"There are three kinds of lies: lies, damned lies, and statistics."
- Mark Twain
In America’s health care debate, Republicans seem to have the upper hand by using a tried and true PR strategy: confuse the living hell out of everyone and shout really loud while doing it. The same can be said for the ridiculously confusing debate over how many people do, and as importantly did, have autism.
Are prevalence rates of autism actually on the rise? If they are, the environment is playing a role, and the crazy parents on this site may actually be on to something. If they’re not, well, this is much ado about nothing. Autism has always been with us, let’s learn to accept our children as they are and bend society to accommodate this very large group of people who have always been with us but only recently acknowledged.
Which is it? Well, a recent “report” from the UK seems to support the position that we have just as many adults with autism as we do kids. You can read the report for yourself HERE, it’s featured prominently on the UK CDC’s website, and appears to support the idea that 1% of British adults have autism, too, so let’s all stop talking about the MMR.
My guess is that this report will do little to move the debate forward. The investigators only surveyed 7,500 people, and their definition of “adult” included anyone age 16 or older. As we all know, the rise in autism appears to have started somewhere in the early 1990s, which means “adults” between 16-20 are really part of the age of autism. Further, the study isn’t peer reviewed and the subjects with “autism” were directly interviewed over the phone but never seen in person (if they were interviewed over the phone, that means they can all talk and process questions, which is another head-scratcher.)
At any rate, I’m not here to nitpick the UK report because it appears to be absurdly useless, and I’d rather redirect you to the simple math that we can all do to look at some of the unsupportable arguments the other side uses to try to convince the world autism has always been with us at the exact same rate.
Devilish, devilish details
OK, here’s a quick quiz for you: your stock portfolio loses 75% of its value. Bummer. But, you’re a risk-taker, and you’re going to stick with stocks. So, how much does your now much smaller portfolio need to grow to get back to breakeven?
Most people, knee jerk, answer 75%. But, the right answer is 400%. Numbers can confuse.
Most people, knee jerk, have a hard time really internalizing the difference between an autism rate of 3.3 per 10,000 and an autism rate of 100 per 10,000. They know the second number is a lot bigger, but perhaps don’t internalize the practical application of this difference, so let’s consider a real-world example:
In 1987, a peer-reviewed study was published called “A Prevalence Study of Pervasive Developmental Disorders in North Dakota” which aimed to count how many kids had a PDD/autism diagnosis in the entire state. The researchers looked at all 180,000 children under the age of 18, and determined that North Dakota’s rate of autism was 3.3 per 10,000. Here’s a summary from the authors:
“Of North Dakota's 180,986 children, ages 2 through 18, 21 met DSM-III criteria for infantile autism (IA), two met criteria for childhood onset pervasive developmental disorder (COPDD), and 36 were diagnosed as having atypical pervasive developmental disorder (APDD) because they met behavioral criteria for COPDD before age 30 months but never met criteria for IA. The prevalence rates were estimated at 1.16 per 10,000 for IA, 0.11 per 10,000 for COPDD, and 1.99 per 10,000 for APDD. The combined rate for all PDD was 3.26 per 10,000 with a male to female ratio of 2.7 to 1.”
This was a thorough study. The children with an ASD were assessed in person by a doctor. The data was published in a journal. It was peer reviewed. It was replicable. They found 3.3 per 10,000 kids had autism. Could the researchers have been wrong? Was the real number actually very different? Maybe. Perhaps the real rate was as high as 5 per 10,000 or as low as 2 per 10,000. But, ballpark we are talking about 3.3 out of 10,000 kids with autism or roughly 1 in 3,300.
Today, we know autism impacts 1 in 100 kids. Based on the numbers from North Dakota, that’s 33 times more kids than in 1987. But, it’s worse than that if you think about it a different way:
In 1987, if you had a million kids, 330 would have autism. Today, if you have a million kids, 10,000 have autism.
Let me say that again. In 1987, the rate of autism prevalence meant for every one million kids, 330 had autism. With today’s number, about 33x higher, you’d have 10,000 kids with autism.
If you’re to believe the numbers from the UK, you have to believe that the research on autism prevalence done in 1987 was simply wrong. The researchers in North Dakota missed a ton of kids, and wildly under-reported the actual number of autism cases. How many kids did they miss? Well, and this is the best part of the analysis, if the North Dakota researchers found 3.3 kids per 10,000 when they should have found 100 per 10,000 kids with autism, they missed 96.7% of autism cases in North Dakota.
