Autism refers to a wide spectrum of neurodevelopmental conditions that are linked with several differences in social communication; repetitive and restricted behaviors, interests, activities, and sensory differences. Current estimates suggest about 1 in 36 children are diagnosed with one of these conditions at an early age, though the social recognition and development of the diagnosis for conditions under the umbrella of autism continue to improve over time and hence vary across studies.
In this study, the term “sex” was used by researchers to refer to the sex assigned to a person at birth and to understand sex relative to chromosomal, genital, or hormonal differences. The word “gender” refers to a person’s sense of self and might or might not be identical to the sex assigned at birth. The expression “transgender/gender diverse (TGD)” was used by researchers to explain the experience of any person whose gender identity does not match the sex at birth, including those such as transgender, non-binary, genderfluid, agender, bigender, genderqueer, two-spirit, and so forth. Such individuals who have their gender identity in accordance with the sex assigned at birth are known to be “cisgender”.
The current analyses are actually based on the larger study data that was collected at the Autism Research Centre (ARC) with the University of Cambridge. It set out to compare chronic disease rates and quality of care between autistic adults and their non-autistic counterparts. Data were collected through an anonymized web-based self-report survey using Qualtrics survey technology. Survey items were informed by data from the National Health Service (NHS) and the National Institute for Health and Care Excellence (NICE) in the United Kingdom and the World Health Organization (WHO). Data was collected between July 2019 and January 2021.
Of the individuals who participated in the survey, 4158 were included, 33.6% of the sample (N=1396) were excluded after not providing valid consent, demographic information was incomplete, or age data was unavailable. Individuals with non-confirmed autism status (N=26) were also excluded following these exclusions. As all site visitors were anonymous, the survey had to eliminate repetitions using an algorithm (N=112): For the 12 profile characteristics matched in the respondent pool (autism diagnosis [yes/no], type of diagnosing practitioner, specific autism diagnosis, year of the autism diagnosis, family members of autistic [yes/no], age, country of the residence, sex assigned at birth, current gender identity, ethnicity, highest educational level, and AQ – 10 score).
To further reduce noise due to a priori power calculation, the non-autistic respondents with TGD will be excluded (n=57) from this analysis, as it is a small subgroup. Finally, due to a possible preponderance of categories, we also treat sex assigned at birth as a covariate when statistical modeling.
Thus, four intersex individuals were excluded due to perfect separation issues related to this covariate. The final sample constituted n=2563, with n=1295 being cisgender non-autistic, n=1094 being cisgender autistic, and n=174 being TGD autistic.
The majority of the sample comprised white (82%), UK residents (55%), assigned female at birth (63%), identified as female (58%), and university degree holders (66%). Here, similar biases were observed in each of the groups separately (with the exception of the TGD group, whose modal gender identity was ‘other’). The mean age was 38.9 years (SD=16.1) for the cisgender autistic group, 35.6 years (SD=13.9) for the TGD autistic group, and 42.1 years (SD=14.4) for the cisgender non-autistic group.
Overall, significant differences were observed across all demographic characteristics between the three groups, as well as between the autistic groups, except for ethnicity, for which there was no significant difference between TGD autistic and cisgender autistic adults.
These data were not originally collected to further understanding of transgender/gender diverse people. Other limitations to the contribution of this study to generalizability are recruitment strategies, convenience sampling, and a survey based on self-reporting. Due to our bias towards indigenous white, UK residents with relatively high educational levels among those designated female at birth and who currently identify as female, the findings may have limited applicability to individuals in other demographic categories. Finally, this study does not include information on the experiences of transgender/gender-diverse people outside the autistic spectrum.
Regardless of gender identity, autistic individuals reported poorer healthcare experiences and high health risks, particularly concerning mental health. Transgender/gender-diverse and autistic individuals face further challenges in accessing care compared to both cisgender non-autistic and cisgender autistic individuals.
While the present study provides important information concerning the health and healthcare experiences of transgender/gender-diverse autistic people, there is a pressing need for more research in this area; in particular, the research should use generalizable samples to corroborate the opinions of transgender/gender-diverse autistic people with their cisgender autistic and non-autistic counterparts, and with transgender/gender-diverse non-autistic individuals.
Reference: Green K, Weir E, Wright L, Allison C, Baron-Cohen S. Autistic and transgender/gender diverse people’s experiences of health and healthcare. Mol Autism. 2025;16(4). doi:10.1186/s13229-024-00634-0
