CCPR: Dr. Kolevson, by the time a patient with autism walks into our office, that patient may have already been seen by other professionals. What is the particular role of a child psychiatrist? 

Dr. Kolevzon: Child psychiatrists play something of a unique, and critical, role within autism, and as you said, we are often the next-line referral. Sometimes the patient goes from a general pediatrician to a developmental pediatrician or to a child psychiatrist or sometimes to a pediatric neurologist. There is a lot of overlap in the roles of all three of these medical specialties. And the critical role initially is to establish the diagnosis. While we may not be able to establish the diagnosis by ourselves, we should act as a team leader and bring the right people to the table. This is something I do all the time, and once I bring this expertise together, I can be fairly confident that a given child has autism.

CCPR: So let’s say you are referred a 4-year-old kid from a pediatrician with a concern about autism. How can you be certain that this truly is autism and not some other condition that may be masquerading as autism?

Dr. Kolevzon: The core deficit in autism is the social cognitive deficit. There are a number of features around repetitive behaviors and anxiety that can as you say masquerade as autism, but I am looking very carefully at social engagement, social attention, eye contact, responding to name, reciprocal play. And you want to rule out all these other types of problems. Kids can have a lot of anxiety and social phobia, for example, which can masquerade as autism. They can have ADHD and attentive, impulsive types of symptoms that impair their social function dramatically and can look like autism. And then there are a whole host of other developmental disorders.

CCPR: What kinds of behaviors do you see in your office that cause you to zero in on autism as opposed to other disorders?

Dr. Kolevzon: One of the key features of autism that you can pretty dependably see in the office is a deficit in shared attention, also known as joint attention. A good way to diagnose this is to point at various things and see how the child responds. You make sure the child is looking at you and then just point out the window and see if the child follows your gaze. In that situation, almost all kids, even kids with ADHD, will divert their attention and follow your gaze and look out the window with you. But kids with autism are completely oblivious to it. Or you ask the child, “Where is your mom?” Do they look at their mom? If you point at her, will they follow your point? Or you ask them to point at something “orange” in a picture. Remarkably, a child with autism will not be able to point to orange, even if you are looking right at the color. 

CCPR: That is fascinating. What do you think is happening in their world, in their mind, when they are not following your gaze?

Dr. Kolevzon: This is one of their core social cognitive deficits. They have significant impairment in their capacity to take the perspective of somebody else. What is going through your mind or your internal emotional state are very mysterious to them. They are not particularly interested in what you are doing; it doesn’t have the same salience to them. They don’t recognize emotional expressions or the tone of your voice; they may not even recognize basic facial expressions.

CCPR: What else do you do to elicit diagnostic behavior?

Dr. Kolevzon: You can just call their name. If you call their name and they don’t respond, that is pretty specific to autism. Kids with ADHD may not respond quickly, but they will respond. Also, kids with autism will not imitate reflexively: If you clap, they should clap, and so on; kids with autism may not.

CCPR: What about deficits in language development? How do you pick up on that?

Dr. Kolevzon: There are two levels of that. The first is just not developing language appropriately—for example, they are not using single words by 12 months, or they do not have phrased speech by 36 months. Or if they are fluent in their speech and they are able to use phrases, for example, they might use scripted phrases; they might use phrases inappropriately; they may have trouble going back and forth in a conversation. So you can ask them a question and they can answer it, but then there are not any follow-up questions.

CCPR: Give me an example.

Dr. Kolevzon: 
“What did you do today?”
“I ate lunch.”
“What did you have for lunch?”
“I had French fries.”
That is it. In normal back and forth speech, someone will reciprocate and ask you, “What did you have for lunch?” There will be some sort of shared engagement, but that doesn’t happen in autism. You can even say something like “Tell me what kinds of food you like to eat?” And they will give you kind of a one-word answer: “I like French fries.” They will just repeat the same thing. This goes back to the social deficit. Language is our social currency, and if social interaction doesn’t have meaning for you, then you are much less likely to use the social currency.

CCPR: How do you communicate the diagnosis to the parents?

Dr. Kolevzon: More often than not, by the time parents actually get a referral to a specialist they are relatively convinced that something is wrong, so often there is a sense of relief. “Oh, now I have a name for this and a plan for what to do. All along I sort of knew something was off, but now you are giving me a universe where I can focus my energy; I have a map of the road ahead, and I have a whole community that I can relate to.”

CCPR: But what do you say to parents who ask how you know their child has autism?

Dr. Kolevson: That can be challenging, because in fact autism is a completely artificial diagnosis. It is just a collection of behavioral symptoms, and the diagnosis is made irrespective of etiology, and whatever behavioral symptoms we identify cross a dozen different diagnostic boundaries like many other diagnoses in psychiatry. Parents will sometimes get frustrated when I tell them that I know it is autism “because it fits the diagnostic criteria.” But in fact the criteria just changed two years ago [Editor’s Note: See this article for more on these changes], and they changed 15 years before that. 

CCPR: When families ask me to explain what autism is, I find I am really not sure what to say.

