Friday, September 25, 2009

The Hidden Horde of Adults with Autism Found?

Perhaps the most important question about autism that remains unanswered is whether we are currently in the midst of an autism epidemic. Is autism more common today than it was ten, twenty, or fifty years ago, or are we just better at recognizing it?

The most recent data available suggests that the current prevalence of autism in the United States is 1 in 100 while in the UK it is thought to be 1 in 84. This stands in stark contrast to the prevalence of 4 in 10,000 that was the going rate for decades before.

The skeptics who don't believe that the autism is more common today than yesterday will tell you that there are a variety of reasons for this contrast. They will tell you that the definition of what autism has become broader, that people with autism used to be given different diagnoses, that the greater awareness of autism leads to greater numbers, and that parents are jumping at the chance to have their children labeled as autistic to take advantage of all of the great services available.

All of the reasons are true, to a point. But I don't think that they can, even if combined, take account for more than half of the rise. Just for the sake of argument, assume for a minute that all of the reasons above are completely plausible and that one percent of the population has always had autism. This leaves us with one rather large, unanswered question.

Where are all of the adults with autism?

If the estimate of 1 percent is correct, then there should be over 2 million adults living in the US with autism while in the UK there should be over four hundred thousand. You would think that it would be easy to find a group that makes up one percent of the population, yet all attempts to locate this hidden horde have failed.

That is until now. Earlier this week the results of the Adult Psychiatric Morbidity Survey were released and they purport to show that one percent of adults in the UK do in fact have autism.

The BBC has an article on the survey that says, among other things, that
...the report suggests that, despite popular perceptions, rates of autism are not increasing, with prevalence among adults in line with that among children.

It also suggests that, among adults, rates of autism remain broadly constant across age groups.

The findings do not support suggestions of a link between the MMR vaccine and the development of this condition.
So there you have it, this survey finally shows that the rate of autism is constant across age groups and that ... wait, what does this have to do with the MMR?

That doesn't sound, right. Lets look at what the survey actually found, shall we?

Hmm, they randomly selected households in England and came up with a group of 7,461 adults who agreed to take part in the survey. These adults were screened using a modified version of the Autism Quotient, and 5,329 people where selected by the AQ to move onto the second phase. Of this second set 630 people were selected to be given more comprehensive tests including the current gold standard, the ADOS.

Of this last group 19 were found to have an autism spectrum disorder.

The survey estimates that had all of the 5,329 cases been tested using the ADOS that 77 cases would have been found, but that is based on a lot of statistical extrapolations that is hard to make head or tails of. Actually, if you read the survey it is filled with a lot of talk about the methods and the statistics and how these numbers were estimated and how those numbers were made up, I mean estimated.

So based on the 19 cases, the survey forecasts that a full one percent of the adult population of England has autism. And, regardless of what is in the BBC, there is no mention of the MMR nor is there any data concerning it.

The survey found 19 members of the hidden horde in the UK, that means there are only 399,981 more members to find. How hard can that be?


  1. But I don't think that they can, even if combined, take account for more than half of the rise.

    That sounds like unfounded speculation to me.

    I do think the methodology is a little difficult to understand at first. I've tried to explain it in the comments of this post.

  2. Joseph,

    We have been over this at length and I am pretty sure you understand the reasons why I disagree with you. I understand what the methodology used was, I just don't happen to think that it produces an accurate or verifiable answer.

    It anyone was interested in the discussion I would start with the comments on this post and then move onto the comments on the one that Joseph linked to.

    As for my statement being speculation, well yes, it is. That is why I stated it as an opinion rather than as some sort of fact.

    As for being unfounded, I am sure you are aware of the research that supports that opinion, even if you don't agree with it. So calling the statement unfounded would something of a stretch. But I would agree that this question is far from answered.

  3. I am sure you are aware of the research that supports that opinion, even if you don't agree with it

    You mean the H-P et al. paper? You probably read my critiques of it. (Besides, the artifacts in that paper explain something like 68% of the rise, if I recall correctly.)

    Here's what a good paper critique should contain in my view: Mention at least one thing the authors overlooked which can easily explain why the results of the paper are what they are. This should be clear and obvious to everyone.

  4. "the artifacts in that paper explain something like 68% of the rise"

    Actually, I think the figure is 56 percent of the estimated 600 to 700 percent increase - or roughly 10 percent of the total increase.

    But that is neither here nor there, the point is that my "speculative" opinion is certainly not "unfounded".

    "Here's what a good paper critique should contain in my view"

    Thank you for the suggestion.