"Three Kinds of Lies"

by  Mark Blaxill, MBA

Conference Presentations: Mark Blaxill, MBA

Mark Blaxill, MBA
Director, SafeMinds

Mark Blaxill is the parent of a daughter diagnosed with autism; senior partner of a major strategy consulting firm; Harvard MBA with distinction and Princeton A.B. summa cum laude; and the author of several publications on autism: including "What's Going on? The Question of Time trends in Autism"; "Reduced Mercury Levels in First Baby Haircuts of Autistic Children"; and "Thimerosal and Autism? A Plausible Hypothesis That Should Not Be Dismissed."

 great autism gene hunt.


Good morning, everyone. Andy Wakefield asked me to give a little talk this morning about population and economics, so I am going to share with you the really challenging aspects of the epidemiology of autism in the United States. There are a lot more kids with autism and it's going to impose a tremendous cost on our society, both in terms of money and human resources. And I will tell you about that in some detail and try to make all of the statistics and data as alive and relevant as your children's lives are to each of you parents in the audience. I think one of the things that's useful about conferences on autism spectrum disorders are the concrete ideas which you take home with you to help your child. And I should mention, by the way, that I am not a doctor, I'm a dad. I have a daughter, Michaela, who is nine years old. And so I always look forward to these kinds of conferences because I come away with so many concrete ideas that help me to care for her. I may not give you any medical ideas, but I hope I can sensitize you to the struggle that we face every day as parents. I think there's both a struggle to help our children, and also a desire to make a difference for the direction and course of medical research, scientific activity, and policy. At the end of the day we are all part of a much larger community. The truth about autism is an uncomfortable truth for a lot of people. And unfortunately there are a lot of lies that have emerged as people have tried to cope with the difficult aspects of these truths. And so the title of my talk today is, "Three Kinds of Lies."

I hope I can clarify some of the policy issues. As importantly, I hope to demonstrate how easily the methods and disciplines of science have been misused by epidemiologists who should know better. They seek to conceal some of the alarming facts about these very simple points because the rising population of autistic children and the costs associated with their care have become an enormous public and private liability. I chose the title for this presentation because there's a famous quote that is a wonderful way to frame these issues. I'm sure many of you have heard it--it's been attributed to Benjamin Disraeli, former prime minister of Britain: "There are three kinds of lies: lies, damn lies, and statistics." We'll talk about those. But I also like the rest of the quote, the part that is not as frequently cited. "There are some easy figures which the simplest must understand, but the astutest cannot wriggle out of." When I say I want to talk about the population and economics of the rise in autism over the decade of the 1990's, I want to talk about the lies, but I also want to talk about the numbers and ratios for which there is no wiggle room.

Let me start with one lie. While I'm going to get to some of the specifics of the numbers, I first want to lay out some of the smoke that gets blown around these issues. The first smokescreen is this fairly pernicious and pervasive argument that a lot of people present both directly and by inference. When you dig underneath what it is they're really saying, you come to understand that most of medical officialdom believes the notion that the rate of autism hasn't really increased--that we're just getting better at diagnosing these kids that have always been there. However, you need to know that the argument that autism is an ancient malady is incorrect. This misconception was put forth in a series of articles for UPI called, "The Age of Autism" written by a friend of mine. He included a completely inaccurate quote which I repeat for you here:

"Autism is ancient. Look at Newton, for example. Look at how Archimedes died -- now, that's about as autistic as you can get. Putting concerns with geometry above concerns about the social order, he died when he reprimanded a Roman soldier for stepping on his circles in the sand. Forget the recent assignment of the label, and look at the older myths. For severe regressive autism there was the changeling, a child that suddenly lost speech, the village idiot; for high-functioning autism/Asperger's there was the eccentric genius, the asocial village blacksmith, unmarried because he was unable to notice the advances of women. Forget the words, look for the traits in history. The only advance in the term autism is understanding that Einstein and the village idiot had something in common. One condition covers both extremes."
Now, if we examine this comment we can see how completely inaccurate it is. First please understand that autism is not an ancient disease. Autism was discovered by Leo Kanner and presented in some articles written in the 1940s. There were 11 children in the first of the Kanner studies, all born in the 1930s. Now looking at a much bigger picture, let me take you through a few basic epidemiological calculations. You can start with the history of reported cases of autistic children, all of them born starting in 1931. Prior to that date, autism had not been observed. Were you to pose the notion that autism is ancient, then you have to take the position that before Leo Kanner discovered autism, a certain mathematically calculable cohort of the population would have been autistic. There has actually been a study that's estimated how many people would be statistically likely to manifest this devastating illness. If you take the history of mankind up until Leo Kanner, close to 100 billion human beings had been born. And if you assume that autism is ancient, and you take the current diagnostic rate for autism of 30 per 10,000, then you can make the mathematical extrapolation that there were 300 million individuals, fully autistic individuals -- not Asperger's, not PDD-NOS -- 300 million who lived on earth before Kanner. There is no mention of these people anywhere in the history of man and literature and folklore and myth. In mid-to-late nineteenth century France, some of the finest neurologists who have ever lived explicated every abnormality they ever saw in exhaustive detail—but there was no mention of autism. We never see anything more than the faintest hint that a child exhibited behavior that could be interpreted as autistic. Where were these people? If autism is ancient, we would have overlooked 300 million people by the year 1930. I submit to you that that's an absurd concept. We should take Leo Kanner at his word that, in fact, autism was discovered in children born in the 1930s. And it's actually interesting to read what Leo Kanner wrote about autism in 1943 in his famous article, "Autistic Disturbances of Affective Contact".
"Since 1938, there have come to our attention a number of children whose condition differs so markedly and uniquely from anything reported so far that each case merits – and I hope will eventually receive—a detailed consideration of its fascinating peculiarities,,,,, These characteristics form a unique and polymorphic syndrome, not heretofore reported, which seems to be rare enough, yet is probably more frequent than is indicated by the paucity of observed cases."
Kanner's observations support the likelihood that 1931 is a base year.

