Causation & correlation: What declining special ed rates don’t (necessarily) say about autism diagnoses

February 4th, 2011 → 6:05 pm @ // 5 Comments

Earlier today, California Watch, a project of the Center for Investigative Reporting, posted a short piece on rising rates of autism and declining rates of students with “learning disabilities”:

More than 680,000 students – 11 percent of all California public school students – are enrolled in special education. The number of students diagnosed with autism climbed from 17,508 in 2002 to 59,690 in 2010, the Lucile Packard Foundation for Children’s Health found.

Students with autism represented 8.8 percent of all special education enrollment last year, up from 2.6 percent in 2002. Other health impairments – defined by the state as “limited strength, vitality or alertness, due to chronic or acute health problems,” such as a heart condition, asthma, epilepsy or leukemia – are also on the rise, comprising 7.9 percent of disabilities among special education students.

At the same time, the number of special education students with a learning disability – the most common diagnosis – is falling. In 2002, 52.4 percent of students had a learning disability, compared to 42.3 percent in 2010. Speech or language impairment affects about one-quarter of special education students.

At first blush, this data would seem to lend credence the idea that the well-documented (and much discussed) rise in autism diagnoses does not actually correlate to an actual increase in autism in the population.* Except…well, except for the fact that correlation does not equal causation. Maybe budget cuts in special ed programs are squeezing out students with other diagnoses — while autism, which has received an enormous amount of attention in the media over the past decade, has received a larger share of educators’ attention. Maybe more students who are diagnosed with other learning disabilities are going to private schools. Maybe this is attributable to something no one has even thought of yet.

The point is, we just don’t know. As the California Watch piece itself notes, “The data do not explain these shifts in disability diagnoses.”

* Roy Richard Grinker, the author of Unstrange Minds, has written about the effects of changing diagnostic criteria on autism rates extensively; he documents the way the DSM has evolved in this regard here.


Post Categories: Blog post
Tags: , , ,

5 Comments → “Causation & correlation: What declining special ed rates don’t (necessarily) say about autism diagnoses”


  1. Emily Willingham

    3 years ago

    When I read that, the phrase that caught my attention was this one: “Other health impairments – defined by the state as ‘limited strength, vitality or alertness, due to chronic or acute health problems,’ such as a heart condition, asthma, epilepsy or leukemia – are also on the rise, comprising 7.9 percent of disabilities among special education students.” I’m wondering about that one far more than the other.

    Reply
  2. [...] This post was mentioned on Twitter by Brian Mossop, Seth Mnookin. Seth Mnookin said: Causation & correlation: What declining special ed rates don’t (necessarily) say about autism diagnoses http://bit.ly/gEosDQ [...]

    Reply

  3. Liz Ditz

    3 years ago

    Also see fine reporting from Autism News Beat

    http://autism-news-beat.com/archives/1443

    A basic tenet of America’s ill-conceived and foundering anti-vaccine movement holds that autism diagnoses have risen to epidemic proportions. Among the faithful, this imagined increase is “the initial fork in the road in the autism debate,” one that colors their entire understanding of autism. But is the increase real, or an artifact of how autism spectrum disorders are diagnosed and counted?

    New data from California indicate the latter. The number of special education students with autism in the Golden State has more than tripled since 2002 , as the rate rose form 2.6 to 8.8 percent, but the overall special education enrollment has remained nearly unchanged, according to an analysis of state education data.

    Reply

  4. Doctor Karen

    3 years ago

    I think we’ll be arguing this question about rising autism numbers until some years have passed and we see whether incidence levels off, declines, or continues to rise. My gut feeling – there really is an increase, perhaps not as dramatic as the numbers suggest. I have an autistic child. I know a fair number of autistic children consequently. These kids are not just “a little quirky.” They have serious behavioral issues and communication deficits but can still have normal or even high IQ. I’m also a physician who went all the way through medical school without seeing a child with autism. That was 10-12 years ago, back when autism was much less common (by the numbers). Regardless, it will be interesting to see how things play out in the coming years.

    Reply

  5. jre

    3 years ago

    The single most interesting examination of autism statistics I have read was this post comparing the (rising) rate of autism diagnoses with the (falling) rate of mental retardation diagnoses over the period 1993-2007. The two are almost exactly complementary, so much so that their combined rate has not changed significantly. This raises the possibility that diagnostic substitution may account for most or all of the “autism epidemic.” I have not run across any similar analyses — but I haven’t looked, either. Is anyone aware of any?

    Reply

Leave a Reply