When I first started reading autism blogs, I was very impressed by many of the writers on what has become the Autism Hub. Amongst them, Joseph of the Natural Variation – Autism Blog.
There are a lot of armchair epidemiologists in the autism world. Joseph has always impressed me with his careful and thorough approach.
One blog post of his is where I got the title for this post: Moving Toward a New Consensus Prevalence of 1% or Higher. That post impressed me then. It impresses me even more now.
Take a look at that blog post. It is from February of 2007. Here’s a line from his intro:
The current consensus prevalence of ASD is roughly 1 in 166 or 60 in 10,000, as widely known. I believe this is still an underestimate.
Here’s a later paragraph:
Nevertheless, it appears that when ASD is screened thoroughly in a population, or when there’s a lot of awareness and good ascertainment, prevalence is found to be closer to 1%. This is not new. The following is what Lorna Wing and David Potter said on the subject as early as 1999.
Why bring it up now? Because Joseph is (once again) proven correct. The “new” consensus is about 1% or more.
The National Survey of Children’s Health shows about 1% of children aged 2-17 currently have an Autism or Asperger Syndrome diagnosis. The rumor mill has it that the CDC will release a report with about 1% soon as well.
Anyone surprised? At the least, is anyone surprised that the “official” numbers may be going up? It has been long recognized and discussed that the “official” CDC numbers are an under estimate. The regional variations alone show that to be the case (with the autism “rate” varying by about a factor of 3 between Alabama and New Jersey).
Again, lifting liberally from Joseph’s blog post: he quotes Lorna Wing and David Potter from…1999. Yep. 10 years ago.
Because we concentrated on the children with learning disabilities (IQ under 70) we saw very few with the pattern described by Asperger. We had to wait for the study by Christopher Gillberg in Gothenberg to find out how many children with IQ of 70 and above were also in the autistic spectrum. As described above, combining the results of these two studies gave an overall prevalence rate for the whole autistic spectrum, including those with the most subtle manifestations, of 91 per 10,000 – nearly 1% of the general population.
and,
Kadesjö et al (1999) report a study in Karlstad, a Swedish town. Although this was small scale it was very intensive (over 50% of the 7 year old children seen and assessed personally by the first author). The study found a prevalence for all autistic spectrum disorders for all levels of IQ, of 1.21%!!! Children were followed up four years later and had the diagnoses confirmed.
Joseph also listed the following studies in his post:
About 1% of children in the South Thames region have an autistic spectrum disorder
and
Pervasive Developmental Disorders in Montreal, Quebec, Canada: Prevalence and Links With Immunizations by E. Fombonne, et al.. This 2006 study showed 1.076%.
More recently, we have the study by Simon Baron-Cohen’s group that showed a prevalence about 1.5%
At the time that I started reading autism blogs there was a recurring theme. Every three months the California Department of Developmental Services would publish their latest data. Rick Rollens would put out comments for the press and David Kirby would blog it, both concluding that the data showed evidence of an “epidemic”. As I recall, one quarter there were simultaneous claims of “See the autism rate went up, there’s an epidemic” and “see the autism rate went down, there’s an epidemic”. Every quarter, bloggers like Jospeh, D’oC and Prometheus (and others) would debunk the claims of epidemic.
Little has changed except that the CDDS isn’t publishing quarterly reports. Now we have more infrequent reports on autism prevalence, so we have to do the debunking less often.
The real question here is whether the prevalence of autism is really increasing. It is a very good question. It just isn’t one that we can answer with the data we have. That won’t stop people from claiming they have proof of an “epidemic”.
Left Brain Right Brain 
“It has been long recognized and discussed that the “official” CDC numbers are an under estimate. The regional variations alone show that to be the case (with the autism “rate” varying by about a factor of 3 between Alabama and New Jersey).”
I know this might be a radical idea, but perhaps the reason why the rates vary from region to region is that the rates actually do vary from region to region. The fact that the rates vary doesn’t really imply that the data is wrong in one spot and right in another.
As as for the “we told you so” implication of 1 percent I have to wonder if we will be hearing the same line a few years from now when the rates go to 1 in 50 and then to 1 in 20.
Our “understanding” of autism and our ability to recognize it seems to just keep getting better and better. At the rate our recognition is growing in another 30 years we are going to be so good at it that we will be “recognizing” it in 50 percent of the population.
MJ,
Pretty unlikely. If so, why aren’t people begging for research to be done on the low autism rates in Alabama? I’ve brought that question up many times on this blog. The simple answer is that even those claiming an epidemic don’t believe that the rate is really lower.
You do realize that the CDC doesn’t actually measure the real number of autistics, right? They count the number of people already identified as having autism and who are listed in the databases. Those are very different ways of counting. The sad fact is that we are still not accurately identifying all of our autistics.
Another consideration, there is no reason for the autism rates to vary by ethnicity within a single state. There are studies which say that there shouldn’t be a racial or ethnic difference in the US. Yet the CDC numbers do. A lot.
I don’t know of anyone predicting that. Should the measured rates (again different from the real rates) go to 1 in 20, I think everyone will be surprised and shocked.
