Monday, November 05, 2007

No Real Autism Increase? Any Evidence To Back That Up?

The latest bandwagon to roll across the autism terrain is the argument, already presented by some as fact, that the stunning increases in estimated rates of autism are due entirely to expanded definitions of autism and the availability of special education monies for those with an autism label. On the latter point it is argued that doctors are more inclined today to diagnose autism because an autism label will trigger special education resources in the United States.

I am willing to accept, without substantial evidence, that the changes to the DSM-IV, in 1994, would have had a significant impact on the numbers of autism diagnoses, particularly in the first few years after the 1994 changes. The mere expansion of diagnostic criteria certainly seems on its face to logically indicate that more people would be caught by the diagnosis. And as an active participant in autism discussions, and advocate for my son who has a classic Autism Disorder diagnosis, I have argued that there are many high functioning autistic persons purporting to speak on behalf of "autistics' who bear little resemblance in their intellectual and communication skills to my son.

The incentive to diagnose autism argument might also have merit to some extent although I am not sure how much of a factor this would be in the increases. There would also be strong reasons why doctors and parents want not want to place an inaccurate label on their children or patients. An erroneous diagnosis could result in inappropriate treatment.

But where is the evidence that the massive increases in recent years is due solely to these perceptual or social factors rather than to actual increases? Many of the diagnostic changes took place in 1994. I have been an active autism advocate since 1999. At that time, 5 years after the 1994 DSM changes, the prevailing figure was 1 in 500 persons on the "autism spectrum". Soon it changed to 1 in 250. Just 2-3 years ago the figure was lowered to 1 in 166 and only in the last year to 1 in 150 - in the United States as a whole. In New Jersey the estimated rate is 1 in 98 and in the United Kingdom 1 in 100.

Do the 1994 changes account for a stunning rise in estimated autism diagnoses from 1 in 500 to 1 in 150 in the US as a whole, 1 in 100 in the UK and 1 in 98 in the State of New Jersey? These are very large increases. Where is the evidence that such increases are entirely attributable to the 1994 diagnostic changes? And why are there current geographic discrepancies.

Furthermore what about the unified theory of autism? Genetics has long been known to play a very large role in autism causality. But there are also theories which look to both genetic and environmental factors of autism causality. If the environment plays a role, in some cases at least, in triggering autism, is it safe to assume that the stunning increases of recent years are entirely attributable to the 1994 diagnostic criteria changes?

Where is the evidence?


Anonymous said...

Yes, plenty. For example, see this and this by me, this by prometheus, and of course Gernsbacher et al..

One would have to be pretty closed minded to not consider the likely possibility that we are not experiencing an autism epidemic given this evidence.

Unknown said...


If you want to comment on a subject here make your comments here please. Don't direct me to other sites.

I am not closed minded at all. I acknowledged the obvious already in my comment. The expansion of the diagnostic definitions and criteria in 1994 undoubtedly expanded the numbers of people considered autistic, including most in the Neurodiversity movement.

But that does not necessarily explain ALL the increases from 1 in 500 in 1999 to 1 in 250 to 1 in 166 to 1 in 150 and in the UK 1 in 100 and 1 in 98 in NJ. What is the evidence that ALL the increase even that showing up after 1999 is due to the 1994 changes?

Anonymous said...

What is the evidence that ALL the increase even that showing up after 1999 is due to the 1994 changes?

A exact comparison clearly cannot be made, because there are no epidemiological studies that are comparable that way. In IDEA, however, if you look at 8 and 9 year olds, the whole population of children in that cohort is stable. So that in itself suggests nothing unusual has ocurred.

It doesn't matter how much administrative prevalence has increased, if all we know is that it is a lower bound of actual prevalence.