Clear, engaging, and practical updates on clinical psychiatry.

Earn CME for listening to the podcast with a multimedia subscription. Listen for free here or using Apple Podcasts, Android, or Stitcher.

Previous Post
Next Post

Effect Size Matters: A Mini Master Class in Autism Intervention

Podcast, Volume , Number ,

Print Friendly, PDF & Email

This deep dive podcast breaks through the dogma of autism intervention and helps you to understand the three main branches of autism intervention and their very different mechanisms, the quality of autism research with a huge shift in our understanding of what works, and addresses such questions as how many hours are really needed for intervention?

Published On: 8/19/2021

Duration: 27 minutes, 21 seconds

Related Article:Effect Size Matters: The Seismic Shift Toward Naturalistic and Developmental Interventions in Autism,” The Carlat Child Psychiatry Report, April/May/June 2021

Got feedback? Take the podcast survey.


Dr. Feder: As clinicians we are often faced with questions from families about the “best” program for autism intervention. While children with autism may receive a range of services including speech and language therapy, occupational therapy, and social skills help, the anchor to most programs is some form of behavioral health treatment. The latest issue of The Carlat Child Psychiatry Report includes an interview with Dr. Micheal Sandbank, who’s an assistant professor of special education at the University of Texas, Austin. Dr. Sandbank specializes in early childhood special education, and she and her colleagues published a systematic review and meta-analysis that has taken the Autism research community by storm. In this podcast, Mara and I will discuss Dr. Sandbank’s research and the three main approaches to autism intervention.

Welcome to The Carlat Psychiatry Podcast.

This is a special episode from the child psychiatry team.

I’m Dr. Josh Feder, The Editor-in-Chief of The Carlat Child Psychiatry Report and co-author of The Child Medication Fact Book for Psychiatric Practice.

Mara: And I’m Mara Goverman, a Licensed Clinical Social Worker in Southern California with a private practice.

I’m also a regional director at Positive Development Institute.

Before we dive into our interview with Dr. Sandbank, let’s take a few minutes to briefly summarize the three main intervention approaches for young children with autism.

Dr. Feder: In the past couple of decades, the field of autism intervention has evolved into three main evidence-based approaches, traditional Applied Behavioral Analysis (ABA), Developmental Relationship-Based Intervention (DRBI), and Naturalistic Developmental-Behavioral Intervention (NDBI).

ABA is the best-known type of intervention. It is based on operant learning theory, meaning that behavior is learned based on what happens before the behavior (antecedent) and what happens after it (reward). Some drawbacks of the ABA approach include poor maintenance of skills, poor generalization of learning to new situations, and reliance on adults to inform children what to do (prompt dependency).

Mara: In contrast to behavioral intervention, DRBI is a parent mediated intervention (PMI) where the primary focus is on supporting parents and other caregivers to build and use warm, meaningful interactions to help the child to function better in communicating, learning, and problem solving. The strategies of developmental interventions are distinct from behavioral intervention approaches in that they use free play without direct didactic instruction or contingent rewards. An adult takes a child’s interest and builds on it, while making the activity an emotionally meaningful experience. The best-known model is DIRFloortime® or simply Floortime®.

Dr. Feder: In an effort to address some of the drawbacks of traditional ABA, the third approach to autism intervention, NDBI, incorporates more choices for children to gain their buy-in to treatment. The learning is carried out in natural situations such as play or daily routines and involves parents, and the rewards given are related to the child’s interest. The goals are still more discreet like with ABA, but the choices make for more naturalistic learning.

Mara: Over the past decade or so, the standards for study designs in autism research have changed dramatically.

Dr. Feder: For instance, in the early 2000’s high quality randomized controlled trials were practically nonexistent in the field of autism research. And in 2011, a systematic review from Pediatrics revealed that, out of all of the studies they identified that investigated early interventions for young children with autism, only two were considered high quality randomized control trials.

Mara: A similar systematic review from 2017 set out to quantify the amount of randomized control trials in the field of early autism intervention research. They found a total of 48 randomized control trials, so we do see this upward trend of randomized control trials becoming the standard study design within the field.

