An autism epidemic?

My post on vaccines and autism got over 16,000 hits, which dwarfs anything else I’ve said about anything.

So, in the spirit of “give the people what they want,” I offer another little tidbit that is useful for thinking about these sorts of things.

In my clinical work, people will often ask me “Why is there so much more autism these days?” And, it is this supposed increase that many anti-vaccination efforts are intended to both combat and explain.

We certainly hear more about autism. But, does this mean that there is more of it around?

One difficulty is that diagnostic criteria change—what it takes to label someone as “autistic” has not always been identical, in all places, and in all times. It is, in short, “easier” to be labeled an autistic today than it used to be.

Let’s take a more straightforward example: suppose we define hypertension (high blood pressure) as anyone with a systolic blood pressure > 150. We find a number of people in our town with high blood pressure. But, later we redefine high blood pressure to be>140–perhaps we’ve found that it remains dangerous to have pressures from 140-150 as well as over 150.

With this new definition, we diagnose everyone we caught before, plus a whole bunch of “new” people that have suddenly become sick—but only under the new definition. Nothing has changed, except the way we define the world and how we label things.

I recently read a book that summarizes this sort of thing as it relates to autism specifically:

…several recent studies suggest that the autism epidemic may be an illusion. In one investigation, researchers tracked the prevalence of autism diagnoses between 1992 and 1998 in an area of England using the same diagnostic criteria at both time points (Chakrabarti & Fombonne, 2005). Contrary to what we’d expect if there were an autism epidemic, the authors found no increase whatsoever in the prevalence of autism over time. Another study found evidence for a phenomenon termed “diagnostic substitution”: As rates of the autism diagnosis soared in the United States between 1994 and 2003, diagnoses of mental retardation and learning disabilities combined decreased at about an equal rate. This finding suggests that diagnoses of autism may be “swapping places” with other, less fashionable, diagnoses. The same trend may be unfolding in the case of diagnoses of language disorders, which have become less frequent as autism diagnoses have become more popular (Bishop, Whitehouse, Watt, & Line, 2008). All of these studies offer no support for an autism epidemic: They suggest that diagnoses of autism are skyrocketing in the absence of any genuine increase in autism’s prevalence. As a consequence, efforts to account for this epidemic by vaccines may be pointless.[1]

Both clinicians and parents may get better at choosing diagnoses, and may be influenced by other factors too. After all, it is more hopeful to label a child with “autism” (which can often be helped considerably with early and aggressive intervention) than with “mental retardation” (which may be much less amenable to improvement, depending upon the cause).

Government and schools also have an influence, since they may create certain programs or funding helps for families with autism—this gives them a real incentive to be formally diagnosed and recorded and tracked, which means they will be there when researchers go looking to determine the prevalence of autism. I certainly provide far more paperwork for people with autism now than I did ten years ago.

There’s an old medical aphorism that says, “You can’t diagnose what you don’t think of.” It’s true. And, you will also tend to diagnose more of what you think of more often.

And, we tend to think of things, or even bother looking for them, when there’s some good reason to do so.

After all, there are a couple of hundred different viruses that cause the common cold. But, I don’t have clinical tools to easily distinguish them, since I can’t do a single thing about any of them that makes a difference.

But, if there was a drug that could cure types #1-50, while doing nothing for #51-200, I’d probably look a lot harder for types#1-50. And, as a result, diagnoses would increase substantially, even though the actual prevalence and incidence of the disease doesn’t change at all.

Some of the increase in diagnosis of major depression is probably due to this sort of phenomenon: as medications that are better tolerated were developed, both doctors and patients had a reason to pay attention and to make formal diagnoses: because there was the potential of doing something.

So, we might look at the “increase” in autism diagnosis as a good thing. It could be that we’re finally recognizing a needy group that has been there all along, and we are doing so partly because we can help them. And, that recognition will lead to us being able to do even more.

And, that’s not something to panic about or try to “fix” at all. Least of all by forgoing vaccines.


[1] Scott O. Lilienfeld, Steven Jay Lynn, John Ruscio, Barry L. Beyerstein, 50 Great Myths of Popular Psychology: Shattering Widespread Misconceptions about Human Behavior (Wiley-Blackwell, 2009), 175. ISBN 9781405131124

Advertisements