Do children outgrow autism?



Interviewer: I also wanted to have you come in and talk today. We've been talking a little bit all this week, but I'm not a researcher and I never understand the research. But there have been two big studies. And we've talked endlessly about the criteria changing, and I wanted to have somebody come in and weigh in on this study that got so much press earlier on this week that they're using the word "outgrow autism" and it just has me going nuts. Talk to us about this study, what it really is.

Interviewee: I only had a chance to read the actual study this morning on some of the press releases that I got this morning. For people who don't know, the study was published in the Pediatrics Journal.

Interviewer: Which is a pretty reputable place.

Interviewee: It is indeed. But the study itself was some researchers who were looking at comorbid diagnoses of children with autism. And while they were looking at that, I believe then they were interested in how that affected diagnosis, because they noticed that a lot of children, when they were looking at their database, they noticed a lot of the children who had diagnosis of autism, earlier on, he also had comorbid diagnoses, later on didn't have or reported not to have their diagnosis of autism anymore.

Interviewer: Already we're at a place where as this was reported in the media, nobody was talking about the fact they were looking at comorbid.

Interviewee: Right. I can't remember, I actually just read it this morning, but I don't believe that the authors of the research study actually talk about outgrowing autism at all. That's very much a media interpretation. It seems like, to me, that they've kind of run with this and decided that children can outgrow autism.

Interviewer: And they really have run with it. Every media outlet that I've looked at this week that covers anything about autism, from my local television station to big programming morning news, they've been using the word "outgrow autism."

Interviewee: Right. In fact, the authors, I feel like from just reading that this morning, I feel like the point of their study, they are talking about this comorbidity of diagnosis and looking at why some children are not having the diagnosis later on of autism. Their point was, "Oh, if we can figure out if there is a relationship with comorbid diagnosis or what is going on when the children being diagnosed, we can tailor intervention or more reliable diagnosis," or something like that. That seems like the point of what they were trying to get, trying to start. And it is a starting point. When I'm reading the study, there's a lot of things are popping into mind about how much the media have run with it. Because first of all, the study itself, it totally relied on a database that was already someone else's data. They drew their information from a survey that was done in 2007. I can't remember who did the survey. I don't know if it was a government survey.

Interviewer: It was a phone survey.

Interviewee: Yes. And basically they phoned parents, I believe, one phone call and asked them a bunch of questions over the phone and then they published the results or whatever of the survey. So these researchers went and took their data from that. So that automatically raises the alarm bells for me, because it's survey data, we know is not that reliable for a start. And it is secondary sourcing. They are going to another source to pull data instead of going and collecting their own specific data for their purposes. There is no experimental design there. So they can't produce the rigorous results that you get from an experimental design. Really they've got a lot of correlations going on. And correlations, I don't know if you've talked about that before.

Interviewer: We have not talked about correlation.

Interviewee: Correlation, basically, is when you say that something is related to something, but you can't assume that there's a cause. They just happen to be occurring alongside each other. But because you're not doing any manipulation or any kind of scientific experimental manipulation, you are not able to say that one thing causes another. I'm trying to think of a really silly example. But I drink coffee every morning. It would be like me saying coffee drinking causes me to do something random that I happen to do every morning that I also happen to do in the morning. It's just happening alongside each other at the same time. One is not causing the other.

Interviewer: Right. You can't leap to the idea that they are interrelated.

Interviewee: Exactly, right. So there is no way that from the data that they collected that they can assume anything really.

Interviewer: Right. But even there, it's saying that a little bit in the actual article, right?

Interviewee: Yes, they did. They admit that. The authors themselves admit that. The other few things about the design and so on of the study that bothered me was that when you look at the demographic kind of stuff, you noticed that, I think, a vast majority of the parents involved were white, highly educated. They had a health insurance for at least year, and they were about 200% or 300% above the poverty level. So it's a very narrow sample in my mind. It's very specific to that population of people. Autism affects across the board. What other things? None of their media coverage, I note really, I don't think considered what happened between the original diagnosis and the loss of a diagnosis. We have no idea whether those children had intervention or what happened to them between A and B.

