Talk:Causes of autism

From Wikipedia, the free encyclopedia

This is a controversial topic that may be under dispute. Please read this page and discuss substantial changes here before making them.
Make sure to supply full citations when adding information and consider tagging or removing uncited/unciteable information.
Archive
Archives
  1. Archive 1 (June 15, 2007)

Contents

[edit] General (multicausal) theories

I thought this theory ought to have a mention as it appears not to be discredited in any way. I guess that some changes to surrounding context could be in order. I'm not a skilled Wiki editor so I'd guess best if I leave that to others. --Idealiot

As a genetics-based theory this belongs in Heritability of autism, not here, so I'm inclined to revert the change. If you want to put it in Heritability of autism I urge you to put it in its proper place, with context, and with a brief discussion as to why the theory is worth looking at. There are a lot of theories, many of which nobody ever refers to once published. Everybody thinks their theory is the one, but Wikipedia doesn't have space to discuss them all at length. Eubulides 00:47, 26 July 2007 (UTC)
Dear Eubulides, you make essentially two points, both wrong. Firstly RP Clarke's theory is not genetics-based any more than it is environment-based (rather it is gene-expression-based). It says absolutely nothing about the heritability of autism, which is not surprising since it is a theory about the causes of autism instead.(I continue in next para.)
There is a categorization issue here, as gene expression and heredity are not the same, but in an encyclopedia most would file gene expression under genetics. Certainly Wikipedia does: for example, Category: gene expression is a subcategory of Category: Molecular genetics. The Heritability of autism page, despite its name, is really about genetics. (Perhaps it should be renamed but that is a different subject.) That page already talks about gene expression and so is a better home for this sort of thing. Eubulides 16:15, 26 July 2007 (UTC)
Your reply only becomes relevant at its last sentence where it becomes wrong. Do I really need to tell you again that it is a theory of causes of autism and not of heredity or genetics - it is just as much about environment. As a genetics lecturer rightly said, there wasn't much genetics in it. Please take a rest from this wasting of other wiki-users time Eub, you do not own this page.--Idealiot 23:26, 26 July 2007 (UTC)
I don't see why the reply's last sentence is wrong. First, that page does already talk about gene expression; see its discussion of Samaco et al. 2005. Second, gene expression is the main focus of Clarke 2003; it argues that autism is caused by abnormalities in gene expression. Since Clarke's theory postulates a genetic cause for autism, any coverage of it belongs in the genetics-of-autism article. Eubulides 02:41, 27 July 2007 (UTC)
A friend once said that you cannot win an argument against a fool. Sadly, those wishing for competently-written autism pages on wikipedia are going to have to await the passing on of this person who is evidently devoid of ability to see the flaws in his own output.--Idealiot 07:42, 28 July 2007 (UTC)

Secondly, you put it in a category with a "lot of theories" which are ignored. But have you ever read any of them? (You evidently have not read this one.) The vast majority of such theories are with good reason ignored, because they have some fundamental flaws of reasoning or evidence, or are relatively flimsy speculations (e.g. the TV theory which you leave in). But RP Clarke's antiinnatia theory is very different from such-like. It is true that Wikipedia does not have space to discuss all the dodgy theories that exist, but that does not apply in respect of a substantial peer-reviewed work such as that of RP Clarke which has encountered no criticism. The Wiki article purports to be a proper uncensored review of autism theories. As such it should not exclude all mention of a substantial, unfaulted theory. Sure, it is not the function of Wikipedia to give a deviant presentation, but nor should it be participating in a mindless authoritarian mutual parrotting of other reviews presenting a false, unreasoned "consensus". It should strive to surpass mere Britannicas! (... else why bother?) For these reasons a reinstatement is in order, a bit shortened perhaps (though you will note that the original account of the antiinnatia theory did not indulge any space to mention any of its actual ideas). --Idealiot

Yes, I've read many theories of autism that are unreferenced in the peer-reviewed literature. Monotropism is one example (though it's recent enough that perhaps it's just taking time for followup studies to percolate through the system). And I did read the paper you referenced, which is why I suggested moving it to the other article for discussion there. Eubulides 16:15, 26 July 2007 (UTC)
It remains the case that you have not made any real reply to my previous points and so they remain right and you are simply wrong.--Idealiot 23:26, 26 July 2007 (UTC)
I have not replied to points about the theory's merits here, because I am restricting my discussion here to the issue of whether Clarke's theory is suitable for this page. The merits of the theory itself are better discussed in Talk:Heritability of autism. Eubulides 02:41, 27 July 2007 (UTC)
Please see my proposed abridged reinstatement at end of the talk page of Heritability of autism--Idealiot 14:09, 26 July 2007 (UTC)

