In one sentence
This systematic review of 26 studies finds that autistic girls and women are often underdiagnosed or diagnosed later than boys because standard tools and criteria were developed mainly around male presentations of autism.
What the researchers did
The author searched four databases in December 2025 and included 26 studies published between 2015 and 2025 on gender differences in autism diagnosis. The studies used mixed methods: observational designs, psychometric work on how well tests work, qualitative interviews, and some technology-based approaches such as eye-tracking. Study quality was generally rated good to very good, with mostly low risk of bias.
What they found
- In the Spanish school context discussed in the review, the diagnosed boy-to-girl ratio reaches about 4.6 to 1, which suggests many girls may still be missed.
- Girls and women often show fewer obvious externalizing behaviors (such as acting out) and more subtle social-communication differences.
- Social camouflaging — hiding autistic traits to fit in — is common in females and can delay or block diagnosis.
- Widely used instruments (for example ADOS-2 and SCQ) may be less sensitive to female autism profiles.
- Co-occurring conditions such as ADHD can further complicate and delay recognition in girls.
- Later diagnosis is linked with poorer mental-health outcomes in adolescence and adulthood.
What this means for families and therapists
- Autistic girls may look different from the classic male stereotype: quieter, more socially motivated, or skilled at masking difficulties.
- Assessment works better with several information sources — parent report, school observations, and self-report when possible — plus attention to camouflaging.
- If a girl struggles socially or emotionally but does not meet traditional autism criteria, a clinician experienced with female autism profiles may still be worth seeking.
Limitations and what we don't know yet
- Only English- and Spanish-language studies were included.
- Most studies were observational, so cause-and-effect conclusions are limited.
- Study quality and sample sizes varied.
- The review focused on Level 1 autism (requiring support) and may not generalize to all support needs.
- More work is needed on how race, ethnicity, and socioeconomic status interact with gender bias in diagnosis.
This is a plain-language summary of Systematic Review of Autism in Women: Invisibilised Profiles and Gender Bias by Muiño Tato A., Revista de Neurología (2026). Source license: CC-BY-4.0. It is not medical advice — talk to a qualified clinician before changing therapy.