America’s False Autism Epidemic

NYPOST.com COMMENTARY

The apparent epidemic of autism is in fact the latest instance of the fads that litter the history of psychiatry.

We have a strong urge to find labels for disturbing behaviors; naming things gives us an (often false) feeling that we control them. So, time and again, an obscure diagnosis suddenly comes out of nowhere to achieve great popularity. It seems temporarily to explain a lot of previously confusing behavior — but then suddenly and mysteriously returns to obscurity.

Not so long ago, autism was the rarest of diagnoses, occurring in fewer than one in 2,000 people. Now the rate has skyrocketed to 1 in 88 in America (and to a remarkable 1 in 38 in Korea). And there is no end in sight.

GETTY IMAGES

The 10th Annual Walk Now for Autism Speaks on Saturday in Pasadena, California.

Increasingly panicked, parents have become understandably vulnerable to quackery and conspiracy theories. The worst result has been a reluctance to vaccinate kids because of the thoroughly disproved and discredited suggestion that the shots can somehow cause autism.

There are also frantic (and probably futile) efforts to find environmental toxins that might be harming developing brains, explaining the sudden explosion of autism.

Anything is possible, but when rates rise this high and this fast, the best bet is always that there has been a change in diagnostic habits, not a real change in people or in the rate of illness.

So what is really going on to cause this “epidemic”?

Perhaps a third of the huge jump in rates can be explained by three factors: the much-increased public and provider awareness of autism, the much-reduced stigma associated with it and the fact that the definition of autism has been loosened to include milder cases.

Sixteen years ago, when we updated the DSM (the official manual of psych diagnoses) for the fourth edition, we expanded the definition of autism to include Aspergers. At the time, we expected this to triple the rate of diagnosed cases; instead, it has climbed 20 times higher.

That unexpected jump has three obvious causes. Most important, the diagnosis has become closely linked with eligibility for special school services.

Having the label can make the difference between being closely attended to in a class of four versus being lost in a class of 40. Kids who need special attention can often get it only if they are labeled autistic.

So the autism tent has been stretched to accommodate a wide variety of difficult learning, behavioral and social problems that certainly deserve help — but aren’t really autism. Probably as many as half of the kids labeled autistic wouldn’t really meet the DSM IV criteria if these were applied carefully.

Freeing autism from its too tight coupling with service provision would bring down its rates and end the “epidemic.” But that doesn’t mean that school services should also be reduced. The mislabeled problems are serious in their own right, and call out for help.

The second driver of the jump in diagnosis has been a remarkably active and successful consumer advocacy on autism, facilitated by the power of the Internet. This has had four big upsides: the identification of previously missed cases, better care and education for the identified cases, greatly expanded research and a huge reduction in stigma.

But there are two unfortunate downsides: Many people with the diagnosis don’t really meet the criteria for it, and the diagnosis has become so heterogeneous that it loses meaning and predictive value. This is why so many kids now outgrow their autism. They were never really autistic in the first place.

A third cause has been overstated claims coming from epidemiological research — studies of autism rates in the general population. For reasons of convenience and cost, the ratings in the studies always have to be done by lay interviewers, who aren’t trained as clinicians and so are unable to judge whether the elicited symptoms are severe and enduring enough to qualify as a mental disorder.

It’s important to understand that the rates reported in these studies are always upper limits, not true rates; they exaggerate the prevalence of autism by including people who’d be excluded by careful clinical interview. (This also explains why rates can change so quickly from year to year.)

So where do we stand, and what should we do? I am for a more careful and restricted diagnosis of autism that isn’t driven by service requirements. I am also for kids getting the school services they need.

The only way to achieve both goals is to reduce the inordinate power of the diagnosis of autism in determining who gets what educational service. Psychiatric diagnosis is devised for use in clinical settings, not educational ones. It may help contribute to educational decisions but should not determine them.

Human nature changes slowly, if at all, but the ways we label it can change fast and tend to follow fleeting fashions.

Dr. Allen Frances, now a professor emeritus at Duke University’s department of psychology, chaired the DSM IV task force.