Authors
Ashley Chay, Tobias Schuwerk
Published in
Autism : the international journal of research and practice. Pages 13623613261499409. Nov 19, 2026. Epub Nov 19, 2026.
Abstract
Autism is diagnosed at higher rates in males, with female-to-male ratios commonly cited as 1:4, though recent estimates suggest ratios closer to 1:3 or even 1:2. Despite growing calls for improved female representation in autism research, the extent and predictors of sex-based inclusion disparities remain poorly characterized at the field level. This preregistered systematic review and meta-analysis used meta-regression to identify study-level predictors of variability in female representation, drawing on a small, randomly selected sample of the retrievable literature: of 17,244 deduplicated records, 1,330 were randomly sampled for screening, 888 were coded for eligibility, and 250 studies (N = 81,333 participants), published between 2000 and 2025, were eligible for analysis. Overall, females comprised 19.3% (95% confidence interval [CI] [18.8%, 19.9%]) of participants-closely matching the 1:4 ratio but falling short of 1:3. Critically, studies using community-based diagnostic confirmation included significantly more females than those relying on Autism Diagnostic Observation Schedule (ADOS) cutoffs. Publication year was significant, with model-predicted female representation increasing from 12.5% in 2000 to 19.4% in 2025 among studies using ADOS cutoffs. Study type was not significant. These findings highlight diagnostic confirmation method as a key driver of sex representation in autism research, reinforcing the need to move beyond single-instrument cutoffs to ensure equitable inclusion.Lay AbstractIf autism research studies mostly boys and men, then what we know about autism may not fit girls and women. We looked at 250 autism studies from the last 25 years. Together, they included more than 81,000 people. We asked two things. First, how many autistic girls and women took part in this research? Second, what choices affected that number? We found that only about one in five participants were girls or women. That roughly matches older estimates of how common autism is in girls and women, but newer evidence suggests the real number is higher, closer to one in four. So girls and women still appear to be underrepresented. There is some good news: representation has improved over time. We found that studies from 2025 included far more girls and women than those from 2000. Our most important finding was about the Autism Diagnostic Observation Schedule (ADOS). This is one common tool used to diagnose autism. Studies that required the ADOS included fewer girls and women. Studies that also accepted other tools to diagnose autism, such as a doctor's opinion, included more. This matters because the ADOS was first designed and tested mostly on boys and men, so it might miss autism in girls and women. In short, who takes part in autism research depends on the choices researchers make. Better choices could include more autistic girls and women. This could lead to better understanding and support for them.
PMID:
43053743
Bibliographic data and abstract were imported from PubMed on 19 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 0
- Comments 0