Tuesday, February 28, 2006

Omega 3 and autism: how to account for dietary aversion?

Recently The Economist, a news magazine, highlighted Omega 3 fatty acids in its science section. This got some attention, including this BBC article which confuses The Economist with an academic publication. The thesis is of great interest to parents of children with autism; although the article speaks of "lower IQs" the traits allegedly related to low fish intake sound primarily autistic:
BBC NEWS | Health | Oily fish makes 'babies brainier'

... Looking at the effects of Omega-3 intake on 9,000 mothers and their children, the team found mothers with the lowest intake of the essential fatty acid had children with a verbal IQ six points lower than the average.

While those with the highest consumption of mackerel and sardines and other sources of Omega-3 had children, at age three-and-a-half, with the best measures of fine-motor performance, researchers said.

Low intake of the crucial fatty acid also appeared to lead to more problems of social interactions - such as an inability to make friends.

Research leader Dr Joseph Hibbeln said "frightening data" showed 14% of 17-year-olds whose mother had eaten small quantities of Omega -3 during pregnancy demonstrated this sort of behaviour.

This compared with 8% of those born to the group with the highest intake, he said.

... Professor Jean Golding of Bristol University set up the original research - the Avon Longitudinal Study of Parents and Children -15 years ago to look at the predisposition to disease.

She told the BBC: "The baby's brain needs Omega-3 fatty acids. It doesn't create its own fatty acids so it needs to be something that the mother will eat." ...

The richest sources of Omega-3 are larger fish which eat other fish, but research shows that the larger the fish the more pollutants, such as mercury, they contain.

For this reason Mr Holford recommends women consume two portions of wild or organic salmon, trout or sardines weekly.

Seeds such as flax, pumpkin and hemp are good sources of Omega-3 for vegetarians, but large quantities need to be consumed to gain the same effect.

This might translate to two tablespoons of seeds daily, Mr Holford said, but women can also use a high quality Omega-3 supplements.
Hmm. This may be one of the most flagrant examples of confusing correlation with causation I've seen in the past few years.

Let's consider autism. Autism is a genetic disorder with complex inheritance and variable penetrance. The parents and siblings of autistic children show "autistic traits". Most autistic people have strong food aversions. It would not be surprising if persons with autism, or their relatives, avoided oily fish. If so, could that explain the entire relationship between Omega-3 intake and lower IQ and social isolation?

Even if Omega-3 intake was helpful to infants who didn't have autism, what's to say it would help those with autism? It might even hurt them.

Bottom line, this is an interesting study. It would be very good to know what the animal models tell us. I would not be surprised, however, if 10 years from now Omega-3 was not considered terribly important.

PS. I've noticed the Google Ads appearing on this page tend to favor 'alternative' therapies and remedies. That's not my gig, and if they keep doing that I'll take them off the page. The only reason they're there is to provide a service to readers -- I don't make anything from them.


Joseph said...

I agree that correlation does not imply causation, but this is often assumed. The only way to tell in these cases is to do a double-blind placebo-controlled trial. I read somewhere that Omega-3 does have some benefit for autistic children, but that's probably just because Omega-3 is good for you. There's no evidence of brain damage or malfunction across the board in autism. In fact, more recent evidence suggests it's simply low socio-linguistic intelligence coupled with high visio-spatial intelligence.

John said...

Joseph, sorry for the delay in publishing your post. I thought I'd authorized it but I found it still in the queue today.

Todd said...

With the hippocratic oath, there is the statement "at first, do no harm."

There is no harm in pregnant women ingesting sources of DHA-rich Omega 3 fatty acids and there is a lot of evidence that it does tremendous good. A huge longitudinal study in England resulted in the government REVERSING entirely what the US FDA said -- i.e. that women should avoid large fish due to risk of mercury poisoning. Apparently the benefits were so great, they outweighed the potential harm.

There is a study showing that autistic children digest fatty acids more quickly than other children. And there are studies that show a significant improvement in the condition of autistic children owing to their intake of Omega 3s.

While there is no particular evidence of causation, perhaps, the same can be said for the presence of mercury in vaccinations. Causation of autism has never been proved; it's superstition to suggest otherwise, and people who don't have their children vaccinated risk the health of society as a whole.

If I have to choose, my bet is that Omega 3s are just what the doctor ordered. And I staked my child's health on that bet -- and she is acknowledged by strangers to be precociously aware and engaged and stupendously happy. Causation or just good genes? I guess I'll never know..

John Gordon said...

Cumulative response to comments:

Joseph - I don't interpret the autism literature the same way you do. I think there's lots of evidence of malfunction but that's against a very diverse set of minds.

Todd - I have seen only case control studies on Omega 3. Those are worth very little. If you know of any experimental studies with a reasonable animal model I'd love to read them.