If you don’t think our understanding of autism has changed a lot since 1980, you haven’t been reading the literature.
I remember Joseph’s post, and the idea that the average prevalence is high and stable at 1%. So no surprise there.
It’s interesting to look at the National Survey for Children’s Health site, because you can compare the questions asked in the 2007 survey with the ones from the last survey in 2003, and you can also break down the results by subgroups to see that prevalence wasn’t 1% across the board.
From one survey to the next, the question changed. In 2003, they asked whether a doctor had ever said that your child had “Autism,” and the yes response was 0.5%. In 2007 they asked whether a doctor had ever said that your child had “Autism, Asperger’s Disorder, pervasive developmental disorder, or other autism spectrum disorder” and got a yes response of 1.0%. So the most recent figure was for the spectrum of ASD, where the prior figure was not.
In the subgroups, males were 1.7% and females were 0.4%. Whites were 1.2% compared to 1.0% for Hispanic and 0.6% for Black, Multi-racial and Other. Where the primary language in the family was Spanish, 0.6% of the kids had an ASD, compared to 1.4% of English-speaking Hispanic kids and 1.0% of Non-Hispanic kids.
They also asked parents to rate the severity level of their child’s ASD. The average figure was the same – 0.5% – for “mild” and “moderate or severe.” But in the subgroups, very low income familes had 0.3% mild vs. 0.7% moderate/severe, while the highest income level had 0.6% mild vs. 0.4% severe. In the ethnic groups, Whites tend slightly toward the mild side, 0.6% mild vs. 0.5% moderate/severe. Hispanics tended overall in the other direction, but among kids from primarily Spanish-speaking households, 0.1% were mild and 0.4% were moderate/severe.
So I think these figures show that there may be underserved populations who have access problems based on income or language barriers.
Anne,
great observations. That is exactly the sort of thing I like to do–take a closer look at the details. I am chagrined to see you do it first and better!
Thanks for the heads up. When I wrote that post two years ago I anticipated that in the future certain people would use the inevitable rise in diagnosed/administrative prevalence of ASD in order to raise alarm and/or argue their pet hypotheses. Look no further than the latest David Kirby post for an example of this.
Joseph,
it is a good assumption that Mr. Kirby read your post.
If he didn’t, he is negligent. If he did, well, he is ignoring important information. Would that be new?
NJ has had it right for a while now…1% was no surprise.
The biggest surprise in the data (perhaps shock is the better word) is the rate of 0.6% of parents that identified that they “Were told child had autism, Asperger’s Disorder etc. at some point, but are not currently [on the spectrum]”
Also very interesting how AoA and Huffington posts are spinning this data faster than a particle accelerator, taking things like the above rate of ‘former autistics’ combining it with the ‘actual autistics’ rate and coming up with wild figures like 1 in 38 boys having an ASD diagnosis at one time…
That’s pretty funny. That would give Prometheus or Orac a run for their money in terms of geeky humor.
I have a paper on my desk I need to blog. One small section–losing a diagnosis isn’t new. One study showed a followup of 39 autistics where 75% had lost their autistic behavior.
That study is from the 1970’s.
I guess I could wind up my unlicensed nuclear accelerator and spin out a claim that biomed is resulting in fewer people recovering now than before.
livsparents,
here is a great example of the spin. Here’s the question asked in the survey:
David Kirby’s statement:
There is a pretty big gulf between having a health professional had autism and having an actual diagnosis. Your pediatrician could say, “I think [CHILD] has autism. Please go to early-start for an intake”. Not the same thing at all.
One should look into the idea that kids are “losing” diagnoses. But one should take this “data” with a grain of salt right now.
Too bad they didn’t ask “at what age” when asking if the parent was told the kid was autistic.
Sullivan, you said –
“Pretty unlikely. If so, why aren’t people begging for research to be done on the low autism rates in Alabama? I’ve brought that question up many times on this blog. The simple answer is that even those claiming an epidemic don’t believe that the rate is really lower.”
I didn’t say lower, I said it varied from region to region. Some areas such as NJ have a very high rate while others have a low rate. I very much think that “those claiming an epidemic” do believe that the rates are not the same in all places. And if there is a environmental component to autism then I would expect the rates to vary.
On a related note it would be very interesting if there were a comparison done between an area with very high rates and one with very low rates to determine if the children in that area were born there or moved there for services or some other reason.
“You do realize that the CDC doesn’t actually measure the real number of autistics, right?”
As far as I know no one has ever measured the amount of “autistics”. They attempt to estimate the number of people, and a lot of time just children, with a DSM diagnosis that falls under one of the PDD categories.
“They count the number of people already identified as having autism and who are listed in the databases.”
The survey that you refer to from the NSCH isn’t “counting in database” rather it was a phone survey. If you are going to try to enlighten me at least get your facts straight.
Second, do you have any reason to think that the CDC is counting people differently between its surveys? The third estimate from the CDD has not released yet but the talk seems to be that it will also show that the rate is about 1 percent now.
“Another consideration, there is no reason for the autism rates to vary by ethnicity within a single state.”
