Boys are four times more likely than girls to be diagnosed with autism spectrum disorder (ASD), yet a growing body of research shows that the condition is more common in girls than previously thought, strongly suggesting that new methods are required to diagnose the disorder at younger ages.

A new study from Children’s Hospital of Philadelphia (CHOP) examined differences in the way girls and boys on the autism spectrum used certain types of words during storytelling. This study found that autistic girls used significantly more “cognitive process” words such as “think” and “know” than autistic boys, despite comparable autism symptom severity. The results were recently published in the journal Molecular Autism.

The authors suggest that identifying differences like these opens the door to making sure girls with ASD receive the diagnosis and support they need to achieve the best possible quality of life.

“In order to place these findings in context, it’s important to understand that because girls tend to exhibit different traits than autistic boys do, they are often incorrectly diagnosed or missed entirely by standard diagnostic tools. That discrepancy also skews the research literature,” explained lead author Julia Parish-Morris, PhD, a scientist in the Center for Autism Research and faculty member in the Departments of Child Psychiatry and Biomedical & Health Informatics at CHOP. “Autism studies have historically included three to six times as many males as females. This means that we don’t yet know enough about gender differences in autism, and so we miss girls whose traits differ from those of boys.”

A misdiagnosis means many girls do not receive early intervention and that standard interventions may not be appropriate for meeting girls’ unique needs. Many autistic women are not diagnosed until they are adults and report significant social challenges and a profound sense of being different from their typically developing peers.

“Autism is a social condition diagnosed using observable behavior, so we wanted to study an observable skill that relates to social ability,” Parish-Morris said. “We chose storytelling because it involves much more than grammar and vocabulary; it relies on a sense of social appropriateness and sheds light on what speakers decide is important to convey.”

The researchers focused on how participants used nouns (object words) compared with cognitive process words, because prior studies found that reduced use of cognitive process words predicted social challenges in ASD. Although past studies used primarily male samples, the results were assumed to generalize to girls, and prior ASD studies never included enough girls to test whether sex differences that exist among typically developing individuals persist in those with ASD.

Parish-Morris and her co-authors studied 102 verbally fluent school-aged children who either had a diagnosis of ASD (21 girls and 41 boys) or were typically developing (19 girls and 21 boys), and were matched on age, IQ and maternal education. The children viewed a sequence of pictures involving a fisherman, a cat and a bird, and told a story based on what they saw.

Results revealed that autistic girls used significantly more cognitive process words than autistic boys did, even when they had similar levels of autism severity. Girls with ASD and typical girls used comparable numbers of cognitive process words. Interestingly, autistic boys and girls both used more nouns than typically developing children, demonstrating object-focused storytelling. Thus, autistic girls showed a unique narrative profile that overlapped with typical girls and boys as well as with autistic boys.

“Through storytelling, we were able to identify key similarities and differences in the language patterns of autistic girls and boys,” Parish-Morris said. “These findings suggest that sex-informed screening and diagnostic methods may help us identify autism in verbal girls at an earlier age, which should spur efforts to develop appropriate, personalized early interventions resulting in improved support for girls and women with ASD.”

In this webinar, Tychele Turner, Ph.D., focuses on the genetic aspect of sex bias in autism. Particularly, she highlights her recent work examining de novo variants, separately in males and females, in approximately 30,000 families.

Boys are four times more likely than girls to be diagnosed with autism spectrum disorder (ASD), yet a growing body of research shows that the condition is more common in girls than previously thought, strongly suggesting that new methods are required to diagnose the disorder at younger ages.

A new study from Children’s Hospital of Philadelphia (CHOP) examined differences in the way girls and boys on the autism spectrum used certain types of words during storytelling. This study found that autistic girls used significantly more “cognitive process” words such as “think” and “know” than autistic boys, despite comparable autism symptom severity. The results were recently published in the journal Molecular Autism.

The authors suggest that identifying differences like these opens the door to making sure girls with ASD receive the diagnosis and support they need to achieve the best possible quality of life.

“In order to place these findings in context, it’s important to understand that because girls tend to exhibit different traits than autistic boys do, they are often incorrectly diagnosed or missed entirely by standard diagnostic tools. That discrepancy also skews the research literature,” explained lead author Julia Parish-Morris, PhD, a scientist in the Center for Autism Research and faculty member in the Departments of Child Psychiatry and Biomedical & Health Informatics at CHOP. “Autism studies have historically included three to six times as many males as females. This means that we don’t yet know enough about gender differences in autism, and so we miss girls whose traits differ from those of boys.”

A misdiagnosis means many girls do not receive early intervention and that standard interventions may not be appropriate for meeting girls’ unique needs. Many autistic women are not diagnosed until they are adults and report significant social challenges and a profound sense of being different from their typically developing peers.

“Autism is a social condition diagnosed using observable behavior, so we wanted to study an observable skill that relates to social ability,” Parish-Morris said. “We chose storytelling because it involves much more than grammar and vocabulary; it relies on a sense of social appropriateness and sheds light on what speakers decide is important to convey.”

