Why is there controversy with Early Start Denver Model?
From Alycia Halladay, PhD, CSO of the Autism Science Foundation:
There has been some back and forth about a brand of early intense behavioral intervention for ASD called Early Start Denver Model, or ESDM, in the past few weeks. Unfortunately a valid and important scientific debate has crept into the mainstream calling question into the use of early intensive behavioral interventions, period. That’s nonsense. I am very afraid that the result of these scientific debates on type and brand and magnitude of effect, will lead to a distrust in early intervention for symptoms of ASD. I think during all the back and forth on may be clouding the message to families about the importance of early intervention. The magnitude of the effects are variable, they may not be monumental, but early behavioral intervention is important. Now after reading this, I’ve got you curious and you might be thinking: what is going on? I’ll explain.
In 2010, a research group published the first randomized control trial of an early intervention for kids who had been diagnosed with ASD between 1 ½ years of age to 2 ½ years of age1. The intervention was Early Start Denver Model, or ESDM, but as I will mention later, it falls into a larger collection of naturalistic developmental behavioral interventions. These interventions have things in common: they are implemented in natural settings, involve shared control between child and therapist, utilize natural contingencies, and use a variety of behavioral strategies to teach developmentally appropriate and prerequisite skills. They use principles of ABA, which helps the child know when to respond and when the clinician should provide feedback. There are several of these now, thankfully, because very few existed before 2010. Others (but not all) that fall into this category are called JASPR or Joint Attention, Symbolic Play, Engagement, and Regulation, or SCERTS which stands for Social Communication Emotional Regulation and Transactional Support. There is also a Canadian version of naturalistic developmental behavioral interventions called the Social ABCs in Canada. What brand you receive may depend on what you have access to in your community or research going on in your area. Since 2010, these types of interventions have also shown to be effective using randomized clinical trials. I’m not saying these interventions are identical, because they are different in what they emphasize and how they are trained to be delivered. Some of these interventions mostly target social and communication skills, whereas other, such as ESDM, target all domains of skills, including things like fine motor skills. It’s good to have choices. I am not speaking on behalf of the researchers, but I don’t believe any one of the clinician researchers who study these interventions went into the research thinking that any one particular named intervention should be the only one used by everyone. I would be thrilled if my daughter had access to any one of these named interventions. Many of them have been adapted to be parent-delivered, which is incredibly important because let’s face it, 2 year old kids spend most time at home with their parents and it’s important for parents to at least learn what they could be doing in the home and in different settings to promote different behaviors and redirect other behaviors. Getting these home-based models to work has been more challenging, because parents sometimes say, and I get it, I don’t come to your job and tell you what to do so don’t come to my home and tell me what to do. But generally, they are well liked. Without going into too much detail about these, I really hope you have the time to read a more comprehensive discussion of these interventions here. The link is to an article that is open access, so enjoy.
This first randomized clinical trial compared the progress of kids offered ESDM to kids who received whatever was available in the community prior to 2010. It used both therapists and parents. Remember, it was published in 2010, so the work was done earlier to 2010. The results of this study showed that, after two years of intervention, children who received ESDM made larger gains in IQ, language, and adaptive behavior compared to kids who received whatever was offered in the community at the time. In addition, some kids changed from an autism to a PDD-NOS diagnosis, which means they gained some skills in social and cognitive ability1. The study was relatively small, but it did, and does, show that ESDM intervention delivered early resulted in gains for young kids with ASD. The team then took their findings a step further by showing that brain function was changed to look more like typical children compared to those receiving community based intervention after two years2. They kept following up these families and showed that improvements were sustained 2 years after the intervention, when the kids were now 6 years old 3. In fact, other groups have replicated their findings (just the behavioral findings, not everyone has access to brain activity, or EEG monitors). This includes improvement cognitive function in young children 4,5. Because of these gains, the cost of autism services in kids who had received ESDM was lower after they left the ESDM study6. Studies outside the US have also shown ESDM to be effective. All of these things are good news even if the gains are not what some people would consider “drastic”, functioning was improved in the short and long term, and it corresponds to how the brain functions.
