Dr. Catherine Lord is widely viewed as the world’s foremost expert on autism diagnosis. The Department of Justice lawyer who examined Dr. Lord in the thimerosal portion of the Omnibus hearing spent what seemed to me to be a very long time just going over Dr. Lord’s credentials and accomplishments as they are considerable.
I was listening to the recording of her testimony yesterday, again, and was reminded of how much I had learned from listening to Dr. Lord’s testimony about what is now known about the early months and years of autistic children. I already knew the basics of what she was explaining but there were some fascinating details that I didn’t know. Links to two audio clips that contain the following testimony are found at the end of this blog entry. Once again, I did the transcribing of the audio, and once again I’m guessing the Dept. of Justice lawyer is Ms. Ricciardella:
Ms. Ricciardella: Does any of your research or research of others support a distinct subtype of regressive autism?
Catherine Lord: No. I mean as especially as we have looked at the toddlers… it’s clear that even, even these very large studies where we felt like we were asking parents many, many questions in great detail probably do not get at what the essence of what happens in those early months because the changes are more subtle and our ability to observe them is so much dependent on the context. It dependent on when do you see a child and what are you looking for. So I think that that has that has moved us, and I think much of the field, toward a sense that there isn’t a regression or not a regression the question is the degree and type of worsening that occurs, how long it lasts and how many skills a child has before that occurs.
Ricciardella: Now in terms of the clinical outcome of a 5 or 6 year old with autism. Is there any marked difference in the clinical outcome of a child who had what I’ll term “early onset autism” versus a child who did indeed have regression.
Lord: Most studies have found no difference at all. The studies that have found differences have found these relatively small differences in verbal skills.
Ricciardella: Now you touched on earlier doctor that you are continuing to research the phenomenon of regression is that correct?
Lord: That’s right
Ricciardella: And you are conducting a longitudinal study. Is that correct?
Lord: That’s right.
Ricciardella: And what information is emerging from that study with regard to regression?
Lord: With that study we have been doing is seeing children who are at risk for having autism, either because they have a sibling with autism, so they may not have any behaviors associated with autism but they have a sibling and their parents are eager to have somebody follow them, or something has occurred, or something has has been seen, often identified by parents, but sometimes by a pediatrician, for example the child has had seizures in the first year of life and so someone is concerned that this child might develop autism.
And we see the children once a month, have parents fill out the same forms each month and then we do standardized assessment, a toddler version of the ADOS. So we do a standardized observation of the child’s social behavior with us and with the parents every month.What has come out of this is that the trajectories are much less clear than we would have thought from retrospective descriptions years later of what the children are like, and when we have tried to sort that out, I think that there a number of implications. One is that different skills are changing at different rates and at different times. So that you have, for example, eye contact is typically getting worse for almost all of the children for from 12 month to 24 months. And social engagement responsiveness to someone trying to get the child to interact with them typically is getting worse in children who have autism diagnoses say by the time they are 2 1/2.
So those things are changing but they actually cycle back around, so they get worse for a while and then for some children they start getting better again.
We also have other skills, for example response to joint attention, or response to somebody pointing, or trying to get the child to look at something, and that for a number of kids gradually gets better even at the same time that some of these social skills are getting worse.
So I think what we’ve realized is it’s just much more complicated changes in development than we thought.
And that these things we used to think only happened in kids who had regressions are actually happening in almost everybody who has autism, because there are some children who look very different from typical children at 12 months, but those are few and far between, and in fact in our follow up study that is not necessarily predictive.
The kids who are not making eye contact at 12 months are not the most autistic kids at age 3. So many things change during that toddler period and I think our conceptualizations of what regression is are partly based retroactive trying to figure out what happened and didn’t happen, which is quite different than when we can see it happening right before our very eyes.
…
(the second audio clip)
Ricciardella: Doctor are you are aware of any evidence showing that the etiology of regression in autism is different from that in “non-regression” for lack of a better word.Lord: No. And I think again that the idea that there aren’t these clear patterns makes it much harder to draw conclusions about etiology. Because basically could arbitrarily divide these kids up in millions of different ways. So far, … people have tried to divide them up and haven’t found differences in etiology. But it’s not even clear that we know how to divide them up, or that they can be divided up.