That means in 1987, the pediatricians, psychologist, and other forms of screeners (not to mention all the parents!) in North Dakota were missing 96.7% of kids with autism and just letting them slip through the cracks. These kids, all 96.7% of them, were sitting right next to you in class and you, and their parents and doctors, never even knew it!
Today, the argument du jour in explaining away the fact that 96.7% of kids with autism fell through the cracks in the late 1980s is that the criteria for autism have changed and broadened, thus creating more kids with a diagnosis. Of course, no one making this point tells you exactly how those criteria have changed, so it’s an effective way to pour cold water on a debate.
The North Dakota study, the one above that produced 3.3 kids per 10,000 with autism, used the DSM-III criteria for autism. Today, we use the DSM-IV criteria for autism. Is DSM-IV radically more expansive than DSM-III? Let me ask a different question: was DSM-III so narrow as to miss 96.7% of the kids who today have an autism diagnosis?
There’s only one way to know, let’s look at the actual DSM-III criteria for autism:
DSM III (1980): Diagnostic criteria for Infantile Autism
A. Onset before 30 months of age
B. Pervasive lack of responsiveness to other people (autism)
C. Gross deficits in language development
D. If speech is present, peculiar speech patterns such as immediate and delayed echolalia, metaphorical language, pronominal reversal.
E. Bizarre responses to various aspects of the environment, e.g., resistance to change, peculiar interest in or attachments to animate or inanimate objects.
F. Absence of delusions, hallucinations, loosening of associations, and incoherence as in Schizophrenia.
That’s it. That’s the DSM-III criteria for autism. Parents, what do you think? Remember, 96.7% of the kids of parents reading this site should NOT meet the criteria above (for the record my son, unfortunately, meets the 1980s criteria for autism, too…)
Anyone? Anyone with a kid with an ASD diagnosis who would have been given a clean bill of health in 1987? Remember, 96.7% of you should be out there! I’m going to take a risk here and say that those criteria sound a hell of a lot like our kids today.
I look at scientists and doctors who say autism hasn’t grown, who say it’s all expanded criteria. Then, I go look at the details, I read the old criteria. I run the simple numbers, I read the published studies. And, I say to myself (and anyone who will listen): how can you be so stupid, or so immoral, or so uninterested in the worst health tragedy of our time, and try to convince the world that everything is just fine? The numbers, and the details, scream the truth.
North Dakota does a double-check
Oh, and back to North Dakota for a second. Turns out the scientists and doctors who did that study in 1987, the one showing 3.3 kids per 10,000 with autism, they were damn serious about making sure they were accurate in their count. You see, they followed the same birth cohort, the almost 200,000 kids who made up their original study in 1987, for 12 years. They published a second study, thirteen years later in January of 2000, called “A prevalence methodology for mental illness and developmental disorders in rural and frontier settings”. What did the study conclude? Hear from the authors:
“The results of the prevalence study [the original study in 1987] were compared with the results of a 12-year ongoing surveillance of the cohort. The 12-year ongoing surveillance identified one case missed by the original prevalence study. Thus the original prevalence study methodology identified 98% of the cases of autism-pervasive developmental disorder in the population. This methodology may also be useful for studies of other developmental disorders in rural and frontier settings.”
So, these researchers went back 12 years later and checked their work. With a couple of hundred thousand kids, they found they had undercounted their original estimate of prevalence of autism in North Dakota by exactly one child. If you believe the numbers coming out of the UK, these researches should have found over 1,800 kids with autism. They didn’t, they found 59. Why? Because that’s how many kids had autism in the late 1980s, a hell of a lot less than today, that’s why!
Stop the madness
Dishonest agencies and scientist are trying to normalize the rate of autism before our very eyes, even though no supportable details exist to back them up. If autism was always here, just the way it is today, than no one can be blamed for anything. Gene research can continue. Vaccine schedules can keep growing. Why bother with recovery? Autism is just part of our society, just ask Lisa Jo Rudy.
Join me, parents, and call bullshit on these idiots, call bullshit on the attempted whitewash of what has been done to our kids.
96.7% of kids were missed? That is, in a word, impossible.