Food and ADHD: lessons for autism?

I admit that I am a little late in getting to this topic. Late because for quite a while now (years in fact) questions have been raised about a possible link between food, diet and some cases of ADHD (Attention Deficit Hyperactivity Disorder), late because the study published by Pelsser and colleagues which forms the bulk of this post has already been through the blog mill, and late because I had to really think about how to write this post - unlike my others!

First things first. For many years some parents of children diagnosed with ADHD (or ADHD-type symptoms also co-morbid to other things on some occasions) have reported a connection between what their child eats and their child's subsequent behaviour. It was the work of Dr Feingold and a few others who set out the stall that certain types of food preservatives, colours and flavours might be implicated in behaviours associated with ADHD. There has, since the early years of this work, been some significant to-ing and fro-ing about whether Feingold's assertions were correct.

In 2007-2008 two things happened: McCann and colleagues published the results of a tightly controlled study on the effects of food additives on hyperactivity, concluding that certain preservatives and artificial colourings did impact on behaviour. A review by Schonwald in the American Academy of Pediatrics Grand Rounds journal was accompanied by an editorial which all but admitted that following the McCann paper there may be a connection; to quote:  ".. that even we skeptics, who have long doubted parental claims of the effects of various foods on the behavior of their children, admit we might have been wrong".

OK so far? The next "stage" in this saga was the publication a few days back now, of a study from Lidy Pelsser and colleagues (like McCann et al, also published in the Lancet). Pelsser and the group based at the ADHD Research Centre in the Netherlands which includes Jan Buitelaar, have been very, very busy in the area of food and diet in connection to ADHD, publishing results from quite a few trials (see here and here). The result of their most recent study concluded that "a strictly supervised restricted elimination diet is a valuable instrument to assess whether ADHD is induced by food". I am not going to dissect the study here; feel free if you wish to.

So what we have is some good evidence that diet and food could affect some children who present with ADHD or ADHD-type behaviours - emphasis on the "some" children because as we know science is about probability not certainty and we would be hard pressed to say all children show a connection. You may ask then where autism comes into this?

Well, first off, we know that autism and ADHD can occur alongside each other and the association seems quite strong. We also know that some people with autism seem to be affected by diet; whether as a consequence of co-morbid conditions such as coeliac disease or through less well-defined connections. The types of diet most commonly linked with autism are those excluding gluten- and casein, but there are several more being quite widely discussed. I will stay with the GFCF diet because that's what I know ('a cobbler should stick to his last' and all that).

The most recent Pelsser study described their diet (at least in the first phase of their study) as consisting of a few foods; just rice, meat, vegetables, pears and water. Notice that a few things are missing from this diet including our old friends gluten and casein (although I believe small amounts of wheat were included on specific days for some participants for RDA reasons). When however you compare the Pelsser results on ADHD with some of the data from autism, a few "similarities" start to occur. For example, during our most recent 2-year ScanBrit trial of the GFCF diet for autism we included a few measures outside of autism aimed at examining ADHD-type behaviours (for those of you interested it was the ADHD-IV scale by DuPaul and Power) given the previous suggestions that such areas of functioning may be affected by diet. What we found and reported on was that hyperactivity and inattention sub-domains seemed to indicate some "group" reduction following intervention.

Fair enough I hear you say, a non-blinded study showing that ADHD-type behaviours presented by children with autism "seemed" to show a significant group reduction effect after diet? One study: it shows nothing. Well not quite. Another recent trial by Cynthia Johnson and colleagues also looked at the effects of a GFCF diet on children with autism. Similar to ScanBrit, the trial was randomised but not blinded and compared a GFCF diet to a healthy low sugar diet for autism over the course of 3 months. Their results: overall no significant gains on the GFCF diet vs. the healthy low sugar diet but reading the full-text to their paper, an interesting finding: a significant reduction in ADHD behaviours for the GFCF group (p = 0.043) (Table 5) which were also examined during their study.

The skeptic in me says OK possibly a fluke event in both cases; the reduction in ADHD-type behaviours could be down to the study design (non-blinded), other things the participants were doing at the time of study or merely because more attention was given to them as a consequence of being "studied". A small, little, tiny voice is however telling me: "mmm, this is interesting, does this need a little more study?" I will let you answer that question yourselves.