April 11, 2006
So Much for the Autism Epidemic
Orac at Respectful of Insolence points to some new research that strongly suggests that the Autism "epidemic" either doesn't exist or is much smaller than many have claimed.
The big problem is something called "diagnostic substitution." In special education programs, "autism" was not a required category until created by the Americans with Disabilities Act of 1990. Suddenly, "autism" diagnoses for special education sky rocketed. At the same time things like "mental retardation" and other categories for mental disabilities declined. The researcher, Dr. Paul Shattuck, notes that this pattern has been observed in the past and that it was not the case that there was epidemic. In short, the problem is that there was a better diagnosis/classification scheme put into place and the primary data source that is often used to justify the "autism epidemic" claim is tainted and cannot be used to determine if there really is an epidemic.
Now, that last part does leave open the possibility that there was a surge in the rate of autism in the population precisely at the time that there were new diagnosis/classification categories. However, this research still implies that even if there is an increase that such an increase is not as was initially suspected. Here is what Dr. Shattuck has to say on this,
Steep growth in administrative prevalence after introducing a new category is a common pattern that was also seen in the other 2 reporting categories newly introduced in the 1990s (TBI and DD). As with autism, in the first few years these categories were used it was not uncommon for states to report very few children with a primary diagnosis of TBI or DD. The prevalence for these categories also had nowhere to go but up. Suggestions that special education trends substantiate the existence of an autism epidemic would logically also have to either claim an epidemic of brain injury and DD or explain why the same pattern of growth in these 2 categories does not represent an epidemic as it does for the autism category.--emphasis added(Orac also provides the helpful hint that TBI= traumatic brain injury and DD=developmental delay.)
In short, if we are going to use the trends in a dataset to justify the existence of an epidemic for a given ailments, then logic requires that we do the same whenever we see other similar trends. Why would we expect an epidemic of TBI? Seems kind of weird to me, and given that in three cases where a new category is added we see similar trends in the data it seems that the most reasonable explanation for the observed trends was the introduction of the categories. The other option is to attribute epidemics to completely unknown causes.
This was one of the my first guesses as to the "rise in the number of cases with autism" back when I first heard about it. I was assured by people who were in favor of the epidemic hypothesis (e.g. Dwight Meredith) that this was not the case. However, autism was not even a required category in virtually all special education programs across the nation until after 1993.
Here is how Dr. Shattuck puts it with an example in an interview he gave,
For instance, consider data collected in Wisconsin: In 1992, 18 children were counted in special education programs as being autistic. By 2002, that number had jumped to 2,739."The conclusion is that the prevalence of autism has grown by 15,117 percent. This is ridiculous," Shattuck said. "No credible clinician or scientist in the field would ever suggest there were actually only 18 children with autism in all of Wisconsin in 1992."
The idea that rate of growth of autism is over 15,000 percent isn't ridiculous, it is blindingly ludicrous.
A quick Google search turned up this kind of nonsense that careless research can lead too. Add in opportunistic politicians and you have a recipe for junk science. Lets look at the dates here. Hmmm...the article is published in 1999, so just in time for that new category to pop into the "developmental services system" under the new category. Now how many of those children were already in the system, but under a different category? Whoops, didn't think to stop and ask that question.
The Department of Developmental Services report, "Changes in the Population of Persons with Autism and Pervasive Developmental Disorders in California’s Developmental Services System: 1987--1998" was released to the legislature this morning. While it confirms the increased incidence, the report does not examine factors leading to the increase. The report was required as a result of legislation developed after parents, human services professionals and educators expressed concern that they were seeing a dramatic increase in children with autism.
Gee, could it be because in 1993 there was no autism catagory for special education, and then in 1994 there was and all those kids who had autism were now migrating to the right catagory? Also, I strongly suspect that the report looks at the number of people entering the "system" and not things like, the number of people per 1,000 people in the state, or looking at population growth in California as well. That is, if the population of California grows by say 3% from 1987 to 1998 then there are 38% more people in California which, holding all other things constant there would be 38% more people with Autism entering the Department of Developmental Services simply due to population growth.
The bottom line is that large changes like a 273% increase could have very reasonable explanations and shouldn't necessarily lead to hysteria. And as Orac has already discussed, it makes the chemical castration of children with autism even more dubious.
Update: I found the report (pdf), it is on the Department of Developmental Services website. Here is a nifty tidbit,
In this report the incidence of autism is not measured. The unit of measure reported here is the rate of occurrence of persons with autism or other PDDs in the regional center and developmental center system during a specified period of time.
In other words, the 273% number isn't the same thing as there has been a 273% increase the incidence of autism--i.e. it isn't necessarily becoming more frequent.
Also of interest is this picture from the report. Note that the population the picture refers to is the State population. If Autism were increasing in the population that time series for Autism would show an upwards trend. If there is one there it is very, very slight and possibly statistically meaningless.
Posted by Steve at April 11, 2006 08:32 PM | TrackBack
