Toolbox Emerging tools and techniques that may advance autism research.
News / Toolbox

New database holds clues to severe end of autism spectrum

by  /  2 December 2015
Gender balance: More than one-quarter of children in the Autism Inpatient Collection are girls.

© / JynMeyerDesign

A new database provides genetic and behavioral information about children with autism who have been admitted to hospital psychiatric units. The database, described 10 November in Molecular Autism, gives researchers an unprecedented glimpse of the most severe cases of autism1.

Many children with severe autism have intellectual disability and little or no language ability, making it difficult for researchers to distinguish the children’s autism symptoms from their other difficulties. They are also prone to aggressive outbursts and self-injury. As a result, they are underrepresented in existing data repositories for autism research.

To address this gap, researchers from six U.S. hospitals with inpatient psychiatric units that specialize in the assessment and treatment of autism collaborated to create the Autism Inpatient Collection. In 2014, they began enrolling children aged 4 to 20 years who are suspected of having autism and who have at least one English-speaking parent or guardian.

Trained clinicians diagnose the children using the Autism Diagnostic Observation Schedule. The researchers then collect samples of blood or saliva from each child and his or her parents for genetic analyses. They also ask parents to complete a set of questionnaires that probe their child’s intellectual and cognitive abilities, language skills, challenging behaviors and psychiatric symptoms.

The new study describes the first 147 children with autism in the database. More than one-quarter of them are girls, who are less likely than boys to have autism but are thought to exhibit more severe symptoms when affected. More than half of the children speak few or no words, and 43 percent have intellectual disability.

Children in the database who are later found not to have autism will serve as a comparison group in future studies, the researchers say. They plan to expand the collection to include more than 1,600 people with autism and their caregivers.

The researchers assign each child and caregiver a Global Unique Identifier that allows them to keep the genetic and behavioral data anonymous and link it with information about each participant that may exist in other repositories.

As the collection grows, the data it contains could reveal whether certain genetic variants track with intellectual disability or language impairment, for example, the researchers say. It may also highlight the challenges of those with intellectual disability, language impairment and self-injurious behavior.

  1. Siegel M. et al. Mol. Autism 6, 61 (2015) PubMed

7 responses to “New database holds clues to severe end of autism spectrum”

  1. Ethyl says:

    According to the CDC’s latest poll of parents, only 16.7% of kids with autism are considered Intellectually Disabled. Severe Autism is disappearing.

    • Quinn Harbin says:

      More likely that expanded diagnostic criteria has given us many more cases of “mild” autism. Research that has looked at incidence rates of the most severe type- first diagnosed by Kanner- have found no change over time.

  2. EJ says:

    “he new study describes the first 147 children with autism in the database. More than one-quarter of them are girls, who are less likely than boys to have autism ”

    So fed-up of reading this. It is factually wrong. Even articles on your own website question this assumption.

    Tony Attwood has said on record that the gender ratio is 50:50 and I have long believed this to be the case myself.

    How can such bald statements be made, when clearly using diagnostic tools researched only on males and females presenting differently, 10 times less females referred for ASC assessment in the first place and other bias issues?!

    • Nicholette Zeliadt says:

      Thanks for your comment. Researchers are certainly investigating the male-to-female ratio in autism diagnoses, and many suspect that the ratio may differ depending on whether you’re talking about people at the severest end of the spectrum or the opposite end. As we reported in October (, researchers now think that the gender bias for those who have high intelligence quotients is not as big as previously thought, maybe because girls at this end of the spectrum are better than boys are at hiding their symptoms or because, as you point out, male-based tests aren’t as good at picking up autism in girls. But few suspect that the overall ratio will end up at 50:50, even if biases in diagnosis are resolved.

      • EJ says:

        You’d be surprised. There are likely thousands of undiagnosed females out there. When you take into account all the elements contributing to the current gender bias, it will very much level the playing field. Just the aspect alone of the different presentation of females would probably do most of it. Consider that 10 times less females are referred for assessment than males! Consider the fact that females are often misdiagnosed with other conditions such as borderline personality disorder, anorexia (a symptom of the underlying autism) and generalised anxiety disorder (another symptom of the autism).

        I am a female with autism, with 2 autistic daughters. I am not so “rare” as your comments would lead people to believe. I also know another autistic woman with an autistic daughter in the same school as one of mine and I have known others. I strongly suspect an acquaintance with 3 daughters none of whom could cope in school and were home-educated, are all autistic, including the mother. It takes one to know one as they say. I also have several autistic female online acquaintances. How likely is it that I happen to live in some bizarre cluster scenario? Something in the water…I think not.

        In this video Tony Attwood goes on record as saying he believes the gender ratio is 50:50:

        Diagnostic rates of the autism sub-type PDA are actually slightly more females than males.

      • EJ says:

        Here is the info on there being more girls with PDA:

        P1 “This paper was first presented at the World Autism Congress held in Cape Town, South Africa in 2006. It provides details on a syndrome which was identified over a long period of time by Professor Elizabeth Newson, often in work done jointly with this author, Phil Christie. In the many diagnostic assessments conducted at the Child Development Research Unit based at the University of Nottingham, she found there were children referred with a possible diagnosis of autism who did not seem typical in that they shared some of the features but had other very different behaviours and characteristics. There were also more girls affected than boys.”

      • EJ says:

        Here’s more:

        and debate continue as to whether this is a different condition than
        autism, whether it is perhaps the female presentation of autism, as more
        girls have been identified with the PDA profile than boys…”

Leave a Reply

Your email address will not be published. Required fields are marked *