Perkins School for the Blind Transition Center

Overcoming the Challenges of Assessing Young Children for Autism

Assessing young children for autism can be very challenging. It can be quite difficult to be confident in a diagnosis of autism at the earliest ages due to many factors that I’ll highlight below. At the same time it is very important for parents with concerns about their child to have an evaluation as soon as possible. Once an evaluation is complete and the parents receive a diagnosis for their child (when this is warranted), then the appropriate treatment can be provided. As we say in early intervention, “the earlier the better” the intervention to achieve the best outcomes for the child.

The main challenge of the evaluator is to answer the question, “Are you sure he/she has autism?” For the evaluator to answer, “Yes, I have confidence in my diagnosis of autism,” the evaluator has to overcome the following challenges. The evaluator must be clear about the diagnostic criteria for autism and how to identify these criteria in the very youngest children, the evaluator must have sufficient knowledge of autistic disorder, the evaluator must have sufficient supervised training and experience, have adequate skill, be able to understand and sort out variability in a child’s performance on the day of testing, be able to validate their own observations with the parent report, and finally be able to bring all of this information together in a diagnosis and coherent report.


Diagnostic Criteria of Autistic Disorder and Presentation in Young Children


A diagnosis of any early childhood developmental disorder is made by determining if a child meets the criteria of that disorder as specified in the latest edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR, 2000). The 3 main features of Autistic Disorder are: Impairment in social interaction, impairment in communication, and restricted repetitive and stereotyped patterns of behavior, interests and activities (DSM-IV-TR, p. 75). The challenge in assessing the youngest children for autism is to determine what is considered “normal” and to know if the child’s behavior should be considered an impairment in social interaction. Let’s consider evaluating an infant of 12 months for possible autistic disorder. First, one must consider what is normal. At 12 months of age a typically developing infant is highly social towards his or her primary caregiver, for example. Typical social interactions include a high level of eye contact, gesturing for adults (such as reaching for them), moving towards the primary caregiver and preverbal sounds. The infants’ movements and gestures might be accompanied by a variety of sounds such as crying, screeching, sounds of delight, laughter, or perhaps a word or a word attempt such as “ma.” A significant impairment in social interaction in a 12 month old infant could include one or more of the following characteristics. The infant may not make eye contact at all, or it might be limited compared with what is typically expected at the child’s age. The infant might be very “quiet” and might even be considered a “good baby” in that the baby does not fuss much. The infant might not gesture much toward the primary caregiver, or perhaps they might not mind when the caregiver is not present. Keep in mind that these characteristics are only possible indicators of an impairment in social interaction. Determining if any group of characteristics represents a difficulty in development or signs of autism is very subjective and can only be determined when there is a sufficient amount of data. The data come from 3 sources: direct observation of the infant or child, parent report, and results of formal testing. All of the data of the child’s behavior must be examined and compared with what is normal and what is considered atypical. Finally, a diagnosis of autistic disorder is made if the behaviors of the child (based on a sufficient amount of data from the 3 sources) align with the criteria for this disorder. There are formal measures for autism (such as the ADOS and the CARS) and they provide information, along with observations and parent report, to help the evaluator determine the appropriate diagnosis.


The Evaluator’s Knowledge of Autism


In addition to understanding how Autistic Disorder presents in the youngest children, the evaluator must know about the disorder overall. This includes understanding the latest theories about causes of autism, prevalence, sex differences, course of autism over the lifespan, treatments available, and possible outcomes as well as the programs available to assist parents. Having this knowledge makes it easier to speak with parents and to answer the typical questions that arise when parents are faced with this diagnosis.


The Evaluators Supervised Training and Experience


Over the last 15-20 years the Early Intervention Program has expanded rapidly throughout the United States and other developing countries. This has resulted in a large increase in services and evaluations provided to the youngest children. Despite the increase in services provided under this program the number of evaluators receiving training in this area has not kept pace with the demand. Evaluators, such as psychologists or special education teachers, for example, are required to maintain a license or certification in their area. However, beyond their license or certification, there are no specific requirements that the evaluator have formal supervision or training in the assessment of young children. As the field matures there will likely be further specific requirements in order to ensure that the evaluator is fully competent in this area. The more supervised training the evaluator has, and the more experience the evaluator has, the more likely it is that the evaluator will be able to provide evaluations with an accurate and valid diagnosis so that early help will be received by children who need this help.


The Child’s Variability in Performance (Test/Retest Reliability)


The behavior of young children changes rapidly from moment to moment and day to day. That is one of the delights and challenges in assessing them. One minute they might be laughing and in an instant they might be crying. Factors affecting a child’s performance include the time of day of testing, the place of testing, the person(s) accompanying the child in the evaluator setting, the child’s reaction to the evaluator, and whether the child is hungry, thirsty, and well rested. The experienced and effective evaluator will assess each of these variables and carefully consider them when working with the child. In addition to considering these variables, the evaluator will try to optimize the child’s performance. A short list of techniques to optimize the child’s performance include ensuring that the child is well fed, that testing occurs ideally when the child is well rested, is in a comfortable environment (such as home, or if in an office, then one that is child friendly and appealing), and that the evaluator typically allows the child time to warm up by speaking with the parent first, then playing briefly with the child, and finally testing the child with appealing toys and/or testing materials. Evaluators receiving formal training in the evaluation process through graduate courses or further training should be well versed in the issue of test/retest reliability and carefully consider the variability in a given child’s performance when making a diagnosis.


Validating the Observations of the Child with the Parent Interview


Despite the best efforts of the evaluator it may not be possible to obtain an ideal or optimal performance of the child on the day of testing. While young children are typically seen for one evaluation session, a child can be seen for further visits if indicated. No matter if the child is seen for one session or more the evaluator must compare their observations with the parent report. It is ideal when the parent states that the child’s behavior was typical on the day of testing. However, when the parent states that the child did not behave typically, then the examiner must probe for more detail into the differences between the parent report and what was observed, and possibly consider another testing session. Keep in mind that the parent or primary caregiver has many more observations of the child than the evaluator. In addition, they may have a strong wish either for their child to receive or to not receive a diagnosis of autistic disorder. It is certainly very understandable that a parent would not wish for their child to receive this diagnosis and this might affect their report of the child. All of this makes it more challenging for the evaluator to make the final diagnosis.

Making the Diagnosis: Putting it All Together


Finally, the effective evaluator must make the final determination of whether or not the child presents with sufficient characteristics to make the diagnosis of Autistic Disorder, Pervasive Developmental Disorder-Unspecified, or perhaps another disorder. There is no substitute for gathering as much data as reasonably can be gathered in a short period of time. Then, the evaluator must simply compare the child’s behavior with the available diagnoses and make a determination. The more experience the evaluator has, the more likely it is for him or her to have confidence in an appropriate diagnosis that will then allow the child to obtain any needed services. Remember that there are now several effective treatments for autism and the sooner the intervention the better.


Scott Mesh, PhD, is CEO at Los Niños Services. Scott can be reached by phone at (212) 787-9700 or by email at To learn more about Los Niños Services, visit their website at

Have a Comment?