What, are you joking? Of course there are reasons to think that autism rates would vary in ethnicity. Autism is at least partly genetic and the different racial/ethnic groups do have suble differences in the genetic make up. That could easily lead one or more groups to have different rates of autism.
And then there is the whole debate around the Somali populatiuon in Minnesota – which if I am not mistaken you have written about. So how can you say that “there is no reason” ?
“I don’t know of anyone predicting that”
Kirby?
“Should the measured rates (again different from the real rates) go to 1 in 20, I think everyone will be surprised and shocked.”
Then I will quote you on that line in another 10 years or so if the rates get that high.
“What, are you joking? Of course there are reasons to think that autism rates would vary in ethnicity.”
I’ll find the citations for you shortly. Apparantly, you don’t read the literature.
Granting that autism is genetic, it would be anything but surprising if it occurs at different rates in different populations and subpopulations. It’s just a matter of how widely the genes involved has apread. As it seems probable that the genes involved in autism are recessive, an especially significant issue is how often carriers bear children together, an occurence most likely (if the genes are there at all) in small and isolated populations.
ASD prevelance rates are likely to continue to rise since the development of self reporting or parent reporting questionnaires are of dubious validity in seperating Autism Spectrum Disorders from Schizophrenia Spectrum Disorders which would include Baron-Cohen’a Autism Spectrum Quotient (ASQ) and his Childhood Asperger Screening Test (CAST). Baron-Cohen has changed the nomenclature from a screening test to his new definition, Childhood Autism Spectrum Test and is now using the dubious test as a diagnostic intrument rather than a screening instrument.
Piven’s group in North Carolina developed the Broad Autism Phenotype Questionnaire (BAPQ).
http://www.springerlink.com/content/v3k63263251k5068/
Sugihara et al, used the BAPQ in patients with schizophrenia ans schizoid pesonalty disorder and found that the BAPQ does not discriminate between Autism Spectrum Disorders and Schizophrenia Spectrum Disorders.
http://www.springerlink.com/content/m47x302w01614816/fulltext.html
Piven’s group, the authors of the BAPQ, in a later study published this month, could not distinguish High Functioning Autism from Schizophrenia.
http://www.ncbi.nlm.nih.gov/pubmed/19671209?
As Sugihara wrote:
“The broader the concept, the higher the risk of inclusion of non-specific phenotypes”.
Kanner, in 1965, also commented on increasing, but dubious, rise of incidence of autism that occured within a few decades of his original paper:
http://neurodiversity.com/library_kanner_1965.html
“This sage advice was not heeded by many authors. While the majority of the Europeans were satisfied with a sharp delineation of infantile autism as an illness sui generis, there was a tendency in this country to view it as a developmental anomaly ascribed exclusively to maternal emotional determinants. Moreover, it became a habit to dilute the original concept of infantile autism by diagnosing it in many disparate conditions which show one or another isolated symptom found as a part feature of the overall syndrome. Almost overnight, the country seemed to be populated by a multitude of autistic children, and somehow this trend became noticeable overseas as well. Mentally defective children who displayed bizarre behavior were promptly labeled autistic and, in accordance with preconceived notions, both parents were urged to undergo protracted psychotherapy in addition to treatment directed toward the defective child’s own supposedly underlying emotional problem”.
I suspect the numbers for female autistics are still off. Girls are less likely to be diagnosed with any sort of LD than boys, and those girls who do get diagnosed are generally diagnosed earlier and are more severely affected.
It’s been a well known problem in special ed circles for years. If girls really had a lower rate of LD for just about any type of learning disability, they’d probably have better school outcomes than boys. They have worse outcomes.
It may be a social issue with parents and teachers referring students for screenings at different rates, or it may be a difference with how affected girls behave and ask for help. Boys may act out more aggressively and therefore be more likely to get noticed. Girls are better taught overt social skills since their primary caregivers and teachers are generally also female. Or a combination. Or there may actually be a big difference in rates between genders. I’ve read all of those theories.
At any rate, I suspect the rates are still off. That’s all.
On the issue of autistic girls, the apparently low rate is consistent with (and evidence for) a genetic disorder linked to a recessive gene.
“On the issue of autistic girls, the apparently low rate is consistent with (and evidence for) a genetic disorder linked to a recessive gene”.
There is also a skewed male:female ratio in leprosy (3:1). Is leprosy a genetic disorder? Actually, there is a strong genetic component involved in leprosy etiolgy. Twin studies in leprosy have reported the same high (60-85%) concordance rates in identical twins and the same rapid fall off reported in fraternal twins concordance rates (5-20%) that has been reported in autism twin studies.
http://www.ncbi.nlm.nih.gov/pubmed/11279529
Heritability estimates for leprosy have been calculated as high as 80%
http://www.ncbi.nlm.nih.gov/pubmed/511131?
The cause of leprosy is infection after exposure to myobacterium laprea.
The genetic data in autism has been interpreted as evidence for a genetically transmitted disorder, but this is not evidence, it is interpretation of published data.
The genetic data published in autism cannot distinguish genetic tranmission from genetic susceptabilty.
I haven’t seen a RAJ “leprosy” comment in a long time.