Thursday, September 11, 2014

Early Intervention for Austim: A miracle cure the media cannot resist


A new pilot study suggests intervention as early as 6 months could reduce or eliminate autism symptoms in children at high risk. Published on Tuesday in the Journal of Autism and Developmental Disorders, media quickly jumped all over the very small study despite its highly speculative findings:

 Pilot Intervention Eliminates Autism Symptoms In Babies Huffington Post

Autism Therapy in 6-Month-Old Babies Eliminates Symptoms in Limited Study Newsweek

Treating Infants for Autism May Eliminate Symptoms NBC News

Study finds early treatment for infants may remove signs of autism Fox News

Taking Action Early May Protect Against Autism WebMD

Could early intervention erase signs of autism? CNN

Could early intervention reverse autism? CBS News

Earlier Help for Children at Risk for Autism Wall Street Journal

The Wall Street Journal hedged its bets and didn’t say much of anything. Not so for Huff Po and Newsweek who declared this latest children’s health discovery a sure thing. The rest of the lot had the good sense to throw in a question mark or otherwise acknowledge that the early intervention, however successful it might prove in the future, is not yet the second coming. 

Did I mention this study intervention involved a mere 7 children 6-15 months old? That's how many kids I could sit in my dearly departed minivan, I couldn't possibly call it a sample.

Despite the tentative nature of the study, Huff Po ran with the results delivering no hint of caution in its opening lines:  

A small new pilot study has found that parents can help significantly reduce symptoms of autism in babies who haven't even reached their first birthdays simply by changing how they play and interact with them.

Oh my. Get on it parents, right now! Huff Po didn’t bother highlighting the fact this was indeed a tiny study. They didn’t stress the exceptionally few number of youngsters involved or the lack of randomized design and thus difficulty establishing cause and effect. It’s absurd, no impossible, to do significance tests on such a limited pile of data. 

The only nod to caution in the entire article gets quickly buried from an outside autism professional gushing about the results: 

While the new research is a pilot study and as such is highly preliminary, Gerard Costa, director of the Center for Autism and Early Childhood Mental Health at Montclair State University, called the outcomes for this small group of children and their parents "wonderful" and "very encouraging."…"The findings are clearly saying what we've always felt to be the case, which is that early intervention can make an enormous difference," he said.

Enormous difference. We got the cure for autism right here folks.

Even news sources that did highlight the small number of participants and speculative nature of the results went to lengths to tell why we should be impressed by the results.

NBC News did the classic bait and switch in a mere 2-3 sentences:

“With only seven infants in the treatment group, no conclusions can be drawn,” the [study co-authors] wrote.
However, the effects were striking. Six out of the seven children in the study had normal learning and language skills by the time they were 2 to 3.

(To their credit, the authors did remind everybody that the study is a tiny first and should not be taken as definitive evidence.) 

Finally, CBS News found an expert who wasn’t at all impressed by the results:  

"This study is groundbreaking in certain regards," said [Dr. Lisa Shulman, director of infant and toddler services at the Children's Evaluation and Rehabilitation Center at the Albert Einstein College of Medicine Shulman]. "It pulls together various streams of current research in a meaningful way.

Groundbreaking? The CBS intern tasked with getting a good quote deserved the night off for this one.  

WSJ showed a bit more nuanced than most but just like its fellow media, reported on the great rise in autism cases over the past decade or so:

The number of children identified with autism has risen sharply since 2002, and the latest figures from a 2014 CDC report estimate 1 in 68 U.S. children are affected by autism or a related disorder. That climb could result from a combination of more children affected with the condition, plus greater awareness of it, experts say.

And like everyone else, WSJ failed to mention that other factor, the expanded DSM criteria back in the 1990s. Apparently the experts who attribute the rise of autism in part to the DSM changes didn’t “say” anything this week. 

Nor did the articles bother explaining the results could be due to factors other than the intervention itself. The families in the study were involved with the institute performing the intervention, namely because they had older children diagnosed with autism. Some families agreed to participate in the intervention (essentially  coaching parents how to better interact with their babies). Some families did not. Something different between these two groups could account for the seemingly miraculous results. It's just not clear at this point. I know the results seem encouraging but having done quite a few pilots I know sure results don't always pan out regardless of their potential value.

2 comments:

Andrea Riley said...

I recall seeing this on the news, someone looking into the camera with dramatically pathetic eyes, encouraging parents of 6 month olds to *really play* with them, meaningfully (yes they actually said that, so much for non - meaningful fun play). Apparently the more meaningful the less autism? Lemme add that to my list of things to do.

Polly Palumbo, Ph.D. said...

Andrea, this is the kind of great tv moment that even we could not make up! I am going to co-opt that term - "non-meaningful fun play."