Leo Kanner literally wrote the book on child psychiatry. He was one of the world's leading experts. People came from around the world to bring their children to Johns Hopkins to see Leo Kanner. He received his MD in 1921 and yet he never saw a child with autism before those children in the late 1930s. Dr. Bernard Rimland underscored this point. Bernie wrote a book in 1964, a famous book called, "Infantile Autism," in which he underscored the rarity of this condition:

"It should be understood that early infantile autism is a very rare disorder. In 1958 Kanner wrote that he had seen fewer than 150 cases in 19 years -- less than eight patients per year despite the fact that his clinic served as "a sort of diagnostic clearing house" and that the patients "recruit themselves from all over the North American continent and include some patients referred from other countries, including as far away as South Africa."
In Europe, Dr. D. A. "van Krevelen, after noting that he had become skeptical about the existence of autism because no cases had appeared in the European literature for nearly a decade following Kanner's identification of the syndrome, wrote that when an autistic child was finally brought to him, she was ‘as much like those described by Kanner as one raindrop is like another.' Van Krevelen, who was considered the European authority on autism, had only seen ten cases in the decade following the publication of his first paper." (Rimland) So the first lie to remember today is this: autism is ancient. It's not ancient, it's new.

We move on to damned lies. In the 1960s, Bruno Bettelheim was considered the world's leading expert on autism. The New York Times wrote:

"Foremost among the psychiatric experts unraveling the mysteries of autism is Bruno Bettelheim. Words cannot describe his wisdom or his compassion."
The compassion of Bruno Bettelheim was captured in this quote from 1981:
"All my life, I have been working with children whose lives were destroyed because their mothers hated them."
This is a true quotation. We have talked about the "refrigerator mother" theory as though it's kind of quaint, well, that's the great accomplishment of medicine—ultimately it's possible to move forward. It's really important to know how pervasive this theory was, how pernicious the theory was, and how wrong this "leading expert" was. Let me now digress a little bit to talk about that.

Bruno Bettelheim was a fascinating man. Actually, although Bruno Bettelheim liked to characterize himself as a Freudian-trained psychotherapist, he was really a lumber dealer in Vienna. He reinvented himself as a trained psychologist after getting out of Dachau. A few years later, he wrote an article that made his reputation. It was called Individual And Mass Behavior In Extreme Situations, and he wrote about his experience in concentration camps. Later, and based on that experience, he made a surprising connection from concentration camps to autism. This is what he wrote:

"Thus what was startling about the experience in the camps was that though the overpowering conditions were the same for many prisoners, not all succumbed. Only those showed schizophrenic-like reactions who felt they were not only helpless to deal with the new situation, but also that this was their inescapable fate. These deteriorated to near-autistic behavior when the feeling of doom penetrated so deep that it brought the added conviction of imminent death. Such men were called "moslems" in the camps and other prisoners avoided them as if in fear of contagion."
He proceeded to draw a conclusion from this that was just plain bizarre. His argument was that in Germany there were extreme situations in the concentration camps; there was an oppressor, specifically the Nazis. The victims were these "Moslems", and their schizophrenic-like behavior was a consequence of the "irrational emotion" that overwhelmed them --their fear for their lives. Bettelheim's argument was that autism was a product of the same conditions. The extreme situation was the family home. The oppressor was the mother. The victim was the autistic child. And the reason that children turned autistic was that they were similarly fearful for their lives. This is his argument which I quote below:
"Despite the incredible variety of symptoms among several hundred schizophrenic children we have worked with over the years, they all shared one thing in common; an unremitting fear for their lives...the more autistic the schizophrenic child, the more debilitating his symptoms, the greater is his mortal anxiety. Autistic children in particular not only fear constantly for their lives, they seem convinced death is imminent; that possibly it can be postponed just for moments through their not taking cognizance of life."

