Graphic of female profile with her brain highlighted in orange

Solving the Problem of Late Autism Diagnosis

With a $12 million grant, autism researchers at the University of Virginia aim to illuminate the causes of late diagnosis, particularly for women and gender-diverse people – and to build a better diagnostic tool.

Laura Hoxworth

Growing up with two military parents, Jess Walters moved around a lot – and at each new school, struggled to connect with peers. 

“I would engage, but it was more like a mirror. Anybody who paid attention could probably pick up on the fact that I was really just repeating the behavior of people around me.” 

Deep down, Walters said they always felt different. “There's this feeling of, ‘I can't necessarily put a finger on why, but I'm different and weird, and people don't accept me as I am.’ That feeling takes a toll on you. You're not allowed to be yourself, because yourself is alien to this world.” 

Now, Walters understands that masking, or mirroring others’ behaviors in social situations, is a common trait among people with autism spectrum disorder (ASD). It is especially common among people who are diagnosed later in life – like Walters, who received an ASD diagnosis last year at 30 years old. 

The problem of late diagnoses 

The average age of autism diagnosis in the U.S. is between ages four and six.  

Diagnosis is the first step to accessing support – and generally, the earlier the intervention, the better the outcome. “We're constantly trying to move that age of diagnosis as early as possible,” said Micah Mazurek, a professor in the School of Education and Human Development and director of UVA’s Supporting Transformative Autism Research (STAR) initiative.  

For women, gender-diverse and LGBTQ people, however, the average age of diagnosis is much later. 

With a new $12 million grant from the National Institutes of Health, a team of researchers from across UVA will focus on illuminating why people like Walters, who identifies as non-binary, wait so long for an accurate diagnosis – and on building a better diagnostic tool. 

In recent years, UVA autism researchers have made great strides in understanding how sex, gender and autism intersect. In 2021, UVA researchers discovered a significant difference in the genes that underpin the condition in girls and boys. They also identified specific ways the brains of girls with ASD respond differently to social cues such as facial expressions and gestures.   

But early autism research focused mainly on boys, and diagnostic processes have not caught up with newer findings. “If we're only looking at things that are typically present in boys, we're going to miss autism presentation that may look completely different,” Mazurek said. 

The problem is not just about accessing timely support. Late diagnosis, defined in the grant as 12 years or older, is also linked to greater mental health difficulties. One recent study found that people diagnosed with autism in adulthood are nearly three times as likely as their childhood-diagnosed counterparts to report having psychiatric conditions. 

For Walters, as they grew up and navigated high school and college, they excelled in academics, graduating from UVA with a degree in psychology. But they still struggled with social situations and simpler everyday tasks. “I would get very frustrated with myself, because it's like, I can write an A-plus 10-page essay in two hours. But I can't do the dishes? 

“I internalize a lot,” they said. “There was a lot of, ‘I'm not good enough.’ Looking back, I can see that I was actually really unsupported for my needs – and unaware of my needs.” 

Envisioning a better diagnostic tool 

Like many late-diagnosed autistic people, Walters’s route to diagnosis has been long and winding. 

Following a misdiagnosis in 2016, Walters had to drop out of graduate school. “That misdiagnosis cost me several years of being medicated with terrible side effects and major depressive episode,” they said. 

But, by following their intense interests in language and psychology, Walters got involved in autism research at UVA. In 2020, they audited a graduate course where they learned about the prevalence late diagnoses among female-identifying and LGBTQ people. 

Walters's existing knowledge about autism and local support systems helped them easily access an assessment. But for many others, that process can be confusing and difficult to navigate. 

With this grant, researchers hope to make that process easier. 

The Autism Center of Excellence Network grant, funded by the National Institute of Health, is one of the top autism research grants in the country. This latest five-year award builds on a decade of cross-institutional, interdisciplinary autism research led by Kevin Pelphrey, a leading autism expert affiliated with the School of Medicine, the School of Education and Human Development and the Brain Institute.  

Pelphrey’s team has worked on the initial development of a self-assessment screening tool for adolescents and adults. With the new grant, they will be able to test, refine, and validate it with a much larger and more diverse group of people. 

Researchers, at the lead site of UVA as well as several other institutions, will combine high-tech methods – like brain imaging, machine learning, and remote data collection using wearable sensors – with medium-tech (questionnaires delivered by web-based and smartphone apps) and low-tech (interviews and behavioral assessments). 

To accomplish this, data scientists and clinicians will work hand-in-hand, allowing the team to pair rich data with a deep understanding of what it represents. Just at UVA, experts from the Brain Institute, STAR, the School of Medicine, iTHRIV, and Data Science are all involved – each bringing a valuable perspective to the project.  

“The interdisciplinary collaboration is essential,” Pelphrey said. On the medical side, experts will examine brain function and biological markers of autism. The STAR team offers extensive experience with behavioral measures and interviews in applied clinical settings. Data analysts will figure out how to bring it all together. 

“Ultimately, the tool will be a concise series of questions that an individual or caregiver can answer via a smartphone, web page, or pencil and paper,” Pelphrey said. “This tool will provide a diagnostic access point for adolescents and adults at risk for late diagnosis, who are routinely missed by current methods.” 

Community-driven research 

At the center of it all is the most important stakeholder: people with autism. "Our priorities are driven by our autistic partners and the autism community,” Mazurek said. 

“Embracing a community-engaged research approach, a sex, gender, and ethnoracially diverse stakeholder team of clinicians, self-advocates, autistic people, and parents, all with professional and/or lived experience with late-diagnosed ASD, will collaborate with us,” Pelphrey said.  

“This precision-medicine approach will accelerate ASD diagnosis in the community, allowing for appropriate support and improved outcomes." 

In addition to developing a screening tool, the project aims to better capture the lived experience of autism, including the many benefits and strengths that go along with autism. This kind of collaboration will also help researchers eventually share what they discover with the entire autism community. 

“We’ve learned from autistic adults that not receiving a diagnosis until adulthood can be very difficult – and can have negative consequences for their mental health and well-being,” Mazurek said. “Figuring out a way to make that better and to use their experiences to inform the development of better tools could really help with quality of life for a lot of folks.” 

Diagnosis, of course, is just the beginning of an autism journey. Walters, who is now an artist and disability advocate living in Charlottesville, said many things need to change for autistic people to feel safe – including more nuanced representation in media and greater understanding and acceptance of neurodivergence throughout the general public.  

But diagnosis is often the first step to unlocking something invaluable: a community of people who understand. 

“In many ways, my autism diagnosis has been freeing,” Walters said. “It makes my life make sense. It gives me validity. It also gives me this group of other humans that don't find me alien, which is a tremendous joy.” 

News Information

Research Center or Department

  • Supporting Transformative Autism Research