"Autism epidemic? More likely we're just better at diagnosis"

[Orangutan]

Just came across this at Ars Technica,

I guess we kind of knew this though.

http://arstechnica.com/science/news...more-likely-were-just-better-at-diagnosis.ars

 

[AlBell]

Even more likely more and more people have figured out how to make a buck (for you furriners, that's US $1.00) off the diagnosis.

 

[WhatRoughBeast]

There is a remarkably simple explanation which accounts for much of the apparent rise in autism: serious cases used to be called mental retardation. I've lost the link I once had, and I'm too lazy to go look for it, but it is instructive to look at the combined total of cases labelled disabled. In the last ten years or so the M.R. totals have nosedived, and this reduction almost equals the growth of autism. Accounting for increased population, the combined total has increased by about 5% per year over the last 10 to 15 years, which is nothing like the rise in autism alone. There is no obvious breakthrough in preventing retardation, so changes in diagnosis seem to make sense as the cause for the decline.

 

[Beth]

Just came across this at Ars Technica,

I guess we kind of knew this though.

http://arstechnica.com/science/news...more-likely-were-just-better-at-diagnosis.ars

I'm not following their math. They said

With over 14,000 households identified, interviewers fanned out, ultimately completing over 7,000 interviews that provided a first level of screening for ADS diagnoses. From that pool, 850 individuals were chosen for a second, more in-depth interview; 630 people completed that [second interview].
...

Based on the scores obtained using the Autism Diagnostic Observation Schedule (a standard diagnostic test), the authors identified 19 adults who qualify as autistic. That's a rate of 9.8 per 1,000, or roughly the same rate as its appearance in children.

I don't see how they get the rate of 9.8 per 1,000. I get a rate of 2.7 per 1,000 assuming 7,000 as the population they got the 19 autism diagnoses. Since it was actually more than 7,000, it seems to me it should be slightly lower than that.

Any ideas on how they computed that number?

 

[blobru]

I'm not following their math. They said

I don't see how they get the rate of 9.8 per 1,000. I get a rate of 2.7 per 1,000 assuming 7,000 as the population they got the 19 autism diagnoses. Since it was actually more than 7,000, it seems to me it should be slightly lower than that.

Any ideas on how they computed that number?

Good eye. 19 autism diagnoses might be a misprint (69/7000+ = ~9.8/1000)*, I guess. *Or output by the authors' adjustment formula mentioned in passing. I don't feel like signing up to post -- any ars technica members (don't see it's been addressed yet in the blog comments)?

 

[Emet]

I'm not following their math. They said

I don't see how they get the rate of 9.8 per 1,000. I get a rate of 2.7 per 1,000 assuming 7,000 as the population they got the 19 autism diagnoses. Since it was actually more than 7,000, it seems to me it should be slightly lower than that.

Any ideas on how they computed that number?

Does this help? (perhaps you already saw it)

Epidemiology of Autism Spectrum Disorders in Adults in the Community in England
http://archpsyc.ama-assn.org/cgi/content/abstract/68/5/459

 

[Dancing David]

Even more likely more and more people have figured out how to make a buck (for you furriners, that's US $1.00) off the diagnosis.

Highly unlikely. The reason most children are diagnosed is because they have the exact same problems they would have fifty years ago, except most of the more serious kids on the spectrum would be called retarted and locked away.

 

[money]

There is a remarkably simple explanation which accounts for much of the apparent rise in autism: serious cases used to be called mental retardation. I've lost the link I once had, and I'm too lazy to go look for it, but it is instructive to look at the combined total of cases labelled disabled. In the last ten years or so the M.R. totals have nosedived, and this reduction almost equals the growth of autism. Accounting for increased population, the combined total has increased by about 5% per year over the last 10 to 15 years, which is nothing like the rise in autism alone. There is no obvious breakthrough in preventing retardation, so changes in diagnosis seem to make sense as the cause for the decline.

You might be referencing this data...

 

[Miss_Kitt]

Bruce Voigt appears to have written the Dr Bronner's soap labels. Nice to have a celebrity in here!

More seriously, I think in ASDs (autism spectrum disorders) we have a lot of changed diagnoses, some increased awareness (diagnosis of borderline cases) and possibly some minor actual increase. I would be very interested to see what work has been done looking at whether the increased number of assisted-fertility births is at all linked. Couples that could not have reproduced previously are now able to do so, with unpredictable results. There was some research in IIRC Israel that showed a correlation between the father's age at conception and ASD rate; there may be other such issues involved.

