Wednesday, May 11, 2011

Korean Prevalence study: Media hype and fodder for fringe groups?

The media has been making much lately of an autism prevalence study done in Korea in which a nearly 3% autism prevalence has been found. "Is autism more common than previously thought?" is the rhetorical question of the day.

Gadfly has now read this study in its entirety (excepting some supplementary data things that appear online). Michelle Dawson, on her TMOB comment board, as of a few days ago, reported that the study was behind a paywall and she still had not read it. I checked out the abstract and was able to download the whole study and read it yesterday, so apparently something changed in two days.

On a less superficial analysis of the study itself, it would seem the media is once again indulging in "yellow journalism" and a misrepresentation of the facts. Two separate groups were studied. They were very different. In statistics, they are what would be called a bimodal sample. There was a high probability group, who had been previously diagnosed and was receiving a variety of special education services. Another group was screened from the general population of a small area of Korea (not the entire country) and this was referred to as a general population group. They had never attended special education school, never received or requested
any services and apparently never received a diagnosis of autism, though they were all between the ages of 7-12. Of the 2.64/100 prevalence reported it was 1.89/100 in the general group but only .75/100 in the high probability group. Therefore, the high probability group had a lower prevalence than what has been found in recent CDC surveys here in the USA.

The study also found a male to female ratio in the high probability group of 5.1:1 which is fairly close to most reported figures in the literature. Interestingly, the general population group found a much lower ratio of 2.5:1 of males to females. To the best of my knowledge, no study of gender differentiation of autistics has found such a number. This includes ratios in the higher functioning groups which are often reported as 10:1 or higher, much higher than for the general population of autistics. Female neurodiversity activists have claimed, contrary to other evidence, that autism in females has been underestimated and that they are typical of autistic people, might they use the results of this study to bolster their arguments?

Blogger Harold Doherty seems to enjoy writing about the high percentage of intellectual disabilities in persons diagnosed with regular autistic disorder as opposed to all ASD's. The number of persons in the general population sample with intellectual disabilities was 16%, whereas it was 59% in the high probability group. Though apparently some neurodiversity bloggers have questioned Doherty's assertions, the results of this study would seem to give his arguments at least some validity as the ratio for autistic disorders to all ASD's in the high probability group is 2.6:1, whereas it is inversed at 1:2.6 in the separate general population group that was sampled.

Another limitation of this study was the response rate, 63% in the general population. The relatively low response rate in the general population could have easily inflated the prevalence, as parents concerned about a diagnosis would have been the ones to respond.

Missing data may have accounted for some of the results, as in 6.1% of the cases one of the two pages of the ASSQ surveying questionnaire were missing, the authors made some adjustments and this could have affected results.

Statistical adjustments or "weighting" were made to factor in the prevalence based on nonparticipants. The authors neglect to show any data for these.

The ASSQ survey and the diagnostic instruments such as the ADOS were of North American origin, yet were being used on a people of a completely different culture. Could this be a factor?

Most striking of all is the fact that after the survey was completed by the general population students, only 33% of them had a full assessment by a doctor giving them a diagnosis! This, in my opinion, really reduces the credibility of the validity of the diagnoses and consequently the prevalence numbers.

The authors claimed the students were able to perform okay in regular school in spite of their autism because of the structured Korean 12 hour school day. I doubt any of the authors were ever autistic children themselves with more than intact intelligence and had to spend 8 years in special education because of profound behavioral problems. As someone with first hand experience in this area, I dispute the authors contention.

Not surprisingly, this study was funded by that circus sideshow without a tent, autism speaks. This is an organization that gives rogue doctors who state that autism is a harmless condition half a million dollar research grants and who put a wealthy high school dropout on their scientific advisory board amongst M.D. and Ph.D. scientists. This individual, though certainly wealthier than 99.9% of all autistics, myself included, solicited and was awarded money for his own personal use from this organization. Apparently, for some reason, a study like this would be much more costly to do in the united states.

One wonders why autism speaks would fund a study of such questionable validity. They seem to want to appease to the ND anticure movement such as Mottron and his disciples as well as John Elder Robin Hood-in-reverse. They also want to seem to appeal to the "thimerosal causes autism" crowd and others who state that some agent in the environment has caused a huge autism increase in the past 25 years. Another likely reason was that in these recessionary times, funding a study which would show such high rates of autism could be used as propaganda purposes, claiming that autism is a much greater problem than realized and this would induce more funding. Not that the half million dollar grant given to Mottron or the free money given to quasi-millionaire John Robison could not be spent on something more useful like helping autistic adults find jobs or finding more brains to autopsy, or on neuron regeneration research to find a cure. I will be interested in seeing what sort of spin AS will put on this very (at least in my opinion) questionable study.

I wonder if this study will result in the start of a wild "fishing expedition" to find the environmental culprit that caused the prevalence of autism in Korea to be so high in spite of the huge differences in the two groups of children studied, and the questionable validity of the diagnoses (at least to me) of the general population children.

Will the deranged hate mongers in that ugly cult called "neurodiversity" use this as fodder to claim that there is near parity between the sexes in autism ratios? Will they use this as an excuse to claim that autism is no big deal because most of the kids in this study are doing so well?

Teen aged neurodiversity blogger Catatab Tabimount is using this study
which she apparently has never bothered to read in its entirety to bolster arguments for special education and social skills training.

Will others use this study to exploit their own political platform? Will the media continue to use this study (which I doubt any of the journalists reporting on it have read as I have) to continue hype about autism? Only time will tell.

7 comments:

usethebrainsgodgiveyou said...

I call Bullshit on the study. Not because of numbers, but because of ETHICS or lack of them.

If autism were determined to be a learning difference...and they were doing it to change the schools to make things easier I'd say fine. They are only labelling. Hell, all of us got a label, we just haven't gone to a psychiatrist yet to get it.

But there's too much dang goofy shit going on in the world of autism. What is causing a 325% yearly increase in ""autism"" diagnosis in the mental health industry? Money, perhaps?

At that trajectory, in a little over 3 years, autism should be one in 4. In my family, it is 4 of 10 the way I figure it. BFD. We all do the best we can with what we got.

JediKnight2 said...

There should be some case studies on which autistic individuals (high functioning people) are better or worse off compared to others on the spectrum, with and without co-morbid conditions and the severity of the condition(s).

The Fauxtist said...

Many people whom want to marginalize the low functioning will shout "Co-morbidity!". Its a convenient way to say "That's not real autism because they have a 'co-morbidity'" (i.e. SIB or epilepsy). The only people who refer to co-morbidity in autism tend to be the same people who follow the line of thinking that Autism is no different than Asperger Syndrome. They think that low and high functioning is something arbitrary and socially constructed.

Jake Crosby said...

"They also want to seem to appeal to the "thimerosal causes autism" crowd and others who state that some agent in the environment has caused a huge autism increase in the past 25 years."

Nope; it has no appeal to us, Jonathan.

jonathan said...

Nope; it has no appeal to us, Jonathan.

It? Don't you mean they? I was talking about autism speaks.

If autism speaks has no appeal to people who believe thimerosal causes autism, then I wonder why Sally Bernard has been on their board of directors for so many years.

Jake Crosby said...

I was talking about the Korean prevalence study, Jonathan.

Anonymous said...

I wonder how many of the children whom the Korean study says have autism actually *don't* biologically have any autism or Asperger's and *do* have parents who actively discourage them from learning social skills, a la these ones:

http://blogs.wsj.com/juggle/2008/08/01/should-you-try-to-raise-a-geeky-kid/

http://online.wsj.com/article/SB10001424052748704111504576059713528698754.html