I just posted about a forthcoming study on pre-diagnosis autism interventions. Several of the comments were interesting.

Of course there was the standard loony theory:

This study is utter nonsense and a big waste of money. If an older sibling has autism-the only way to prevent another child from becoming autistic is to stop vaccinating the younger child.
Another ridiculous study to try and blame it 100 per cent on genetics or on the parents.

This is a meme with some serious staying power, but it's false. Another comment had a great theory I hadn't heard before:

Personally, I think the SIDS Prevention “Back to Sleep” Program caused the autism epidemic. Having all these kids on their backs all day is definitely not healthy. Supine (back) sleep makes sleep apnea worse (by a 4:1 male to female ratio), causes social skills delays at 6 months, causes motor skills delays at 6 months. Also, between 1987 and 1993 over 30 industrialized countries began “Back to Sleep” campaigns and that coincides with the onset of the autism epidemic. That’s just my 2 cents…..Tom

I couldn't find any references about it except a Google Answers question whose answer said no research had been done. No evidence except a vaguely similar onset equals no reason to believe it.

But one comment pointed to a doctor working on autism who advocates very early detection. He has something called the tilting test, which works as follows:

The Teitelbaums asked parents of Asperger's children for videos of their babies at less than one year old. They then examined the videos of 16 babies frame by frame. The researchers said they found that certain physical characteristics stood out when it came to the babies' movements.

The infants apparently had difficulty sitting and standing without hunching over. They also fell without putting their arms out to catch their fall. Osnat Teitelbaum says the babies did not display the protective reflexes that most infants develop.

She suggests parents try what she calls the tilting test. When carrying out the test, parents hold the baby away from their own bodies, then slowly tilt the baby from an upright position by 45 degrees to the right and left. By eight months of age, most babies will try to keep his or her head directly upright. Meanwhile, a child with Asperger's syndrome tends to keep his or her head aligned with the body. That means his or her head would completely tilt from side to side during a tilt test.

I have no idea how to evaluate this claim, but it's a fascinating one. If you know at 6 months that you have a child with early indications of autism, you will do things very differently than otherwise.

The NYTimes has an article about a controlled study of how parenting might impact autism rates in high-risk kids:

All of the babies recruited for the study will be monitored by specialists and evaluated at the ages of 6, 12 and 24 months. But half the mothers in the study will be coached in a unique intervention that trains them to detect subtle communication cues from their babies. The mothers will be taught how to take advantage of periods when their children are reaching out, engaging infants in eye contact and communicating with them in a lilting voice that captures attention and may make it easier for kids to learn language.

What is interesting to me is the reaction in the comments to the story. Along with the typical negative reactions people have to medical studies (you're denying half the group the treatment!), many people were furious that a scientist might suggest that parenting impacts autism -- since if that's true, then in some sense you might be to blame for your kid's autism.

Here's a sample:

as a mom to a 16 year old son with high functioning ASD, i was saddened to read the article. my heart shattered 14 years ago when I first received the diagnosis and it breaks into even smaller pieces when i read articles that suggest that “socially connecting” to their child in ways such as “cooing” or “gazing into their eyes when they look at you” will “cure” or “improve” the symptoms of autism. no one gave more love, attention, and just plain everything to their child than me and my son is still on the spectrum. he is engaging but in his OWN way and will ALWAYS be that way.

Actually, this isn't about you and your son -- it's an experiment, and even if it works, it will be a small marginal effect rather than a cure or a prevention. Certainly the fact that they are researching this is good, and not some kind of attack on your parenting skills.

Mostly, I'm interested to see what the results of this are. Autism has a very strong genetic component -- but can borderline kids be saved by this kind of intervention, even before they are symptomatic? If so, combined with genetic screening (which doesn't yet exist but will pretty soon), we might actually be able to prevent autism in some cases.

Now that I have a little one on the way, I'm paying more attention to articles like this: watching baby Einstein delays language acquisition:

Led by Frederick Zimmerman and Dr. Dimitri Christakis, both at the University of Washington, the research team found that with every hour per day spent watching baby DVDs and videos, infants learned six to eight fewer new vocabulary words than babies who never watched the videos. These products had the strongest detrimental effect on babies 8 to 16 months old, the age at which language skills are starting to form. "The more videos they watched, the fewer words they knew," says Christakis. "These babies scored about 10% lower on language skills than infants who had not watched these videos."

It's not the first blow to baby videos, and likely won't be the last. Mounting evidence suggests that passive screen sucking not only doesn't help children learn, but could also set back their development. Last spring, Christakis and his colleagues found that by three months, 40% of babies are regular viewers of DVDs, videos or television; by the time they are two years old, almost 90% are spending two to three hours each day in front of a screen. Three studies have shown that watching television, even if it includes educational programming such as Sesame Street, delays language development. "Babies require face-to-face interaction to learn," says Dr. Vic Strasburger, professor of pediatrics at the University of New Mexico School of Medicine and a spokesperson for the American Academy of Pediatrics. "They don't get that interaction from watching TV or videos. In fact, the watching probably interferes with the crucial wiring being laid down in their brains during early development." Previous studies have shown, for example, that babies learn faster and better from a native speaker of a language when they are interacting with that speaker instead of watching the same speaker talk on a video screen. "Even watching a live person speak to you via television is not the same thing as having that person in front of you," says Christakis.

There are also studies that suggest that autism and television viewing among very young children are correlated (though there are reasons for skepticism). The causal link is much less certain, but there's some reason to think that we (both silicon valley types) might have higher risk for an autistic child, so it seems like it's worth being careful.

Plus there's a much more reputable study that found a link between early television watching was associated with attentional problems at age 7. Again, not necessarily causal, since it could be that kids with attention problems are more likely to get put in front of the tv to give parents a break. But all of these different studies add up to my wanting to err on the side of prudence.

And of course the American Academy of Pediatrics recommends limiting TV viewing. Their list of effects contains ones that sound plausible to me (kids parked in front of the TV are more likely to be overweight) and ones that don't (I'm suspicious of claims that the number of murders you see on TV has much effect on anything). What does seem pretty clear to me is that the costs pretty clearly outweigh the benefits, at least for very young (<2 years old) children. Perhaps we'll revisit the issue in a couple of years.