Dr. Feder: This field is rapidly expanding, and the rate at which new evidence has emerged is astounding. This tidal wave of new autism research inspired Dr. Sandbank and her colleagues to perform a systematic review and meta-analysis of these studies.

Mara: They used a recently developed technique that let them compare all the effect sizes for different outcome measures across many different studies.

Researchers often declare that a “treatment works” if it has a statistically significant benefit over placebo. Statistical significance gives some confidence that there is a treatment effect, but it’s not the final word on treatment efficacy.

What we really want to know is the effect size – the magnitude of the treatment effect. Common convention is that 0.20 = small, 0.50 = medium, and 0.80 = large. In psychiatry, effective treatments nearly always generate small to medium effects (compared to placebo). It is now standard practice to report effect size in treatment studies. A study that fails to report effect size may be hiding a minimal treatment benefit.

Dr. Feder: Dr. Sandbank and her team summarized numerous studies on ABA, developmental interventions, and NDBI interventions, and they compared the efficacies of these interventions to one another. Her study really sheds light on the quality of many studies examining early interventions for youth with autism.

[Dr. Sandbank’s response]

Dr. Feder: Not only does her study highlight the lack of basic quality studies in the autism field, but it also demonstrates how research can be intentionally or unintentionally set up in a way that creates an illusion of a positive treatment effect when in reality there may not be one at all.

Mara: Researchers have the ability to select specific interventions and/or outcome measures that have the highest potential of producing large effect sizes. But will a treatment or intervention with a large effect size be clinically meaningful?

Dr. Feder: Not necessarily. This is where generalized versus context-bound outcomes and distal versus proximal development comes into play. Imagine I was teaching a classroom of elementary school children how to spell and read. I want to see how two different teaching styles effect the way children perform on a spelling test. So I split the class in half, and taught one half of the class only spelling and pronunciation of the words that would be on the test (Group A), while I taught the other half spelling, pronunciation, grammar, and syntax, without exposing them to the words on the test (Group B). After 1 month, I gave the children the spelling test, and group A completely outperformed group B with an average test score of 95% versus 50%, respectively. So I wondered, “What would happen if I gave both groups a test where they had to spell novel words correctly and order them into coherent sentences?”. I made the test and gave them to both groups and found that group B scored better than group A. However, the difference between the groups average scores was more modest being only 75% for group B and 45% for group A.

Mara: In this example, the first spelling test represents a proximal outcome, meaning that what the intervention specifically targeted, or in this case what Dr. Feder targeted to teach, is what was measured. While the second spelling/grammar test would be more of a distal outcome, where the children would have to build on what they were taught and apply their new skills to a new context.

Dr. Feder: And this also shows the difference between generalized and context-bound development. The children in group A who were taught to spell the same words that were on the first spelling test performed very well on the first test but struggled greatly on the second test. This shows context-bound development and proximal effects where a discrete skill can be applied to a specific context. While group B experienced a more generalized-like development which is why they outperformed group A on test 2. The skills that group B learned improved their internal development sequence which enabled them to build on their skills and apply what they learned to a new context.

Mara: Coming back to research, proximal outcomes are much easier to measure because they’re usually specific skills such as spelling certain words. Whereas, distal outcomes are more difficult to measure.

Dr. Feder: For these reasons, it’s far easier to do research in traditional ABA approaches that teach specific skills such as new phrases. These are narrow proximal outcomes. It has been harder to conduct research on Developmental or NDBI approaches because you need to measure change across a whole domain on a validated, standardized assessment administered by a naïve assessor. That would be a distal effect. Proximal effects appear large, even though they may not generalize nor result in distal growth. Distal and generalized effects are more clinically meaningful but likely to appear smaller. And Dr. Sandbank took this into consideration in her study.

[Dr. Sandbank’s response]

Mara: Teaching social communication may cascade into language development not just within the intervention, but across multiple contexts.

Dr. Feder: Considering that social communication is a core challenge of autism, it’s important that interventions target it and that they don’t just produce proximal effects but that they promote distal development.

[Dr. Sandbank’s response]

Dr. Feder: If you’re familiar with the American Academy of Pediatrics (AAP) guidelines, you might have noticed that the research data presented by Dr. Sandbank and the AAP guidelines don’t line up.