We're in a bit of a sticky spot trying to compare any developmental issues with earlier data, because we are able to have more 'high risk' babies survive. Pre-term and low birthweight or difficult delivery babies are now surviving and even thriving, so a group of children without a matching cohort in earlier decades may be skewing the data. I know that we are seeing more preterm and low-weight babies, but that most of those are due to the multiple births common to fertility assistance (either ova stimulating drugs or implanted embryos leading to more than one fetus). So we have a challenge removing confounding factors.

I have been so busy with school the last 8 months I haven't been in here much, but I'd love to hear what other folks know about this.

Cheers, MK

 

[Beth]

Good eye. 19 autism diagnoses might be a misprint (69/7000+ = ~9.8/1000)*, I guess. *Or output by the authors' adjustment formula mentioned in passing. I don't feel like signing up to post -- any ars technica members (don't see it's been addressed yet in the blog comments)?

That's a possibility I hadn't considered. Thanks.

Does this help? (perhaps you already saw it)

I did, but don't have access to the journal article itself. The abstract indicates they are using some sort of weighting formula, but the difference seems very large to be accounted for that way.

 

[Damien Evans]

Part of the reason for the increase over, say, 40 years ago is that back then the less severe ASDs, like Aspergers for example, weren't used as diagnoses until much later, 1994 in Aspergers case.

 

[dakotajudo]

Part of the reason for the increase over, say, 40 years ago is that back then the less severe ASDs, like Aspergers for example, weren't used as diagnoses until much later, 1994 in Aspergers case.

This.

See this (http://www.ncbi.nlm.nih.gov/pubmed/11439754 ) paper, or PDF here ( http://eppl604-autism-and-creativity.wmwikis.net/file/view/BaronCohenWheelwrightEtAl2001.pdf )


I took this test, scored 36, that puts me well within the high-functioning autistic range. Been talking to a counselor, it's a reasonable diagnosis. The high-functioning part simply means I've learned how compensate. If you read the paper, you'll see a long tail in the distribution of the control subjects (Fig 1), that's about were I fall in, the undiagnosed HFA.

 

[CelticRose]

Part of the reason for the increase over, say, 40 years ago is that back then the less severe ASDs, like Aspergers for example, weren't used as diagnoses until much later, 1994 in Aspergers case.

This.

I know many autistic people who were not diagnosed until they were adults because when they were children they had no academic difficulties and their social difficulties fell under the "nerdy" stereotype.

 

[Damien Evans]

This.

See this (http://www.ncbi.nlm.nih.gov/pubmed/11439754 ) paper, or PDF here ( http://eppl604-autism-and-creativity.wmwikis.net/file/view/BaronCohenWheelwrightEtAl2001.pdf )


I took this test, scored 36, that puts me well within the high-functioning autistic range. Been talking to a counselor, it's a reasonable diagnosis. The high-functioning part simply means I've learned how compensate. If you read the paper, you'll see a long tail in the distribution of the control subjects (Fig 1), that's about were I fall in, the undiagnosed HFA.

I scored 42 on the same test.

 

[TubbaBlubba]

I scored 42 on the same test.

I took the test for fun. I got 12.


On the other hand, I think I got 180/200 on "The Aspie-Quiz"... I wonder if I should go with the one made by scientists, or random people on the internet?

 

[CelticRose]

This.

See this (http://www.ncbi.nlm.nih.gov/pubmed/11439754 ) paper, or PDF here ( http://eppl604-autism-and-creativity.wmwikis.net/file/view/BaronCohenWheelwrightEtAl2001.pdf )


I took this test, scored 36, that puts me well within the high-functioning autistic range. Been talking to a counselor, it's a reasonable diagnosis. The high-functioning part simply means I've learned how compensate. If you read the paper, you'll see a long tail in the distribution of the control subjects (Fig 1), that's about were I fall in, the undiagnosed HFA.

I couldn't remember what I got on Baron-Cohen's test last time so I just retook it and got a 39.

Linky if anyone's interested: http://www.wired.com/wired/archive/9.12/aqtest.html. It's definitely flawed: several of the questions depend on stereotypes or assume all autistic people have the same special interests. Also, for many of the questions the correct answer for me would be, "No, I don't do that anymore because I have learned over the years not to, but I used to do it when I was a child or teenager." Or, "No, people don't tell me that even though they're thinking it because they're too polite."