The researchers focused on how participants used nouns (object words) compared with cognitive process words, because prior studies found that reduced use of cognitive process words predicted social challenges in ASD. Although past studies used primarily male samples, the results were assumed to generalize to girls, and prior ASD studies never included enough girls to test whether sex differences that exist among typically developing individuals persist in those with ASD.

Parish-Morris and her co-authors studied 102 verbally fluent school-aged children who either had a diagnosis of ASD (21 girls and 41 boys) or were typically developing (19 girls and 21 boys), and were matched on age, IQ and maternal education. The children viewed a sequence of pictures involving a fisherman, a cat and a bird, and told a story based on what they saw.

Results revealed that autistic girls used significantly more cognitive process words than autistic boys did, even when they had similar levels of autism severity. Girls with ASD and typical girls used comparable numbers of cognitive process words. Interestingly, autistic boys and girls both used more nouns than typically developing children, demonstrating object-focused storytelling. Thus, autistic girls showed a unique narrative profile that overlapped with typical girls and boys as well as with autistic boys.

“Through storytelling, we were able to identify key similarities and differences in the language patterns of autistic girls and boys,” Parish-Morris said. “These findings suggest that sex-informed screening and diagnostic methods may help us identify autism in verbal girls at an earlier age, which should spur efforts to develop appropriate, personalized early interventions resulting in improved support for girls and women with ASD.”

Having one child with autism is a well-known risk factor for having another one with the same disorder, but whether and how a sibling’s gender influences this risk has remained largely unknown.

Now new research led by scientists at Harvard Medical School has for the first time successfully quantified the likelihood that a family who has one child with autism would have another one with the same disorder based on the siblings’ gender.

Overall, the results, published Sept. 25 in JAMA Pediatrics, reveal that having an older female child diagnosed with autism spelled elevated risk for younger siblings and that the risk was highest among younger male siblings. They also affirm past research findings that having one child with autism or an autism  portends higher risk for subsequent children, that the disorder is somewhat rare–slightly more than 1.2 percent of children in the study were affected– and that boys have a notably higher overall risk than girls.

The findings can arm physicians and genetic counselors with information useful in counseling families and clarifying the risk for younger siblings in families who already have one child with autism.

“Our results give us a fair degree of confidence to gauge the risk of autism recurrence in families affected by it based on a child’s gender,” said study first author Nathan Palmer, instructor in biomedical informatics at Harvard Medical School. “It is important to be able to provide worried parents who have one child with the condition some sense of what they can expect with their next child. That information is critical given how much better we’ve become at screening for the disease earlier and earlier in life.”

Such knowledge, the researchers added, could be particularly important in light of physicians’ growing ability to detect autism’s manifestations early in a child’s life and intervene promptly.

“This study is a powerful example of how big data can illuminate patterns and give us insights that allow us to empower parents and pediatricians to implement anticipatory and far more precise medicine,” said study senior author Isaac Kohane, head of the Department of Biomedical Informatics at Harvard Medical School.

The newly published results stem from the largest study of its kind. Researchers analyzed de-identified health insurance records of more than 1.5 million U.S. families with two children between the ages of 4 and 18, tracking patterns of recurrence among siblings over a year or longer. Of the more than 3.1 million children in the study, some 39,000, or about 1.2 percent–2 percent of boys and 0.5 percent of girls–received a diagnosis of autism .

The results confirm previous research showing that, overall, boys have a higher risk of autism and related conditions than girls.

The results, however, also reveal a curious pattern of recurrence based on gender: Siblings born after a female child with autism or a related disorder had a higher risk than siblings born after a male child with autism. Male children were, overall, more susceptible to autism than females. In other words, boys with older female siblings with autism had the highest risk for autism themselves, while female siblings with older brothers with autism had the lowest risk.

For every 100 boys with an older female sibling with autism, 17 received a diagnosis of autism or a related disorder. Male children with older male siblings with autism had a 13 percent risk of an autism diagnosis, followed by younger female siblings with older female siblings with autusm (7.6 percent). The lowest risk–4 percent–was observed among younger female siblings who had an older brother with autism.

The investigators caution that families should keep the risk in perspective because autism and related disorders remain relatively rare, affecting roughly 1 percent of the general population.

“Even for the group at highest risk–males with an older female sibling with autism–the odds are still about five to one that the child will be unaffected,” Palmer said. “What we have provided here is context for families who already have children with autism or another similar disorder and need a clearer perspective on recurrence risk.”

The results, the researchers said, underscore the notion that autism and related disorders likely arise from the complex interplay between genes and environment and, for reasons yet to be understood, these conditions disproportionately affect more males than females even within families. The stark gender variance, however, hints at a possible role of inherent biological sex differences that may precipitate the development of such disorders under the right environmental conditions, the research team said.

Autism-spectrum disorders are neurodevelopmental conditions that typically emerge in the first few years. They are marked by a range of brain problems, impaired social interactions and compromised communication skills. The Centers for Disease Control and Prevention estimate that autism spectrum disorders affect 1 in 68 children in the United States, with males having four times greater risk than females–an observation also borne out in the new study.

Yet exactly what portion of these diagnoses are strictly rooted in genetic mutation and how many are influenced by environmental factors has long mystified scientists. While some forms of autism arise from a single genetic mutation, most cases appear to be the result of a complex interplay between genes and environment.