Diagnosing Autism Spectrum Disorder: A discussion with Dr. Sam Goldstein

We sat down with Dr. Sam Goldstein, co-author of MHS’ Autism Spectrum Rating Scales™ (ASRS®), to discuss challenges that can arise when diagnosing autism spectrum disorder (ASD) and emerging trends within the clinical field related to ASD.

The interview below has been edited for length and clarity.

An ASD diagnosis can be a complex and arduous process due largely to wide variations in symptom types and severity, and changes with age. How can clinicians avoid common obstacles or pitfalls while aiming to achieve accurate diagnosis?

All childhood diagnoses require careful analysis and data gathering in multiple ways from multiple sources. The basis of the DSM diagnostic system is polythetic. A child does not demonstrate every single symptom but a sufficient number of symptoms. The diagnosis of ASD is challenging, more so than something like Attention-Deficit/Hyperactivity Disorder (ADHD) or Oppositional Defiant Disorder, because those conditions have very specific descriptions of behavioral symptoms. ASD provides a broad set of examples but not every observed behavior within a category. I don’t necessarily agree that the process is arduous, but it takes time. It begins by obtaining a careful history from parents. This reflects an effort to better understand and appreciate the child’s journey and the family’s functioning. Second, we use parent and teacher questionnaires that have been well-validated and standardized, not only within the general population but on children with various diagnostic conditions. It is important to keep in mind that questionnaires don’t diagnose. They provide valid, reliable, and statistically relevant data about how a particular child’s functioning relates to children in the general population.

Additionally, there is an issue of frequency, duration, and intensity for all symptoms. Questions about symptoms in this regard need to be asked during an interview. Finally, the evaluation progresses with a direct assessment of the child. In younger children with developmental delay and suspected ASD, this involves a structured play interview. A more comprehensive neuropsychological evaluation is needed with older youth to evaluate neuropsychological functioning, academic achievement, emotional development, and behavior. All of this data must be carefully integrated in a comprehensive, logical fashion to arrive not just at diagnostic labels but at an appreciation of the child’s development and a set of data that can be used to begin planning treatment as needed. The best way to avoid common obstacles and pitfalls is to know the subject, possess the training and expertise to complete a comprehensive, thorough evaluation, and never assume that you know everything about a particular child or family.

We know that any child who might have social pragmatic issues may not have ASD. Because getting it right is so important, how can a clinician mitigate false positives and false negatives?

To better understand and streamline the diagnosis of ASD, a milder, similar pattern of social challenges, Asperger’s Disorder, was eliminated from the DSM-5. A new diagnosis, Social Pragmatic Communication Disorder, was created. Essentially, this is similar to the Asperger’s diagnosis. These are children with primary social pragmatic language problems absent significant problems with self-regulation or atypical behavior. The best way to mitigate false positives and false negatives is to appreciate that, as evaluators, we always balance sensitivity with specificity. A sensitive assessment will find all children with problems and identify false positives. A specific test will identify children with the condition but reject children who have a milder form of the disorder.

We know ASD is associated with, or can be masked by, or mistaken for, other commonly co-occurring diagnoses. What makes differentiation so challenging, and how do you think the ASRS helps?

When you examine the validity studies Dr. Naglieri and I completed for the ASRS, it is evident that the ASRS is excellent at separating children with ASD and related symptoms from children in the general population and children with other clinical conditions. Thus, we consider the ASRS a narrow-band instrument, one that is used specifically when questions are raised about patterns of behaviors that could be related to ASD. The co-occurrence or comorbidity of ASD, particularly with ADHD, is fairly high. However, ASD is typically not easily confused with other conditions in its presentation.

The ASRS was the first nationally standardized, norm referenced ASD rating scale created. How have you seen the impact of this notable achievement?

Every year since its publication the use of the ASRS has grown nationally and internationally. It is by far one of the greatest achievements in my career.

The rate of ASD diagnosis has risen significantly in recent years. Studies between 1960 and 1980 found a prevalence of 0.02% to 0.05%. In the early 2000s rates were between 0.3% and 0.6%; by 2006 0.5% to 1.14%. Why do you think the rate of ASD diagnoses has risen so significantly over time?

The issue of the increase in ASD diagnosis really is a question of asking: are there now more children with ASD or, simply, are more children with ASD now being identified? I think the latter is the case. There is significantly greater awareness in the community among parents and educators. A child in the past who might have been overlooked as being shy, socially awkward, or slow to warm up may now be referred for evaluation and found to experience ASD. One of the interesting trends is that the percentage of children with ASD and typical levels of intelligence has likely been increasing as we can better identify these children accurately.

The new DSM-5-TR update has just been released. Considering this and beyond even DSM diagnostic criteria – what changes would you like to see in how ASD is diagnosed?

For the moment, the one thing I would like to see that I don’t believe the DSM has included is an effort to evaluate coping, camouflaging, or masking behaviors. Our initial data set for adults with ASD demonstrates that their level of impairment is directly related to the effectiveness and efficiency of their coping strategies. That is, the more effective they are, the less their symptomatic challenges are observed by others.