Interviewer: I'll tell you what has driven me nuts all week long. And I talked a little about this on this show that okay, this phone survey, they asked them a question, "Did your child get a diagnosis at some point from an expert?" "Yes." Later on the survey, they asked, "Do you think that your child currently has autism?" Around 33%, I think actually, last time I was looking it was a little bit even higher than 33%.

Interviewee: Forty percent, I think.

Interviewer: Forty percent said no. And then we get all these theories that they're like okay, well, it seems like at least a portion of them were kids who were misdiagnosed.

Interviewee: Right.

Interviewer: We have to capture that.

Interviewee: I'm going to stop you there. The first thing that just popped into my mind there is they asked the parents, "Has anyone ever told you, has a doctor or someone else told you that your child has a diagnosis?" They just went on the parent report. They didn't go in and check. I don't think these researchers went in and definitely checked that the child had officially been given a diagnosis as well. And then it's just a diagnosis generally. We don't know what proportion of the children had autism versus Asperger's versus PDD-NOS. All those things might affect comorbid response to treatment and so on. And of course, autism is such a spectrum. We know that too. We've no information about the specifics of each child's autism or groups of children's autism and so on.

Interviewer: So already, there are things that we don't know, questions that arise from this. But okay, so you have this thing and you see this correlation, whatever it is. So the media decides to take a leap. Here is where I lose my mind entirely that we know that there are studies that are out there, many studied now that have shown that kids on the autism spectrum, if they get quality ABA therapy for a certain number of hours over a certain period of time, that a percentage of those kids, a fairly high percentage, higher than even was on this survey, will lose their diagnosis. They no longer qualify for their diagnosis. But instead of leaping to what there is science-based evidence for, and it would still be a leap to say maybe and asking the question, what they leapt towards was, the media leapt and said, "They must have outgrown it. That seems the most logical explanation."

Interviewee: It's interesting because that's the first thing that they jumped to rather than looking at maybe there's a possibility that a small percentage of the children did outgrow their diagnosis, whatever that means, maybe there's another... But then there's all these other things like maybe the child had intervention, maybe they were misdiagnosed. These authors, in particular, are looking at comorbidity. So what that means is that they're looking at the children who have an autism diagnosis and a diagnosis of some other thing. So what they are thinking is maybe the children, not so much misdiagnosed, but that the children with all these other things going on, those things may have been presenting autism kind of symptoms. That is one possibility. So when the child grew older or they treated these other diagnoses, perhaps the autism symptoms went away. That's one possibility to talk about.

Interviewer: So for instance, if a child had a documented hearing issue and was diagnosed with autism, and the hearing issue got taken care of, that could explain... But again, they are totally postulating. There is no evidence.

Interviewee: It's complete speculation. The entire thing is total speculation. It's just all report data.

Interviewer: Not even report data that was intended for this particular subject.

Interviewee: Exactly. And also, it's all based on two questions. They only looked at two of the survey questions. And they're very open questions. So that is interesting to me. And then even when they are taking about the co-occurring condition, I was just saying to Shannon on the break there, even those are very vague. They talk about children who have a past speech problem and autism. And I'm thinking what does a past speech problem mean? You have to have a language delay to have a diagnosis of autism. So it's kind of confusing. Or a learning disability, or they have to have developmental delay, and those are such broad...

Interviewer: Yeah, so what does it mean?

Interviewee: Yeah, the article was kind of like what does that mean. And yeah, it's just not a very good...

Interviewer: But I got to say, what I feel better about having you come in is say that the study was not leaping to this outgrew, that really, for all intents and purposes, it's the media. And the media drives me a little bit nuts in terms of what they'll come out with. And what I worry the most about is that the takeaway for a parent, hearing what the media has been talking about all week long, and we've had people commenting on the site this week that pediatricians have said, "Oh, don't worry. Your child could outgrow this."

Interviewee: Yes. Actually, that's something that occurred to me, because I have heard that before in my practice. Actually, I've heard other things. I've had children who have recovered and worked with those children myself and then had later on, pediatricians say, "Well, they would have got better anyway. They would have outgrown it anyway, that it wasn't the intervention." And that is extremely frustrating, because we know that we don't want to leave our kids just to see what happens.

Interviewer: What a disservice to them.

Leave a Reply