Autism linked to digestive problems?

[Lauree]

I found this while I was ramdomly searching for a cure for my mouth ulcers.

I'm not recommending it or otherwise - I just thought some of you might be interested...

gutdoctor.com

SaintlyDame - I haven't really any links - am at the research literature end of this argument but it's not my area so might have to wait until I'm less busy before i can produce a précis that i'm happy with! I do have a vested interest - mild ASD runs in the family and there is strong evidence of gut / food effects in some ASD family members but not others.

What the effect actually is, I have absolutely no idea. It may be something as indirect as Amber's suggestion above. Or may be something with a more direct immuno-neuro-modulation as I suggested. Or some combination...

While agreeing with the point above re sensitivity on this particular board, I totally agree with DBennett's approach (I am just a lot less informed than DBennett). Indeed I am actually a member of/ advisor to the bad science forum, aka Sense About Science.

DBennett, I wonder if I know you...?

Lol yes it does help to know you're not a member if bad science - not because they don't have anything of interest to say (sometimes they do of course) but because they often post to prove a point and that's not so helpful.

I do take your point about cost and the need for evidence. However, again that leads me back to the need to identify subgroups. In ds1's case he'd had so many antibiotics that anything that helped restore gut bacteria was likely to be useful - and it did seem to be. Incidentally Reading university were meant to be carrying out a research project on prebiotics and autism - I don't think it ever happened (funding issues I think).

Wrt interventions in general I think many parents feel unable to wait for concrete evidence. We've always taken the approach we'll try something if we can afford it, it's unlikely to do harm and it's likely to be applicable to ds1 and his type of autism.

The good news is that things have changed. In the last year I have had 2 paediatricians and 1 neurologist suggest interventions that only a few years ago would have been viewed as fringe. The more people are able to access help from their nhs professionals the less likely they are to stray into the world of the dangerous or into the paths of people making a quick buck. It's been really pleasing to see the NHS catch up with current research - I've had some very helpful discussions with the neuro and paeds this year.

byt the way SaintlyDame - you haven't declared your interest in/ background in science. It's clear you are interested in the science here, but what you've said on this particular set of posts makes me think you're operating outside the scientific research community.

This doesn't mean you are ill-qualified to make judgements, or to be appropriately cynical. Your personal experience means you are sensitive to nuances that others may miss.

But the selectivity that comes with a strongly evidence-based approach, as (for example) seen in DBennett's posts about a lot of things, is useful, and makes the plethora of scientific literature less confusing. Statistics and epidemiology are used for a reason..... they work the way they say they will. Whether you then think "well this epidemiological study doesn't apply to my particular case for XYZ reasons" is your call. But that's a different point from whether or not a particular epidemiological study is properly designed, executed and interpreted.

(end of ramble )

Reading are still doing the work. I can't link as am on iPod but google autism Reading prebiotics and you'll get it.

thanks for that :)

will look into the possibility of running a Sense About Science precis on this stuff. Like a cochrane review but intended for non scientists. The advantage of SAS involvement is as well as having the overriding objective of evidence-based reporting, they have the pulling power to get SBC and others involved, whereas (as a scientist in a mostly unrelated field) i don't.

I realise that massive but I happen to think that the lack of identification of subgroups is a massive problem in autism research. Without identifying subgroups you just cannot make sense of interventions. If a study is treating autism as one thing then I think it is heavily compromised. The MIND study I linked to makes that point. I personally believe that identifying subgroups is the most important step that is needed.

I work on severe autism - a very under represented group in the literature presumably because it's a group that is very hard to work with. Part of the reason I bang on about subgroups is because if you begin to look in detail at paper you find this is a group that is more or less absent from the research literature. This has been recognised fairly frequently - it just hasn't been addressed - and it is difficult to address.

Sbc though really only works with AS/HFA.

Cambridge actively recruit participants to take part in studies. This is commendable. Before you take part you have to fill in questionnaires about your family. These are basically sbc's systemizing/empathising scales. You cannot take part in any studies until you have filled in the questionnaires. But - big but- the questionnaires cannot be answered for someone with severe autism. The questions do not make sense. About 50% of the questions cannot be answered at all. So at that stage there is selection going on. This is fine - providing when published it is noted that the research relates to HFA/AS not 'autism'. That distinction generally isn't made.

I did contact Cambridge and they did agree that they needed a new questionnaire. They removed the need for ds1's to be completed before we could take part in research - but how many people with bother doing that.

very good points there. I would specifically include that kind of thing (recruitment bias in literature, applicability of particular results) in an SAS precis.

Just give me 6 months and I'll be able to think stright enough to get it started!

I should clarify - for studies involving cognitive testing etc the papers are very clear about the group that has been studied. If the work is experimental it will beon HFA/AS and this is stated clearly. I don't have a problem with that - although I'm always pleased to see a paper noting the absence of literature on severe autism.

I have more problem when the research is on for example causes of autism and recruitment bias isn't noted. we took part in a Cambridge study on sibling finger length (related to testosterone in utero) and the broader autism phenotype. If questionnaires such as the one above were being employed then it would be very important to recognise and note the sample population. In those sorts of studies this often doesn't happen.

One simple way of dividing autism is simplex versus multiplex families. If you do that you are broadly dividing along genetic lines as well. It's crude but cheap and would make studies more relevant.