"I believe the initial cause of withdrawal is rather a child's correct interpretation of the negative emotions with which the most significant figures in his environment approach him. This in turn evokes rage in the child until he begins, to interpret the world in the image of his anger. All of us do that occasionally, and all children do it more than occasionally. The tragedy of children fated to become autistic is that such a view of the world happens to be correct for their world."
-Bettelheim THE EMPTY FORTRESS, 1967
This is a damned lie. And this was an awful, awful idea. It was very unfortunate. But Bruno Bettelheim was respected as one of the great experts of the time.

One interesting thing about science is that science is not as linear and orderly a process as many scientists would like you to believe. In fact, there are revolutions in scientific theory. At the height of Bettelheim's influence, the prevailing wisdom among scientists and other experts was that autism was a mysterious disorder that arose as a response to extreme situations. The main cause was the mother's hostility to the child, which in turn provoked a psychological problem that was autism. And then Bettelheim actually had an interesting idea. Instead of taking a hopeless view of the prognosis for autism, he argued for recovery.

With the rejection of Bettelheim's ideas about refrigerator mothers, autism experts converged on a different view: that autism was a specific disease entity. That it was a complex response to genetic and/or biological propensity and had nothing to do with mothers, but it was the syndrome of biology. This was, for its time, progress. However, unfortunately, once you've made that leap, you could then no longer say that recovery was possible. You had to basically sign on to the notion that autism was a lifelong disorder. So instead of having the mother as the scapegoat, it was now accepted as fact that biology and fate and destiny governed the incidence of ASD. This was the great revolution that we went through, a legacy that Bernie Rimland left in the 1960s. It took us away from the dammed lie, but it created it's own orthodoxy and it's own set of statistics and new experts, which are summarized here only briefly. But if you delve into the literature and read the prevailing scientific wisdom on autism, you will find that there are a lot of things that just plain aren't true. This is an example, written by a fellow named Edmond Cook, one of the leading autism geneticists. Just last year, he wrote an article with this claim:

"Autism is one of the most heritable complex disorders, with compelling evidence for genetic factors and little or no support for environmental influence. The estimated prevalence of autism has increased since molecular genetic studies began, owing to loosening of diagnostic criteria and, more importantly, to more complete ascertainment strategies."
I mention this because one of the things that a lot of us observe, and the medical literature now reports, is that autism is increasing exponentially. Yet the experts who have spent their careers in autism will tell you this rising epidemic is wrong--it's a lie. Why do they believe this? Actually, they have a little data. A genetically-driven disease model depends on twin studies that have shown that identical twins are more frequently reported for autism than fraternal twins. But interestingly, it's not a 100 % concordance. It's only about a 77 % concordance. And the concordance rate for fraternal twins is actually not all that low. It's higher than the siblings; it's about 11 %. The other interesting thing is that all these studies were done a long time ago. So they were looking at a very different population. Yet despite such outdated information, the prevailing scientific experts in autism depend on this single bit of data upon which they base all of their views. It's overstated and it's obsolete, yet it has yielded massive investments in research and funding of genetic research. I like to call this the great autism gene hunt. There have been ten major genomic scans from eight separate groups, a quick profile of which I've described here. And this next chart (points) is a high level summary of their findings. I won't go into all the detail. But the point is, they have found absolutely nothing, not a single significant, reproducible result from any of this work, in large part because they are looking for the wrong causal arguments in the wrong ways. Nevertheless they will tell you that their statistics are good.

This next chart summarizes some of the basic themes of an article I wrote, a review of the autism epidemiology literature which was published in public health reports last year. There is a simple set of parameters that you can set up if you want to look at the numbers. There are a lot of autism surveys, about 50, written over the last few decades. You can look at them in a simple way and compare them by doing a few simple things: use the dates of birth of the children, not the dates when the studies were published in order to avoid the problem of age-at-ascertainment bias, and focus on specific geographies, rather than random comparisons based on diagnostic criteria. Using this approach, which I then applied in the article, you can then set a nice clean apples-to-apples comparison of similar populations over time.

What comes out from that is a picture that looks like this (points). I have taken all of the studies published in the United States and United Kingdom over time, and as you can see on the chart (the yellow dots are autistic disorder, full-syndrome autism prevalence rates; the purple dots are full PDD, including PDD-NOS and Asperger's) the rates started out quite low. They then increased dramatically, ten-fold, in a very short period of time. This next chart shows the same effect in a different way with some of the data from California. This is one of my favorite charts. It's a little complex, but it deals with one of the key methodological issues: If you're comparing a six-year-old group and three-year-old group, you're going to get the result that the three-years-olds haven't always found their way into the reporting system. Quite obviously if you're going to try to measure rates of autism, you have to hold the age and diagnosis constant. And I've been able to do this with some of the California literature. You can see here that each of these individual lines is a constant age at the time the data file is cut. As in many states, in California the rates of autism have exploded from about three or four per 10,000 to about 30 or 40 per 10,000. The rates went vertical sometime in the late '80s and early '90s. It's a very disturbing trend.