Ms. Ricciardella: Doctor before this litigation had you ever read in any published literature that thimerosal containing vaccines cause regressive autism only.
Dr. Lord: I had not.
Ricciardella: Are you aware of any study that has ever suggested that hypothesis?
Dr. Lord: No.
Ricciardella: Doctor did you review the report submitted by Dr. Marcel Kinsbourne in this litigation?
Lord: Yes.
Ricciardella: On page 14 of his report, he states that the, “late onset of the regressive subtype and the subsequent remission or relapses become more understandable if autism is due to disease than if it is the aftermath of congenital maldevelopment.” Do you agree with this statement?
Lord: No
Ricciardella: Why not?
Lord: There are many different disorders where the onset occurs later on. We have Huntington’s disease and schizophrenia and sickle cell anemia and all kinds of disorders… where in some cases we know are genetic, but which occur later on. So I think we can’t make a simple inference that because something emerges later that means that somehow someone has caught a disease or had some kind of particular environmental event that caused it.
Ricciardella: Dr. Kinsbourne also draws a distinction between what he terms as classical or congenital autism and regressive autism. Is this a proper distinction?
Lord: I think the term congenital autism means nothing. Because, I mean, as I said it’s a developmental process. We can’t diagnose autism in a brand new baby. And so in all cases something is developing that would lead us into autism. So to make this distinction between congenital and regressive is a false dichotomy.
Ricciardella: … Dr. Kinsbourne has also describe what he terms his “over-arousal model” as an explanation for autistic behavior. Does his over-arousal model accurately describe what is known about autistic behavior?
Lord: I don’t believe so, I mean the over-arousal model has been around for 40 or 50 years and used to describe many different disorders. I think one of the hard things is that it’s becomes very circular. And children with autism do respond to being over-stimulated as do many other kids, and children with autism may respond in more conspicuous ways, and may have a lower threshold. But the problem is that often the behaviors are used to say that a child is responding by over-arousal, for example by flapping or getting very physically excited, or distracted, are the same behaviors that occur when the child is under-aroused. You know, we can get children who have a lot of self-stimulatory behaviors to do these behaviors by putting them in a situation where there’s nothing to do. We also see children do these behaviors when they are very happy or when they are not so happy. So that the behaviors that used to define over-arousal are behaviors that occur in many different contexts.
Ricciardella: Thank you. That’s all I have.
…
PSC Lawyer, Tom Powers: … I do have some questions to ask you as you might imagine based on the report you filed and the testimony you gave today. …Your testimony … is that there’s no phenotype for regressive autism. … Regression within autism is not a distinct phenotype with in autism spectrum disorder, is that correct?
Lord: Yes.
Powers: You’ve also describe regression in autistic children as a striking phenomenon. … How would you describe the difference between a phenotype and striking phenomenon?
Lord: My point about the striking phenomenon is that it is a remarkable experience to watch a child who has been able to do things not be able to do those things. Or to watch a child who has been relatively socially engaged become less engaged and be more and more difficult to engage or attract. But I think that is different from a phenotype because a phenotype implies that there are a cluster of behaviors that are associated with each other, and that there’s something unique about that cluster of behaviors. I think regression is a real phenomenon in autism, but there’s a continuum of regression. … And we can create a phenotype, I can say, well I’m only putting kids who lost words into this group and I’m going to call it the Lord phenotype. But there has been no, nobody has been able to show that that phenotype is associated with anything other than the characteristics which I used to define the phenotype.
Powers: And that would be because, as I understand it, is because autism diagnostically is entirely a symptomatic diagnosis that is there’s not a biomarker … is that correct?
Lord: It’s not, the problems with defining the phenotype aren’t because autism is defined by purely by behavior, it’s because we haven’t been able to find an association between any of these particular phenotypes that people have pulled out, and the ways in which people have pulled out the phenotype.
