and larger effects on caregiver skills and wellbeing,⁵⁰ while a pilot of the WHO program itself found large gains in parent skills but no significant change in the child's autism severity.⁵¹

Money and class appear in who is counted as autistic. A US survey of 205,480 children aged 3 to 17 (2018 to 2023) found that those in households below the federal poverty level had higher adjusted odds of a parent-reported autism diagnosis than those at four times the poverty level, and that lower parental education, food insecurity and unsafe neighborhoods went with higher odds as well.⁵² The direction is not fixed. Older US surveillance data, for 2002 to 2010, recorded autism more often among 8-year-olds in higher-socioeconomic-status areas, which its authors suggested might reflect unequal access to diagnosis, though they could not confirm that lower-status children were underidentified.⁵³ In the 2020 surveillance year, prevalence was slightly higher in more socially vulnerable neighborhoods overall (prevalence ratio 1.06, confidence interval 1.00 to 1.13), a pattern that held for White children but reversed for Black, Asian and Hispanic children.⁵⁴ All of these count recorded or parent-reported diagnoses and cannot separate who is autistic from who is counted.

Diagnosis and support cost families money. In an exploratory US study of adolescents, material hardship was more common than income poverty among families of autistic teenagers (46.4% against 29.8%), and those families were more likely to report income drops and bankruptcy than other families.⁵⁵ In a preliminary study at one tertiary clinic in urban India, about 71% of families seeking autism services spent more than a tenth of their monthly income on healthcare, which the authors call catastrophic; across all costs, the largest expenses were diagnosis, early intervention, schooling and center-based rehabilitation.⁵⁶ A pilot survey in Ontario, Canada, found that financial pressure often began at the diagnostic process and continued as out-of-pocket medical and therapy costs, with caregivers describing public support as inadequate.⁵⁷

Access to assessment varies with geography and with what a family can afford. In a Sydney service evaluation, the slowest group of children waited a median of about 850 days for a developmental assessment, against about 250 days for the rest; that group was more often from culturally and linguistically diverse backgrounds and more often needed an interpreter, and caregivers said the cost of private assessment kept them on the public wait.⁵⁸ A qualitative study across six regions of rural Northern Ontario found that families with fewer resources faced delays at the first stage of identification, with staff shortages and distance adding to the wait.⁵⁹ In the United States, a rapid review of 60 studies using Medicaid claims (the programs that often fund autism services for lower-income families) found the disparities literature concentrated on school-aged children and young adults (ages 6 to 21) and on individual-level factors, with family, provider and sociocultural influences on how services reach families rarely examined.⁶⁰

Being open about being autistic is easier for some people than others. A scoping review of 14 qualitative studies of autism at work found that stigma, worries about disclosure and the pressure to fit in shaped autistic employees' experiences and their use of masking.⁶¹ No study found as of October 2026 has measured whether income or job security changes how openly autistic adults can be, so the link between money and openness is not yet tested.