News The latest developments in autism research.
Profiles Portraits of scientists who are making a mark on autism research.
Toolbox Emerging tools and techniques that may advance autism research.
Spotted A roundup of autism papers and media mentions you may have missed.
Opinion Conversations on the science of autism research.
Viewpoint Expert opinions on trends and controversies in autism research.
Columnists Dispatches from experts on various facets of autism.
Crosstalk Debates and conversations about timely topics in autism.
Reviews Exploring the intersection of autism and the arts.
Q&A Conversations with experts about noteworthy topics in autism.
Deep Dive In-depth analysis of important topics in autism.
Special Reports Curated collections of articles on special topics in autism.
Webinars Presentations by leading experts on their latest research.

Diagnostic tests for women with autism fall short

by  /  11 March 2016
Play, repeat: Play that involves typical girl toys, such as dolls, may mask repetitive behaviors in girls with autism. ©

Women with autism show fewer repetitive behaviors than men with the condition on a standard diagnostic test. This difference may lead to a ‘partial’ diagnosis for many women, qualifying them for only limited services, according to the largest study of adults with autism to date.

Clinicians may diagnose just 1 girl with autism for every 10 boys in children of average intelligence, and her diagnosis may be years overdue, according to some reports. Most diagnostic assessments are based on research in boys with autism and may be biased toward behaviors typically seen in these boys. So it is unclear whether fewer girls have autism, or whether the features of autism in girls and women are simply distinct and harder to recognize.

Either way, girls who go undiagnosed in childhood may find it difficult to obtain an autism diagnosis later, suggest the new findings, which appeared 22 January in Autism1.

A primary reason for this difficulty may be that women don’t show repetitive behaviors and restricted interests — or don’t show them in a way that is obvious to caregivers or clinicians.

“The most striking thing is the difference in repetitive behaviors between males and females,” says Grainne McAlonan, clinical reader in translational neurodevelopment at Kings College London and a senior researcher on the study. “That may have an impact on the way we recognize autism in females, and consequently whether those females get access to services.”

The study points to the possibility that women and men with autism present differently and may benefit from diagnostic tools that take gender differences into account.

Gender gap:

Researchers evaluated 935 men and 309 women, all 18 to 75 years old, who were referred to an autism specialty clinic by their primary physician. They used a standard assessment that relies on parent or caregiver reports, called the Autism Diagnostic Interview-Revised (ADI-R), to document the behavior of 630 adults when they were children. They used a standard clinical measure, the Autism Diagnostic Observation Schedule (ADOS), to directly evaluate the social and repetitive behaviors of another 408 adults for whom parent reports were not available. Researchers administered both tests to an additional 206 participants.

Based on the test results and a psychiatric exam, 72 percent of the men and 66 percent of the women received an autism diagnosis.

Researchers then narrowed their analysis to adults with autism who have at least average intelligence — a particularly challenging group to diagnose. They examined gender differences in autism signs on each of the three diagnostic domains in the World Health Organization’s diagnostic bible, the International Classification of Diseases (ICD)-10: social interaction, communication, and repetitive behaviors or restricted interests.

Most women they diagnosed had dramatic signs in one domain, usually either communication or social interaction, and moderate or mild signs in the other two. By contrast, men typically showed moderate signs across all three domains.

The study highlights potentially important differences in the profiles of men and women with autism, says Thomas Frazier, director of the Center for Autism at the Cleveland Clinic in Ohio, who was not involved in the study. Men with autism tend to have features that span several diagnostic criteria, whereas women often show severe impairment in only one area, he says.

Muddled memory:

About a quarter of the men and a third of the women with autism qualified for only a ‘partial’ diagnosis under the ICD-10 because they showed signs in only two of the three domains.

But these sex differences were relatively minor, given the sample of more than 1,200 adults, says Sander Begeer, assistant professor of developmental psychology at the VU University Amsterdam in the Netherlands. “Given the adequate power of the study, one would expect sex differences to show up here if there were any.”

The women and men differed most starkly in whether they displayed repetitive behaviors and restricted interests, and even there, the difference was only apparent on the ADI-R. This means that parents and caregivers tended to recall far fewer repetitive behaviors during childhood for the women than for the men.

The findings suggest that although women may display repetitive behaviors as adults, researchers who rely on the ADI-R to diagnose might not flag them, resulting in a higher rate of ‘partial’ autism diagnoses in women.

Still, the ADI-R is perhaps the best way to identify these behaviors, McAlonan says. “The repetitive behaviors criteria really must come from the ADI-R because the ADOS is not very good at identifying these behaviors [in adults].”

Doctors in the U.S. who use the criteria in the Diagnostic and Statistical Manual of Mental Disorders may diagnose people lacking repetitive behaviors with social communication disorder rather than autism.

Subtle signs:

“These women had severe social and communication deficits, but they were not getting a full autism diagnosis because they didn’t meet criteria for repetitive behaviors,” says study investigator C. Ellie Wilson, who is now a postdoctoral researcher at the University of Seville in Spain. “This is potentially problematic when you think about clinical support and the resources available to them.”

Girls who appear to lack repetitive behaviors may simply have subtler ones than boys do. Because playing with dolls is typical for girls, someone might not notice, for example, that the girl is spending more time moving the dolls around than imagining a conversation between them, Wilson says.

A study published 2 March in Autism is consistent with previous reports suggesting that girls with autism act out simpler scenarios with their dolls than do typically developing girls. Play complexity is roughly similar in girls and boys with autism, the researchers reported2.

Neither the ADOS nor the ADI-R measure the severity of traits, which may be key to spotting sex differences, says Begeer. Nor did the researchers look for differences in types of repetitive behavior or sensory sensitivities. In studies of adults, such details can be hard to get because the ADI-R relies on parent or caregiver recollections from decades past.

“We really need to do prospective studies to try and identify what sex differences there might be,” Wilson says.

  1. Wilson C.E. et al. Autism Epub ahead of print (2016) Pubmed
  2. Harrop C. et al. Autism Epub ahead of print (2016) Pubmed
  • Planet Autism

    They need to start realising that repetitive behaviours may apply to many things, including how something is said or required to be said by others in an autistic female; which toys (or beauty products in an adult) and how many of one type she collects, her special interests and how what appear to be OCD behaviours are in fact autistic repetitive behaviours. They are looking at it all wrong.

    The gender ratio in autistics is equal. Eventually everybody will realise this. In the meantime, females get failed and misdiagnosed.


Log in to your Spectrum Wiki account

Email Address:



Request your Spectrum Wiki account

Spectrum Wiki is a community of researchers affiliated with an academic or research institutions. To be considered for participation, please fill out this form and a member of our team will respond to your request.


Email Address:

Title and Lab:

Area of Expertise: