I recently had a conversation with a speech therapist, who told me a fair number of her patients are autistic children. I joked about how autism is a very fashionable condition nowadays, like “yuppie flu” in the 1990s, so business must be booming. It was, she said. Her practice is quite busy. However, she added that the perceived increase in autism is mostly down to a broader definition of the disorder.
I’ve long been aware that the claim that vaccines can cause autism is bogus. It was based on a single fraudulent paper, produced by a doctor-for-hire to support plaintiffs in a lawsuit. This fellow, Andrew Wakefield, has since been stripped of his medical licence, and the study has been retracted by The Lancet, which first published it.
It was also clear that it has become common to describe people who are introverted or suffer from social anxiety, as being autistic. During the dot-com boom, the media began to diagnose apparently awkward geeks like Bill Gates as suffering from Asperger’s Syndrome (a mild form of autism). Wired published a major feature, The Geek Syndrome, in 2001. Just over a decade later, New York magazine suggested everyone might be on the autism spectrum. For some intelligent but introverted people, being an “autist” or “aspie” almost became a badge of honour.
Prior to the 1988 film Rain Man, which portrayed an autistic savant, autism was far from the public mind. However, between the rise of the geeks in the 1990s and the vaccine myth of the 2000s, autism cast a large shadow across the popular consciousness. Yet it was easy to ignore as neurotic hyperbole.
Not so, the underlying data. It never even occurred to me to question the statistical rise in autism prevalence, especially among children. Many autism (and anti-vaccine) groups claim that it is an epidemic, based on statistics published by the US Centers for Disease Control (CDC), which one would think is a reliable source of data.
For the period 2000 to 2002, the CDC found one in 150 children had an autism spectrum disorder. This rose to one in 110 in 2006, one in 88 in 2012, and one in 68 last year. That’s not just a minor trend. That does look like an epidemic. If it is true.
But it looks like it is not true. Prevalence (the number of people with a condition) is not the same as incidence (the number of new cases per capita per year). An epidemic is a rise in incidence, not just in measured prevalence. Worse, I took an autism test and passed with flying colours. It looks like I’m an autist myself.
Once you go digging, trying to make sense out of a host of contradictory views, the worst one can say about autism incidence is that the data is not good enough. There is little evidence that autism is really becoming more common, and there is some good evidence that it is not.
The idea of “fashionable diseases” is not a new development born of mass media. The Irish writer and lyricist Thomas Moore, in his journal, records a conversation between the second Marquess of Sligo and Lord Byron, the poet:
“I look pale – I should like to die of a consumption” – “Why?” – “Because the ladies would all say, ‘Look at that poor Byron, how interesting he looks in dying.’ ”
The fact that consumption was fashionable did trivialise it, in many ways. Tuberculosis, as it later became known, was then a very real and dangerous disease, as it remains in many countries today. Yet a romantic idea of consumption, gout or melancholy was very modish among the posturing nobility of the 19th century. An academic project on fashionable diseases notes that much of high society at the time was centred upon the spas, resorts and sanatoriums of Europe, to which the right chronic illness was an entry ticket.
The appeal of consumption to the upper classes may have been callous, considering how common and deadly it was among the working poor, but was it really any different to the “heroin chic”, “goth” and “edgy” affectations that remain popular to this day?
Fashionable diseases also play on fear, which motivates people to buy cures. The allure of selling profitable potions and liniments, or running a luxury spa and sanatorium, provides a commercial reason to perpetuate scary claims about vague diseases that appear to be common.
It isn’t all a conspiracy by the medical industry, however. In cases like autism, there are other strong motives to obtain positive diagnoses. For those who enjoy either state-funded medical care or private health insurance, an autism diagnosis provides access to special care for a child who appears to have educational, social or language difficulties. And that incentive is not only financial. Whether or not a child suffers from a disorder can make the difference between judgemental rejection and sympathetic acceptance.
The causes of autism are not well-understood. It appears that as much as 90% of it is genetic, though it is unclear how much is hereditary and how much is caused by spontaneous mutations in DNA. Whether any environmental factors trigger or worsen the disorder is also still unknown.
There is considerable evidence that broader definitions and higher awareness account for a large part of the rise in diagnoses, however.
A recent study in Denmark tried to assess the impact of changes in Danish reporting practices in 1995, as well as the publication of the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) in 1994.
When the DSM-IV came out it broadened the definition of autism by introducing three distinct conditions, namely Asperger’s Syndrome, autism, and the awkwardly-named “pervasive developmental disorder not otherwise specified”. All were considered developmental in nature, which explains in large part why children are diagnosed more often than adults.
The Danish study found that 60% of the increase in autism diagnoses can be attributed to broader diagnostic criteria introduced by the DSM-IV in 1994 or the inclusion of outpatient diagnoses in national statistics. If you count more cases, you’ll see more cases.
A British study assessed adults who were diagnosed with a developmental language disorder in 1980, and found that under the criteria of a quarter-century later, 25% of them would have been diagnosed as autistic. It was already clear that the DSM-IV led to an increase in diagnosing high-functioning individuals, but a study among intellectually impaired people in the American state of Utah found that almost 60% of those diagnosed as not autistic under DSM-III in 1980 would have met autism spectrum criteria in 2012, and most of the rest displayed some signs of autism.
The latest edition of the DSM, the DSM-5, was published in 2013. Besides changing the version number from Roman to Arabic numerals, they also changed the autism diagnostic criteria again. The intention is that anyone who was diagnosed with pervasive developmental disorders under DSM-IV will be diagnosed as having either autism-spectrum disorder or social communication disorder under the DSM-5. It is unclear whether the diagnostic sensitivity for autism will increase or decrease.
The DSM is one of two standard diagnostic references, the other being the International Classification of Diseases published by the World Health Organisation. It is due for an update to edition 11 this year, and while it includes nine disorders under the category pervasive developmental disorders, it does not appear to differ substantially from the DSM with respect to autism.
Definition changes make it difficult to compare historical datasets. It can be impossible to tell how much of an observed change is caused by higher awareness, different measurement, or different definitions.
There is also evidence that casts the quality of psychiatric survey data into doubt. Such surveys are usually based on self-reporting, or reporting by parents on their children, and can be inaccurate for many reasons.
Time magazine once quoted an autism epidemiologist who noted that since autism is by definition a lifelong condition, the fact that 40% of parents who claim their children have had an autism-spectrum diagnosis also say that the child no longer meets the criteria, “suggests that there may be over-reporting”.
Indeed, it does, and it suggests over-reporting by 67%, which would account for most of the rise in the CDC’s prevalence statistics.
Autism is usually studied among children, because it is a developmental disease and children are typically supervised by at least a parent or teacher. Adults are not as easy to study. The first survey which tried to assess the prevalence of autism among adults found no difference between younger and older adults. This implies, necessarily, that there has been no increase in incidence. There is no “autism epidemic”.
During my research, I completed an autism-spectrum disorder self-assessment, designed for adults with normal intelligence. Such questionnaires are commonly used in psychiatry to evaluate patients. My score was comfortably higher than the threshold that indicates potential Asperger syndrome or high-functioning autism.
Which is kinda awesome, because I’m a geek, and I want my badge of honour too. But it is also surprising, because I don’t flap my hands, and it is somewhat alarming. Yet despite this result, I can’t claim to be autistic, because such a self-assessment is, at best, a screening instrument. It is not meant to be diagnostic, and I fully intend to compromise statistics about autism by not seeking a formal diagnosis.
The next time you bump into someone who claims to have a child on the autistic spectrum, or who wonders where all the autistic people come from, remember that there is no evidence of an epidemic to panic about, and you might be autistic too.
Because there is no epidemic, grasping for a putative cause of a rise in autism is futile. It would be nice to understand the causes of autism, but it isn’t something evil foisted upon us by modernity.
As was the case with consumption in the 19th century, some disorders become fashionable, especially when they have broad definitions and vague symptoms. This trivialises the condition. In raising public awareness, it is possible that we have only managed to stir up unwarranted fear, and to direct attention, research and funding away from those who most need it.
Of course, I’ll stop going on about it if you send me tickets to a luxury spa where I can go and while away a lazy season or two. DM