Dr. Mandell: I was trained in public health with a broad interest in children’s psychopathology and improving mental health out-comes for kids. I did an internship for the city of Philadelphia in 1998, during which time I was asked to look at their very expensive children—those who were using the most mental health services. I found that about half of the hundred most expensive kids in the city were kids with autism. Since then, I have become an advocate for and a researcher on improving quality of care for people with autism in community settings.
CCPR: What’s your view on the prognosis for autism? We sometimes have parents say to us: “Is my child doomed forever?” How do you answer this?
Dr. Mandell: First let me preface this by saying I am not a clinician—I am a psychiatric epidemiologist and health services researcher. But I do talk to a lot of families about this. The prognosis of autism is still a mystery and the best predictors we have of prognosis are the same things that are predictors for kids in general. So the higher your I.Q. when you are a little kid, for example, the better your prognosis is. We are starting to get a sense that reciprocal engagement and joint attention seem to be important predictors of a positive long-term trajectory. But the reality is that these are averages, and I think our predictive models of how children with autism are going to fare over time are still really poor.
CCPR: Why do you think this is?
Dr. Mandell: I think it is the failure of the science at this point. and, increasingly, people in the field talk about autisms rather than autism, with the idea that autism might be as heterogeneous as intellectual disability. There are thousands of different causes of intellectual disability, and I think we will find that it is the same thing with autism. We just don’t know how to measure differences among these types at a behavioral level. The autism diagnostic measures have the highest reliability of any of the measures of disorder in childhood psychopathology, but they were designed to be diagnostic measures; they were not designed to quantify children’s autism in different domains. We are not good at measuring in a quantitative or dimensional way how children are doing, especially socially, and also in pragmatic communication, which I think is the real communication challenge kids with autism face. We have a big measurement issue and, in fact, we are not even sure we are measuring the right things. So it becomes very difficult to figure out what predicts long-term outcomes for kids with autism. I do tell families that the earlier they can get their kids into intervention, and the more intensive that intervention, the better the outcomes for their kids are going to be.
CCPR: What are the strategies that have an evidence base for working for autism?
Dr. Mandell: It depends what you are trying to address. We know that for most of the behavioral challenges and the specific core symptoms of autism the most effective interventions are those in the family of applied behavioral analysis. We used to think of that as meaning very didactic, intensive intervention led by a therapist where the kid is sitting at a table for many hours a day. While that component is still an important part of intervention, one of the exciting things that has happened over the last two decades is the understanding that you can weave these principles of applied behavioral analysis and learning theory into many aspects of the child’s day in a way that is child-directed and pervades the normal activities in which the child would engage. There has become an increased interest in these more developmental approaches to treating kids with autism—techniques that originate out of the Floortime model that Stanley Greenspan developed and popularized (www.stanleygreenspan.com/).
CCPR: What are the overarching principles of these approaches?
Dr. Mandell: Probably the most successful interventions are those that find a way to engage the child in some activity that is reinforcing in and of itself, and use that engagement paired with learning theory to both teach new skills and to make the child want to engage socially. In pivotal response treatment (PRT), they talk about setting up challenges that require a requesting behavior from the child, for example. Greenspan talks about being playfully obstructive in Floortime, which, of course, has the same effect. These two styles of intervention come from very different places, but in practice end up looking very similar. I think one of the reasons why people have been instinctively less interested in computer programs that use the teaching method known as discrete trial training is that they don’t involve the interpersonal engagement that makes up a really important part of the successful interventions.