Reading the Research: Unreliably Tested

Welcome back to Reading the Research, where I trawl the Internet to find noteworthy research on autism and related subjects, then discuss it in brief with bits from my own life, research, and observations.

Today's article underscores how far we have yet to go in terms of early diagnosis for autism and other differences.  Psychology is a very young science, particularly the study of the brain and how people work, so it's understandable that sometimes things simply don't work the way you expect them to.  

This test, the M-CHAT, is a widely-used screening test given to toddlers when there's a suspicion.  It's usually administered by parents to their kids at home, and used to rule out autism.  A positive result is a good sign that you should take your kid to a specialist to get a proper screening with a different test.  A negative result was supposed to let you and your kid get on with your life, unworried about that particular developmental complication... 

As it happens, that negative result apparently isn't to be well-trusted.  The test was supposed to detect 91% of children with autism, but in reality appears to only detect about 40%, which you'll notice is quite a drastic difference.  This is a reoccurring theme in "my family's life with an autistic ___" books, but if your parental gut continually tells you something is unusual, it probably is.  

Now, please understand that while it can be very helpful to catch autistic differences early, which lets you start therapy and support services early, it is not a death sentence or the end of the world if it's missed.  Obviously, I survived nearly to my 21st birthday without an ounce of extra supports, and there are many adults like me.  

These days autism is much more well known than it was when I was growing up, and it's much more likely that someone will say, "hey, so you/your kid, are you/they on the autism spectrum?"  Or some concerned teacher will notice things, or their school performance will be spiky and merit extra supports, which raises the question of what's causing these higher needs.   So there's a lot of chances for an autistic kid to be noticed and get the support they need to live their best life.  

I would guess, if pressed, that the makers of the M-CHAT likely didn't use a proper cross-section of the US population when testing their creation.  Often your test sample is white, middle class families who have time to do this sort of testing, rather than a broader, racially mixed and diverse in household income sample.  This is actually a really major problem in psychology as a whole.  Research is often conducted in colleges, and who mainly has the money to send their kids to college?  Middle and upper class white people.  So those college kids tend to be the test subjects for psychology research.  This is sometimes called a "sample of convenience."

It leaves a lot of people in the lurch, and makes for a poorer, less scientifically rigorous result... and this disappointingly unreliable test is one of many results affected.  

(Pst! If you like seeing the latest autism-relevant research, visit my Twitter, which has links and brief comments on studies that were interesting, but didn't get a whole Reading the Research article about them.)