The Autism Diagnostic Observation Schedule (The ADOS) is a diagnostic instrument that was created by the University of Michigan Professor Dr. Cathy Lord. During the last 2 decades, the ADOS has become the most accepted diagnostic tool for autism spectrum disorders. The ADOS has 4 different overlapping ‘versions’ (or modules) that were designed to be used with individuals of various ages and abilities – including non-verbal individuals. However, the original ADOS is not very useful in the diagnosis of children under the age of 3. For these children, the ADOS is not specific enough. That is, it incorrectly identifies ASD in many children who actually have a non-ASD developmental delays.

But why do we need an autism diagnostic instrument for children under 3?

There are actually a number of valid and important reasons in support of the early diagnosis of autism. When conducting evaluations of children with autism I hear parents describe how they knew that ‘something was wrong’ since their child was very young. This phenomena is not just a clinical anecdote, as it has supported by research studies (see for example Chawarska et al. 2007 DOI:10.1111/j.1469-7610.2006.0185.x) suggesting that in some children, clear symptoms of autism can be identified very early. In addition, a number of studies have shown that early intervention is extremely important in the treatment of autism, thus early identification would help families obtain the interventions they need.

Given the need to have a diagnostic instrument that can be used with children under 3, Dr. Lord and her team at the University of Michigan have been working on a new ADOS module that would reliably identify autism in these young children. The results of these efforts have now been presented in an article to be published in the Journal of Autism and Developmental Disorders. In the article, the authors described in detail the process that led to the development of the ADOS new toddler module (ADOS-Module T). However, I will limit this post to a description of the validation procedures.

In order to test this new module, the authors used the ADOS-Module T in 360 clinical evaluations with children under age 3 conducted at the University of Michigan Autism and Communication Disorders Clinic, and at the University of California-San Diego Autism Center of Excellence. These children included those who eventually would receive a diagnosis of non-ASD developmental delays, ASDs, or no diagnosis at all (typically developing). The ASD children had their clinical diagnoses of ASD based on a “best estimate” procedure conducted by specialists, and based partly on a modified version of the ADI-R. The non-ASD developmental disorder group as well as the typically developing group were also evaluated for ASDs with the ADI and they did not meet standards for ASDs.

So in essence, the ADOS-Module T was employed on 3 groups of children: Children with ASD, children with a non-ASD disorder, and typically developing children. The clinicians administering and scoring the ADOS-Module T were unaware of the eventual diagnoses of these children. This allows the researchers to examine the specificity and sensitivity of the new ADOS module in the correct identification of autism spectrum disorders.

What is sensitivity and specificity? Sensitivity refers to how accurate the instrument is in the identification of autism when autism exist. For example, when a test has 80% sensitivity, this means that 80% of the time when a condition is present the test will ‘catch it’. Specificity however, refers to how well the test differentiates the target condition from other conditions. So for example, a test may have very high sensitivity in that every time the target condition (in this case autism) is present, the test gives you a ‘positive’ result. But the same test my have very low specificity, in that it also gives you a positive result when a different condition is present, so that it incorrectly identifies the target condition as present when it’s not there!

How did the new ADOS Module T perform?

  1. The sensitivity of the ADOS Toddler module was 91%. That is, the test was able to correctly identify 91% of the cases of ASD (based on a cut off score of 12).
  2. The specificity of the ADOS-Module T when tested against non-ASD disorders was also 91%. This means that only in 9% of the cases, the test suggested a diagnosis when the child had been previously identified as not having an ASD.
  3. The specificity of the ADOS-Module T when tested against typically developing cases was 94%. That is, only in 6% of the cases, the test suggested a diagnosis in children who were actually typically developing kids.

These are actually excellent numbers and indicate that the ADOS Toddler Module has excellent sensitivity and specificity. However, the authors also described some general concerns and limitations.

The ADOS, although it is the most reliable and valid diagnostic instrument available, it is still only a clinical tool that must be used in the context of a comprehensive clinical evaluation and it is subservient to clinical judgment. Specifically, a diagnosis of autism is provided only when the person meets the DSM-IV diagnostic criteria. Therefore, clinicians must use their judgments in interpreting and applying the results from the ADOS. There will be cases when the ADOS suggests a diagnosis but the clinician will not provide the diagnosis because the child doesn’t meet full diagnostic criteria based on the DSM-IV.

So you may ask, what is the point? Why do we have the ADOS if all a clinician has to do is go down the list of the DSM-IV criteria and add up the check marks? The ADOS provides for a reliable and valid tool to assess for the specific symptoms included in the DSM-IV criteria and it helps the clinician interpret the child’s clinical presentation as it applies to the DSM-IV criteria. The ADOS standardizes this process so that diagnoses are less dependent upon other factors, such as biases in parental reporting of symptoms, or the skills or training of the clinician in properly indentifying such symptoms. Therefore, the ADOS greatly improves the validity and accuracy of our ‘clinical judgment’.

On a personal note and disclaimer. Dr. Luyster (lead author of the study), Dr. Richler, and Dr. Oti were all my classmates in graduate school and I congratulate them for their wonderful work. In addition Dr. Lord, creator of the ADOS and founder of the University of Michigan Autism and Communication Disorders Clinic, will be my collague this Fall when I join the University of Michigan faculty.

The Reference: Luyster, R., Gotham, K., Guthrie, W., Coffing, M., Petrak, R., Pierce, K., Bishop, S., Esler, A., Hus, V., Oti, R., Richler, J., Risi, S., & Lord, C. (2009). The Autism Diagnostic Observation Schedule—Toddler Module: A New Module of a Standardized Diagnostic Measure for Autism Spectrum Disorders Journal of Autism and Developmental Disorders DOI: 10.1007/

Post to Twitter

Tagged with:

7 Responses to Diagnosing autism in toddlers: The new ADOS Toddler Module enters the field

  1. Brandon says:

    Very interesting. Mr. Lopez I hope you enjoy coming to the U of M. I am in the Adult Autism Social Group up there, so I know some of the researchers. I find the ADOS-T very interesting I will be sure to ask Katie Gotham more about it.

  2. JulieL says:

    I listed to a recent podcast with Professor Margot Prior. She noted that there is recent research in the US and UK stating that children can be, as she stated “picked up as at risk” for autism, at the age 12-24 months old. She was clear to state that this was not to say these would be always accurate, but it would be say a marker for close observation for that child. Do you know of anything regarding the research she speaks of? This new research article you present here seems to be adding to that body of agreement.

    Thanks :)

  3. Pregnancy Questions says:

    Do you plan to keep this site updated? I sure hope so…it’s great.

  4. Naomi Weatherford says:

    Very interesting article Dr. Lopez-Duran.
    I am caring for a child who is believed to have been neglected for 15 months and at 30 months continues to have significant delays in speech, coordination, and social development. Is there a reliable tool you are aware of to determine if the child has an autism spectrum disorder versus continued delay due to maltreatment?

    • Hi Naomi, at 30-months a local pediatric neuropsychologist or autism specialist could provide an evaluation that could guide you in obtaining the necessary services. Most professionals may be a bit reluctant to provide a diagnosis at that age, but they can identify the pattern of strengths and weakness that could help him/her receive the best intervention at this time. Thank you. Nestor.

  5. Gina Marriott says:

    I appreciated your article so much. Is this Toddler Module now available for purchase? If not, do you know when it will be published?

Leave a Reply

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


three + 3 =

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Looking for something?

Use the form below to search the site:

Still not finding what you're looking for? Drop a comment on a post or contact us so we can take care of it!

Set your Twitter account name in your settings to use the TwitterBar Section.