Dr. Kolevzon: That’s because there is no one “autism.” There is a famous quote: “if you’ve seen one child with autism, you’ve seen one child with autism.” because it is so heterogeneous. The core features of autism are behaviorally defined where there are clearly social cognitive impairments, language delays, and repetitive behaviors. The debate is how specific that is because these kids all look so different.

CCPR: What can be done after the diagnosis is given to the parents? What’s the next step?

Dr. Kolevzon: That really depends on what is going on with the child. You basically select your target symptoms, and some of these symptoms will require educational interventions, some of them require behavioral interventions, some of them require speech and language interventions, some of them require physical therapy, some of them require parent support and sibling support, and some of them require medication.

CCPR: That’s a lot of work!

Dr. Kolevson: It feels like a lot to land on the psychiatrist, but a lot of it gets farmed out. You establish the diagnosis. Then you establish the IQ and the adaptive functioning. That gets incorporated into an educational plan, which dictates where the child goes to school, and that educational plan also incorporates things like speech and language therapy, and social skills training, and counseling, and it is the whole package. From the psychiatrist’s perspective, a significant amount gets referred out with the psychiatrist acting as team leader and someone the family consults with.

CCPR: What would it mean for me to be the team leader when I am just sort of doing one small piece, say the medication piece?

Dr. Kolevzon: It means that you are overseeing everything and you are making sure that this kid is making gains and staying on track, and if the school isn’t doing its job you are writing a letter; you are forcing them to update an IEP meeting. If the psychologist is sort of missing a piece that the family is concerned about, you are speaking with the psychologist and trying to jigger the treatment plan a little bit.

CCPR: What are the elements that you are trying to assess at every visit?

Dr. Kolevzon: There are a lot of medical issues that can be part of the etiology, so I am often checking in on problems like seizures, cardiac issues, kidney problems, gastrointestinal problems, and so on. I’ll ask about specific psychiatric symptoms if I am using medicines to target them, such as attention deficit, anxiety, irritability, and aggression. And then I’ll check in on educational issues—what is going on in school, are they making appropriate gains? I also tend to ask about the family functioning in general. How is the sibling doing with all this? How are the parents feeling about all this? Is the family starting to feel more and more isolated because their friends don’t want to hang out with them because their kid is so difficult, so are there additional sort of family supports you can recommend? It sounds like this is a very comprehensive and idealized process, but the reality is that these are things that you want to keep track of, and I do think that as a “team leader” it is the physician’s job to pull all these pieces together.

CCPR: We always like to get experts’ philosophies about medications. There are only two FDA-approved meds for autism: risperidone and aripiprazole.

Dr. Kolevzon: As a result of the FDA approvals, these two medicines have turned into first-line treatments, and more often than not people are too quick to prescribe these medicines. Normally when I end up recommending risperidone or aripiprazole, the family is kind of desperate. When kids are aggressive to the point where they are going to hurt themselves or somebody else, that is an appropriate use of risperidone. It’s also appropriate if they’re in danger of getting kicked out of school or if they’re psychotic. But for anxious, irritable, pain in the ass kid who can’t tolerate changes in routines, I would want to start with things that reduce anxiety or impulse control problems. [Editor’s Note: See this article for more on medications and autism.]

CCPR: Do you see a big difference between risperidone and aripiprazole in terms of side effects?

Dr. Kolevzon: Not really. It is not like aripiprazole doesn’t cause weight gain and metabolic problems; it does. There have only been two really large, well-done studies of aripiprazole. There have probably been a dozen with risperidone, and it clearly causes weight gain and it clearly causes dystonia and those kinds of problems. I do think the metabolic risks are a little bit lower with aripiprazole, but on the other hand I believe aripiprazole is a little bit less effective than risperidone. So once I’m there I usually start with risperidone because I don’t think aripiprazole has some massive advantage over risperidone by any means.

CCPR: When you are looking at anxiety, what kind of meds do you tend to reach for?

Dr. Kolevzon: I will use SSRIs and nonstimulants, such as extended-release guanfacine, clonidine, even atomoxetine, which can bring their anxiety down and enhance their attention, and improve their impulse control. For ADHD hyperactivity-type symptoms, I use stimulants all the time. There are plenty of studies that suggest methylphenidate is extremely effective. And then for the kind of more irritable, aggressive kids, I use antipsychotics or I use anticonvulsants like divalproex sodium.

CCPR: And then within the SSRIs, are there any in particular that you have found easier to prescribe or more effective?

Dr. Kolevzon: I go with the safety data in pediatric populations in general, so Prozac and Zoloft are usually my go-to medications because we have so much more experience with them. They also both come in liquid, allowing you to use miniscule doses. The key thing when you treat kids with autism is you have to acknowledge that overall the response rates are lower than they would be in typically developing kids with anxiety or ADHD, so don’t expect necessarily the same effect. And then also they are exquisitely sensitive to side effects so that you really want to start at much lower doses than you otherwise would.

CCPR: Thank you, Dr. Kolevzon.