Sunday, February 21, 2010

When does autism develop?

One of the many unanswered questions about autism is whether autism is something that a person is born with or whether it is something that happens or develops later. There are currently a few theories about how autism is thought to develop -

Early onset - A child is born with autism and shows subtle signs of it almost from birth. These differences grow over time until the child is noticeably different from "typical" children. It is thought that most people with autism fall into this category.

Regression - A child appears to be developing normally for the first year or two but then has a period where they regress and lose previously acquired skills and become different from other children.

Plateau - A child develops normally, without any early signs of autism or regressions, until they hit a certain point. After this point, they simply simply fail to develop new skills and stop developing normally (i.e. hits a plateau in their development)

That's where a new study, published online this month in the Journal of the American Academy of Child & Adolescent Psychiatry, comes in. This is one of the first studies to follow children and watch which ones develop autism rather than look at which ones developed autism after the fact.

The text of the study is freely available, so if you are interested in the topic I suggest you read it for yourself.

The children included in this study were selected from a larger, ongoing study of infant siblings of children with autism. For this study, researchers selected 25 children who developed autism and 25 that did not. These children came either from a high risk group (sibling had autism) or a low risk group (no siblings with any developmental disorders). Of the 25 children with autism, 22 came from from the high risk group.

Each of these children was evaluated at 6, 12, 18, 24, and 36 months using a variety of tests. The results of the tests were compared to each other and found to be in agreement. Overall, the researchers did a good job getting an accurate picture of how these children developed and what skills they had at each point in time.

The results of the study were interesting, to say the least.

The most important finding was that the children who went on to develop autism were no different from the other children at 6 months. And even more than that, these children seemed to be slightly more social than the other children at 6 months (although this result is not scientifically significant).

For the next 12+ months, the development of children in the autism group slowed down and they failed to keep up with the development of their peers. But more importantly, most of the children showed loss of skills - as in they regressed. In what could be considered a silver lining, the children did not show losses of cognitive ability or language skills as the losses appeared to be of social communication skills area.

The second important part of this study is that most of the parents (83%) did not pick up on these regressions. Keep in mind that these are mostly parent who already had a child with autism and, as such, would be familiar with what autism looks like. It is likely that they would be watching very closely for signs of autism in their younger children. This finding, if confirmed, means that parents frequently miss subtle regressions and signs of autism in their children.

The result of these findings is that our models of how autism develop may be wrong. Autism may not be present from birth in the majority of cases and might start developing sooner than we think and involve more regressions.

The researchers proposed a new theory whereby the symptoms of autism are not prevent at birth but rather appear over a length of time. Under this theory, most children have regressions but it is the timing and severity of the regressions that give the impressions of different models of onset. If the regressions happen early enough they are much harder do detect and make it seem like the symptoms have always been there. If they happen late enough, the results are much more drastic and noticeable and make it seem like it appeared out of the blue.

I think it is important to remember that this is just one study, and a smaller one at that. While the researchers appear to have been very careful these results are still going to need to be confirmed.

But, if the researchers are correct, autism may develop differently than we think.

A Prospective Study of the Emergence of Early Behavioral Signs of Autism
Sally Ozonoff, Ana-Maria Iosif, Fam Baguio, Ian C. Cook, Monique Moore Hill, Ted Hutman, Sally J. Rogers, Agata Rozga, Sarabjit Sangha, Marian Sigman, Mary Beth Steinfeld, Gregory S. Young
Journal of the American Academy of Child and Adolescent Psychiatry 8 February 2010 (DOI: 10.1016/j.jaac.2009.11.009)


  1. This is so confusing
    I don't understand

  2. Hi anonymous,

    Which part did you find to be confusing? The basic point of this bit of research is that the symptoms of autism might not be present at birth but rather develop later. So, the different ways that autism seems to develop might really all be variations of the same type.

    If the symptoms start early enough they seem like they were always there. If they develop a little later you get the appearance of a plateau. After that you get the regression.

    But the key point in this research is that all of the forms involve a loss of previously acquired skills.

    Hope that helps.

  3. I agree there were no descriptions of symptoms or anything!

  4. Can symptoms of Autism change over time? My son was diagnosed with PDD NOS when he was 5. At that time he was bright except his language was impaired. then he got the ADHD label because he showed age-appropriate social skills but he was severely distracted in school and he showed social anxiety. Now he is 10 and he is back on PDD NOS label because his social and communication are extremely underdeveloped. I watch him slip slowly through the cracks of the public school system. They gave him the ASD label and then removed it and now he is in a classroom for kids with learning disabilities. he still can't read independently, he still struggles with computation. All he learns is how to prepare to learn domething and the school year will be over. After reading your article, I am certain that my son is the plateau ASD type. Its depressing to be in this category. By the time you find some kind of closure and make peace with your child's dx, his childhood will be over.
    I am just left with a mountain of guilt and pointless hindsight(woulda shoulda coulda). All the support system is gone. the school district must have patted themselves on the back for conning me out of the Autism program and saving all that money.

    1. Hi Anonymous,

      I'm sorry to hear that you have had a hard time getting an appropriate diagnosis for your son. The short answer to your questions is yes the symptoms of autism can (and often do) change over time.

      I can't point to any specific research on the subject but it is my understanding that the exact symptoms and severity of the symptoms will change over time because of many different factors such as stress level, changes in the person's environment, hormonal changes, social changes, as well as many other factors. I know from watching my own children get older that their behaviors come and go over time.

      I don't know whether they can change enough to go back and forth between ADHD and PDD NOS. There can be a lot of overlap between ADHD and autism and it is possible to have both conditions at the same time.

      Regardless, I can really understand the guilt and the endless games of what if you can play because my wife and I do the same thing. We are constantly wondering were our children would be at if we had done this differently or not did this or did that sooner. All parents whose children have disabilities do that to some extent, although it does seem to be worse in conditions like autism where the outcome is so variable.

      The way that I look at it is this - you do the best you can with what you know at the time and that is really all you can do.

      You can always look back and kick yourself for what you could have done differently but, at the end of the day, there is absolutely no way to know if your child would be in a better place if you had done something differently. They might be in a better place, they might be in a worse place, or they might be exactly where they are now. So you do the best that you can with what you know and just move on.

    2. You are so eloquent with compassion, which is very beautiful. In the autism arena, you run into so much ignorance, anger and blame. I wish more parents saw it this way, as well. Thank you.

  5. I found this article rather interesting. I've been researching the effects of vaccines on humans and their effects. I came across a hepatitis A strand called simian virus 40 and it has contaminated the poli vaccine which millions of people are given. It is one of the many problems with vaccines and is know to causing cancer. It has oncoproteins that mutate genetic material and cause them to go out of control. I was wondering if there is a oncoprotein that may set off certain phyontypes to develop into autism, or anyother disorder.

  6. Wild suggestion, but perhaps the slightly better (but insignificantly so) "social skills" at 6mo were influenced by some kids not showing the same fear response to strangers that most babies that age do? 5-6 mo is the time when babies suddenly become very attached to mum and fearful of new faces, probably not a coincidence that it is when they may start to crawl and be more independent (keep them safe ...). Is it also possible that the "social skills" that "regress" in the autistic babies after 6mo were more like pre-programmed routines designed to elicit caregiving behaviour? It is a hell of a long time since I studied developmental psychology - I know the level of genuine consciousness in very young babies has been a point of debate for a long time. I used to be aware of (then) current thinking and the different stages and so on, I expect much has changed by now. Anyone got anything to shed any light on that?