Just recently, as in a few weeks ago, another study looking at ESDM was published by the same authors that published the 2010 study. This time, the age was pushed down a few months, and it included three sites7. This was great. These changes from the first study increased the number of children involved in the study, and diversified out who was getting the intervention. It added a parent coaching phase at the beginning since the parents are so involved in the intervention. This parent coaching component had been refined by the team for years and was found to really help parents. The pre-defined primary endpoint was a standardized test of language ability rather than cognitive ability or adaptive function, although those things were included. Some things that make it an example of a well-designed clinical trial are: The treatment was manualized. Therapists were trained and monitored for fidelity of implementation. Outcome assessments were conducted by examiners naive to group assignment. Data were managed and analyzed independently by a data coordinating center with strong expertise in clinical trial methodology. Investigators had no access to the data. So, what did they find the second time around, almost 10 years later and with small changes in the protocol?
There were greater improvements in expressive and receptive language in the children who received ESDM, compared to the community group, however, this effect was only seen at 2 out of the 3 sites. Also, unlike the 2010 study, there was no difference in the improvements in cognitive ability, adaptive behavior and autism severity between the two treatments. In fact, in these areas, everyone showed an improvement, regardless of treatment group. Everyone showing an improvement with no differences between groups is a good thing.
A scientific magazine which is read by the community pointed out the criticisms of the latest trial. The article itself calls into question the required size of an improvement in skills that is meaningful. Is an only “mild” improvement not clinically significant? Who makes that judgement? That was left for the reader to decide. There was a greater than expected chance that language ability was more improved in the ESDM group compared to treatment as usual, and both groups had cognitive and adaptive behavior gains across the time periods studied over the course of the 2 years. Cognitive function is measured differently across time, measuring it just in toddlers shows different trajectories8. A finding of improvement in both groups but lack of difference on all measures is an unfair criticism. Kids in both groups received about the same number of hours of intervention. Researchers including Elizabeth Berry-Kravitz who works in community-based settings, have noticed that more and more, knowledge and skill are being brought into the community and this is leading to better community-based treatment. This is likely why there may not have been an effect – the community interventions got better since the original study. That’s something to celebrate. Of course, not all families live in communities that provide high levels of early intervention. We have a lot more work to do to make sure that all children in all communities get access to evidence-based early intervention.
As a parent of a young child with autism, don’t be dissuaded from this particular controversy. Have faith that behavioral interventions have improved over the years, if you get the opportunity to enroll your child in a program or study that examines one of these named interventions, take it, and realize that your participation will improve the long-term gains that your child will make in the short and the long term. Parents and family members who read about this controversy should not be dissuaded from enrolling their children in ESDM programs or any other naturalistic developmental behavioral intervention program.
- Dawson G, Rogers S, Munson J, et al. Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. Pediatrics. 2010;125(1):e17-23.
- Dawson G, Jones EJ, Merkle K, et al. Early behavioral intervention is associated with normalized brain activity in young children with autism. J Am Acad Child Adolesc Psychiatry. 2012;51(11):1150-1159.
- Estes A, Munson J, Rogers SJ, Greenson J, Winter J, Dawson G. Long-Term Outcomes of Early Intervention in 6-Year-Old Children With Autism Spectrum Disorder. J Am Acad Child Adolesc Psychiatry. 2015;54(7):580-587.
- Touzet S, Occelli P, Schroder C, et al. Impact of the Early Start Denver Model on the cognitive level of children with autism spectrum disorder: study protocol for a randomised controlled trial using a two-stage Zelen design. BMJ Open. 2017;7(3):e014730.
- Vivanti G, Dissanayake C, Victorian AT. Outcome for Children Receiving the Early Start Denver Model Before and After 48 Months. J Autism Dev Disord. 2016;46(7):2441-2449.
- Cidav Z, Munson J, Estes A, Dawson G, Rogers S, Mandell D. Cost Offset Associated With Early Start Denver Model for Children With Autism. J Am Acad Child Adolesc Psychiatry. 2017;56(9):777-783.
- Rogers SJ, Estes A, Lord C, et al. A Multisite Randomized Controlled Two-Phase Trial of the Early Start Denver Model Compared to Treatment as Usual. J Am Acad Child Adolesc Psychiatry. 2019.
- Henry L, Farmer C, Manwaring SS, Swineford L, Thurm A. Trajectories of cognitive development in toddlers with language delays. Res Dev Disabil. 2018;81:65-72.