Mara: The AAP guidelines suggest the use of an Autism Diagnostic Observation Scale (ADOS) as a chief diagnostic tool, and the results are taken to a psychiatrist, psychologist, or pediatrician who often doesn’t have expertise in autism but recommends 40 hours per week of intensive behavioral intervention.

Dr. Feder: Many families hope that this will result in the best outcomes, and yet mere access to these hours is not what the research suggests. About a third of kids with autism who are preverbal at 2 are not likely to develop phrase speech by the end of elementary school, and this is related to effective intervention–not access to traditional ABA.

Mara: So, why do the AAP guidelines recommend 40 hours a week? Why isn’t it 10 or 20 hours? I mean 40 hours a week … that’s a full-time job!

[Dr. Sandbank’s response]

Dr. Feder: In this study, Dr. Sandbank’s former advisor, Paul Yoder, compared the efficacies of the Early Start Denver Model (EDSM), which is a form of NDBI, and a traditional ABA intervention and he examined how different intensities of treatment, that is 15 vs 25 hours per week, impacted the efficacy of each intervention. They showed no differences in outcomes by intervention or intensity across the whole group. It was the first study to ask this question in this way.

Mara: And we definitely need more studies like this one because it can be extremely taxing on a young child or a toddler to undergo 40 hours a week of traditional ABA intervention. Not to mention, it can be difficult for families to support their child through this process without seeing clinically meaningful results.

[Dr. Sandbank’s response]

Mara: As time goes on, we will gain more insight into the different interventions and treatment intensities for children with autism. It is imperative that we have high quality randomized controlled trials so that we can improve the treatment of autism with evidence-based interventions.

Dr. Feder: At this moment, the best we can do as clinicians is recommend evidence-based interventions that address the core challenges of autism. We need to inform families that we have high quality evidence that developmental interventions and NDBI’s have modest to moderate effect sizes for improving social communication. But despite the 1000’s of studies in the literature we just don’t have the adequate quality of evidence that can be used to determine the benefits of behavioral interventions on social communication.

Mara: It’s imperative that we familiarize ourselves with the different types of autism interventions because this isn’t a one size fits all kind of deal. What may work for one child and family, may or may not work for another, which means we have to be flexible and work with families to make sure their child is getting the best treatment they can get. We should ask families about which core challenges do they want treatment to address the most. And we need to make sure that they are aware of the different intervention options.

Dr. Feder: And before you recommend a specific intervention, figure out how to make sure that it’s accessible within that particular family’s community. Remember that many of these approaches are being delivered online now, at a distance. Also, families may feel like they’re letting their child down by not being able to fully commit to a 40 hour a week treatment plan. Which is why we need to let them know that we can’t say with any certainty that 40 hours a week of a traditional ABA intervention is necessary, and that they need to pursue a treatment intensity that best suits themselves and their child.

Mara: We asked Dr. Sandbank for a bottom-line message to clinicians and this is what she told us.

[Dr. Sandbank’s response]

Dr. Feder: The print interview is available for subscribers to read in The Carlat Child Psychiatry Report. Hopefully people check it out. Subscribers get print issues in the mail and email notifications when new issues are available on the website. Subscriptions also come with full access to all the articles on the website and CME credits.

Mara: And everything from Carlat Publishing is independently researched and produced. There’s no funding from the pharmaceutical industry.

Dr. Feder: Yes, the newsletters and books we depend entirely on reader support. There are no ads and our authors don’t receive industry funding. That helps us bring you unbiased information you can trust.

Mara: Go to to sign up. You can get a full subscription to any of our three newsletters for $30 off using the coupon code LISTENER.

There are a lot of doctors and nurse practitioners listening, and quite a few students and residents too. Many of you are interested in getting some kind of CME credit for listening.

Dr. Feder: So we’re going to do it! We’re working on putting that program together now and as soon as it’s available you’ll be able to take a post-test for listening and get credits. Everyone who took the survey is going to get early notification when it’s ready because we have their email address.

If you’d like to get that early notification you can still take the survey at

Mara: As always, thanks for listening and have a great day!


Leave A Comment