Diagnosis of autism traditionally requires extensive interviewing and evaluation by behavioral experts, which means that autism spectrum disorders — which can range in severity over a stunning spectrum — can easily be missed.
Today the mainstream news is aflutter with reports that researchers at King’s College London claim they can detect subtle but distinctive differences in brains affected by autism spectrum disorders when comparing them to normal brains. Reporting in the Journal of Neuroscience, the researchers say they detected autism with 90% accuracy. The problem? The study was only of 40 patients — 20 autistic, 20 non-autistic — who were diagnosed by traditional methods and then given an MRI. That means with an error rate of 10%, 4 out of 40 would have been misdiagnosed — whether as false-positives or false-negatives, the BBC article doesn’t say but the Journal of Neuroscience article might (trying to find the actual citation in J. Neurosci‘s tables-of-contents is like looking for a needle in a haystack).
The hope is that the procedure may lead to a clinically available test for autism that will be cheaper and use fewer resources than behavioral diagnostics — but as with most medical studies, the big caveat is “this is a really small sample.” It’s a long way from forty patients to a clinically useful device.
Autism is usually said to affect about 1% of adults in the U.S., but like everything about autism, that’s controversial. The BBC article reports that most of those affected by autism are men. This gender disparity, reported often in the media, is actually grossly overstated; the ratio is something like four or five to one male-to-female, but few sources seem to agree. I see the idea that autism is a “male disease” reported requently in the media; I suspect it may come from the idea, soundbite-heavy but controversial, that autism is a case of an “extreme male mind,” as claimed by experimental psychopathologist Simon Baron-Cohen — yes, Borat‘s cousin.
Other researchers, including a team at Albert Einstein College of Medicine and elsewhere, suggest that girls may carry the gene for autism much more than detected. Dr. Rachel Mills of the UK charity Research Autism suggests that girls may be underdiagnosed and may typically be diagnosed later because they’re less likely to show delayed language development, and girls with autism may seem less unusual than boys with autism, because of different expectations of female vs. male behavior.
Back in January, CNet reported on similar procedures to the Kings College study used at the Children’s Hospital of Philadelphia, not for diagnosis but to research how autistic children process speech and other sounds.
Oh, while we’re talking autism, did you know a Danish corporation, Specialisterne (“The Specialists”), founded by a man whose son is autistic, seeks to use the characteristics of employees with autism spectrum disorder as an advantage in the business world? Best idea they’ve had since getting Beowulf to whack Grendel.
Obsessively stacked Campbell’s Soup Cans courtesy of Andy Warhol and Wikipedia.