A better one is the RDOS test: http://www.rdos.net/eng/Aspie-quiz.php. The last time I took that one I got "Your Aspie score: 156 of 200. Your neurotypical (non-autistic) score: 45 of 200. You are very likely an Aspie."

Of course, neither one of these is a substitute for a professional evaluation, which I can't get due to lack of funds.

 

[TubbaBlubba]

A better one is the RDOS test: http://www.rdos.net/eng/Aspie-quiz.php. The last time I took that one I got "Your Aspie score: 156 of 200. Your neurotypical (non-autistic) score: 45 of 200. You are very likely an Aspie."

Of course, neither one of these is a substitute for a professional evaluation, which I can't get due to lack of funds.

The Aspie-Test is freaking terrible, seriously. As I said above, I got a really high score on it; I'm definitely not in the Autism Spectrum, unless people around me and professionals have been most extraordinarily unobservant over the year.

 

[Speed of Light]

Well, My observation is that high functioning autism is being diagnosed more nowadays. It wasn't really a condition in the past.
My opinion is that there is less tolerence in society these days, and those who are 'different' in some way, are sidelined and considered to have a disabililty.
For example, whereas Asperger syndrome is thought of as a disability by 'normals', by aspergers, it is thought of as 'Wrong Planet' syndrome.

Sometimes, because of the sidelining in different areas of life, the easier option is to become diagnosed and get help.
The main problem that aspergers have, is the attitude of 'normals'.

Other things that I have observed is that asperger diagnosis has become streamlined and diluted - There is a strong parent's lobby and pressure from parents for their children to be diagnosed - and if one person won't diagnose them, they go to someone else.
I also am aware of a significant proportion of people who don't have any social difficulties, but have the problems often associated with asperger - eg anxiety, and these are now being called asperger.
Where I live, there is a new group in a hospital dedicated to 'helping' high functioning autistics - and it is in their interests to diagnose as many people as possible, so as to make it seem that the group is helping a lot of people. Such a group will never have existed in the past.
They help with thngs like 'supported employment' and benefits etc.
They also help with anxiety, depression, mental problems etc. which often accompanies asperger syndrome - but I know a number of people who they help, who are perfectly good socially. but just have mental problems.

Perhaps the saddest fact is that some people have used the 'manipulated statistics' of increased autism' as evidence to further their political aims, whatever they might be.
I cannot help observe that the natural tendency of aspergers to examine cognitively things in detail would prevent them from being taken in by such distorted pseudo-statistics, but 'normals'. like sheep, tend to follow the herd

 

[dakotajudo]

It's definitely flawed: several of the questions depend on stereotypes or assume all autistic people have the same special interests.

All tests are flawed, which is why I linked to a research report, not the test itself. The report shows how this test is validated against reference populations, and it performs quite well.

Is there a similar validation for the RDOS test? The theory ( http://www.rdos.net/eng/asperger.htm ) seems to be, well, pseudoscientific.

Also, for many of the questions the correct answer for me would be, "No, I don't do that anymore because I have learned over the years not to, but I used to do it when I was a child or teenager." Or, "No, people don't tell me that even though they're thinking it because they're too polite."

This is true for myself as well; I took the test multiple times. It does help one identify learned versus innate behaviors, at least, for me, with respect to autism spectrum.

Of course, neither one of these is a substitute for a professional evaluation, which I can't get due to lack of funds.

I don't know about that; I went to a professional to get confirmation of the test, but other than that, counseling isn't of great benefit. The counselor I see, she wasn't sure if we needed continuing sessions, since I'd seem to have learned compensating behaviors. I still go, mainly because it's useful to talk out some frustrations.

 

[CelticRose]

Of course, neither one of these is a substitute for a professional evaluation, which I can't get due to lack of funds.

I don't know about that; I went to a professional to get confirmation of the test, but other than that, counseling isn't of great benefit. The counselor I see, she wasn't sure if we needed continuing sessions, since I'd seem to have learned compensating behaviors. I still go, mainly because it's useful to talk out some frustrations.

The reasons I say that they're no substitute for a professional evaluation are a)the aforementioned flaws, and b) it's too easy to figure out how to "game" the tests and obtain a desired result, whether consciously or unconsciously. Also, a professional can observe things that the patient may be unaware of, in fact, may be incapable of being aware of if they have autism.

Also, I was referring to the diagnosis itself. Whether or not someone needs treatment is an entirely different matter.