Something is happening to our children. This is an important issue because this suggests that there is a crisis. This suggests that there's an environmental factor that has not been addressed. This suggests that there is something dramatically wrong with the environments in which our children have been raised. As you might surmise, there are a lot of people that want to make this go away. And there have been remarkably creative attempts to explain these increases away. There are theories of diagnostic error, that somehow we're substituting mental retardation diagnoses for autism diagnoses, or, in fact, we're just over-diagnosing--the principle that these kids just have transient developmental problems. There is an assertion--I like to call it The Lie That Autism Is Ancient--that we just changed our diagnostic standards. Or that we are now looking at smaller study populations. Or, in fact, it's just legislation. Since we introduced the Individuals With Disabilities Education Act in 1990, now all the states are looking for money. And that's why we're discovering this autism.

One of my favorite arguments is that a change in mating patterns has occurred--that there are increased rates of immigration or in fact what some call "geek" syndrome. This now has the fancy name "assortative mating," which basically claims, ladies, that as fathers we were truly unfit. And you just didn't pick up that we were autistic.

These are all testable hypotheses. And every single one of them has failed every single test. They may seem funny. They are certainly absurd. But most importantly, they are lies. And it is important that we move past them.

Why would people lie? Why do people come up with these creative ideas? I would suggest to you that we're in the middle of another revolution in science here and that we are moving away from this old and convenient set of beliefs that all the experts in the field have grown up with and have grown comfortable with, which is this notion that autism was a rare tragic disorder with constant prevalence, and that it had clear, if somewhat complicated, genetic causes. That the outcome of autism was determined, inevitably, in utero. That it's a problem that exists only in the brain, and that the only treatment one could apply is behavioral therapy.

We're learning that increases in the occurrence of autism don't fit with this model. And in fact, if you accept the fact that autism is increasing, then you have to accept an entirely new disease model. That autism is an alarming and frequent disease with rising incidence. That it's therefore an environmental disease with genetic vulnerability factors. But the outcomes result from preventable events in otherwise normal children, and that, in fact, those events result in a syndrome that spans a multitude of disciplines, including toxicology, epidemiology, neurology, immunology, and gastroenterology. That there are many opportunities for prevention, treatment, and recovery in the face of this constellation of symptoms which we don't understand well.

So we are moving from a world in which we believe that our children are defective to a world where we're acknowledging that our children are sick, and we have to do something about it.

What will this cost? I've taken a stab at some of these calculations. First we have to estimate how many children there are with autism. The CDC is declaring that one in 166 children have autism, which is true with most recent studies if you use the estimates for all autistic spectrum disorders. But you can't simply take that number and multiply one in 166 times the entire U.S. population. If you do, you come up with a number that's way too big. The increases have been much more recent and in a very specific set of rough birth numbers. If you say that in the '90s we've had four million births per year in the U.S., and you take 30 per 10,000 as a full-syndrome autism prevalence rate, 60 per 10,000 as an autistic spectrum rate, that would mean that roughly 12,000 autistic children were born each year, and 24,000 on the spectrum. If you date the epidemic, for simplicity's sake, to 1991, and multiply the annual rate by ten years or so, you get 120,000 full-syndrome kids, and a quarter million on the spectrum. That's a lot of kids.

And then, what's the total cost to society and to families of the disease? This is a very tricky number to calculate. There aren't many good studies on how much autism costs. There is actually one study from the U.K. that, I think, did a pretty thorough job of breaking out services, and out-of-pocket expenses for the families. As all of you know, autism is expensive for families. And the costs for the parents--their time, lost income for the parents giving care, not to mention the pain and suffering—are high. Over 50 % of the total costs of autism, by this estimate, are borne by the parents and the families. If you take this number of $65,000 per year -- and we all know that's going to go high or low, depending on the situation – and multiply by that population of 240,000 kids, roughly a quarter million, you get a 16 billion-dollar annual cost problem in the United States alone. Some people have estimated there's a lifetime cost. I have seen numbers ranging from two to ten million. If you just want to put a net present value on that 16 billion and bring it to a present value, that's roughly a trillion-dollar problem, plus or minus. So this is a massive problem. And even though the population may be smaller than the flawed estimates that some like to toss around, we don't need to claim that autism is ancient to persuade ourselves that this is an epidemic problem. And clearly we need to do something about it.

So that's my explanation of population and economics. Obviously it suggests that there is something that we must do as parents, which is to help foment this revolution in science.