I realize now that if my own (now adult) ASD child been a part of Dr. Lord’s baby siblings study that they would have been able to document a “regression” because, basically all autistic children regress, depending on how strictly you define “regression”. I remember my ASD child as an infant apparently losing the ability to respond to the sound of his/her name, and then regaining that ability. Looking back, I doubt that I could isolate the dates when this “regression” started and when it ended, since is was just aggravating to me at the time and not something I brought to the attention of our family doctor.
(Edited to fix some errors my transcribing. Also, I just listened to the second clip I uploaded to boomp3.com and realized that it is a little shorter than I thought.)
Click here for the first clip.
Click here for the second clip.
For some reason I can only embed one “player” and it must be at the very end of the post. This plays the second clip:
http://static.boomp3.com/player.swf?song=by6i9hwl7_0<a style=”font-size: 9px; color: #ccc; letter-spacing: -1px; text-decoration: none” target=”_blank” href=”http://boomp3.com/listen/by6i9hwl7_0/lord-2″>boomp3.com</a>
Left Brain Right Brain 
I wonder what Catherine Lord knows about paternal age and non-familial autism.
Mr. Feldman, I don’t know. I suppose you could do a pubmed search on something like, “risk factors for autism,” plus her name.
I’m guessing you also know about the purported risk factor for having a grandmother who was comparatively older when she gave birth to a mother of an autistic boy?
I wouldn’t want older men to avoid fathering children out of fear of having an autistic child. Autistic children are as wonderful as non-autistic children.
I forgot to give the link to the site where you can get all of Dr. Lord’s testimony on mp3 (audio recording)
ftp://autism.uscfc.uscourts.gov/autism/thimerosal.html
Her testimony was on “Day 13”, she’s on the first mp3 file for that day, May 28th.
But it’s not even clear that we know how to divide them up, or that they can be divided up.
That’s good to hear from a top expert.
Thanks Ms Clark for your work transcribing the omnibus proceedings.
I also notice in my son (still under 5 years old) this cycle of losing/regaining skills, improving in some areas and regressing in others. For instance, I see that as he gains more language, his rigidity gets worse because he feels he can have more control over things and gets frustrated if he doesn’t have his way. His social skills go up and down on the course of a week or a day. And he only started hand-flapping a couple of months ago, although most of his self-stimulatory behaviors have decreased substantially in the last year.
I’m so glad that Dr. Lord and others are studying early development in kids at risk for autism, because this information is sorely needed, not just for educational and treatment purposes, but also so that parents can have some idea of what they can expect. More information should reduce the uncertainty and fear that people understandably feel when their kids are first diagnosed.
Dr. Lord even briefly discussed the fact that there are later “regressions” beyond the preschool years, but that the ones folks are most interested in are the first ones in the 1 to 3 year olds.
There’s another segment of her testimony I want to transcribe. I’ll try to get at it in the next couple of days if I can.
One thing Dr. Lord has said elsewhere is that it’s common for kids who get very expert diagnoses of “autistic disorder” or “pdd,nos” to switch categories a couple of years later, so that even the most expert diagnosticians find that a kid meets criteria for one disorder and later for another one.
Many parents will claim that a treatment was what caused a child to move to a “less severe” category, but apparently, it’s not uncommon for kids to do that, independent of which therapy they get.
I would suppose that many undiagnosed kids start off looking “autistic” and then end up looking not autistic or maybe mildly quirky without any therapy at all.
A study of lower income kids in rural California might show that. There are tons of undiagnosed toddlers in California (and probably in rural areas of most of the rest of the United States), I think they usually get a diagnosis if they are still plainly autistic by school age, but as toddlers they are not diagnosed and get no services.
If children with autism go through a continum of regression, what are the implications regarding intensive early intervention, particularly related to theories about intervening in the first years?
I think most of what is offered as “early intervention” is without any scientific underpinnings. So as more is learned about the differences in developmental trajectories that autistic children have, and the differences of developmental trajectories that other developmentally different but not-autistic children have… and as they learn even more about the developmental differences of apparently normal children then they can make statements about what it is that autistic children need that is different from what normal children need.
For now what is provided as “early intervention” is usually someone’s idea and that someone has the ability to market their idea and because of that… that is what is provided. In different areas of the country different services are provided, so I understand, based on the providers in that area and what people are willing to pay for.