[edit] Vaccine evidence

Eubulides put in the article that: "the vaccine hypotheses have no convincing scientific evidence". Sorry Eub but where have you been the last decade?! Sure, You, me and a zillion others are very unconvinced by the "evidence" re vaccines. But aren't you just slightly aware that a huge number of others are convinced?! So that claim is not only unreferenced, but it is counterreferenced. I suggest you spend less time on your worthy editing work (albeit much may be of value) and more time reading the things you cut out such as http://cogprints.org/5207 . --Idealiot

The claim was indeed referenced, by Rutter 2005. I am aware that not everyone is convinced. The article has many paragraphs promoting the vaccine theory, with what must be a dozen references on the subject, so it's not clear what the complaint is here. Eubulides 16:15, 26 July 2007 (UTC)

[edit] Organochloride pesticides

Re this change I made to Causes of autism:

  • None of the chemicals mentioned in Causes of autism#Teratogens are proven teratogens; they are all implicated by association only. Also, as far as is known none of them are always teratogens (ethanol, for example, is not). So the reasons given for having a separate subsection for dicofol and endosulfan were not sufficient. If there are to be two subsections I'd suggest titles like "Oral teratogens" and "Airborne teratogens" or something like that (is there common terminology in use here?). But giving organochlorine pesticides their own section is a bit strange and could easily confuse the reader.
  • As far as the citation goes, there shouldn't be two footnotes for just this one study, or even one multipart footnote; that's too much work for both us and the reader, with too little payoff. I don't know what the objection was to the "cite journal" template with the laysummary= argument; I thought it was fine. But if we don't use that, let's just leave out the lay summary entirely.

Eubulides 00:33, 31 July 2007 (UTC)

One argument for omitting the Cone article, by the way, is that in a matter of days it will be available to subscribers only, whereas the main citation will be available indefinitely. Eubulides 00:40, 31 July 2007 (UTC)
You guys too fast! All I get are edit conflicts everything I try to save! :)
Anyways, I'm not going to edit this anymore--if you want to revert I'll abide by that. I'm more intersted in pesticides than autism, and I built most of the endosulfan page. Anyway, neither endosulfan nor dicofol are established human teratogens like thalidomide, DES, etc, therefore I think it is incorrect and confusing to list endosulfan and dicofol in a subsection about teratogens. I don't see how breaking them out into their own section, as I had done, is "strange" or "could easily confuse" anyone. Perhaps the Prenatal enviroment section would be best without any subsections, or with Endogenous Toxicants and Exogenous Toxicants subsections or something. At any rate, listing endosulfan and dicofol as teratogens is incorrect, and i was just trying to fix that. —Preceding unsigned comment added by Yilloslime (talk • contribs) 01:02, July 31, 2007
If this result is confirmed, then endosulfan and dicofol will be known human teratogens, right? So is the problem that the text is lumping together chemicals like thalidomide (which are known human teratogens for reasons other than autism) with these two chemicals (which are not)? If that's the case, I sort of see the point, though I still would rather solve the problem in a different way. 131.179.64.200 05:25, 31 July 2007 (UTC)
'Teratogen' is understood to refer to agents that casuse physical deformities, and I think you'd really be stretching the meaning of the word to apply it to agents that cause autism in the absense of physical birth defects. And either way, discussing endosulfan and dicofol under the heading of 'teratogens' is akin to saying "endosulfan and dicofol are teratogens"—a statement which the scientific and medical communities have not made. Wikipedia is not the place to decide what chemicals are teratogens, we can only report what authoritative sources have to say on the issue. Furthermore, your reasoning should also apply to folic acid--why should folic acid have it's own section and not organochlorine pesticides?Yilloslime 17:01, 31 July 2007 (UTC)
Teratogen refers to any agent that causes a birth defect; the defects need not be physical abnormalities. But to help work around the problem I made this change. Endosulfan has significant teratogenic effects in laboratory rats, so I put in a reference to that effect. Perhaps Endosulfan should mention this too? Eubulides 19:53, 31 July 2007 (UTC)
I think your edit improves the section greatly! I don't have strong objections to adding material about teratogenic effects on rats to the endosulfan page. When I was working on that page I tried to focus mainly on data from humans since there actually are some pretty striking studies from humans, and I felt that going into all the animal data would dilute the human studies and make the page really really long. Yilloslime 20:48, 31 July 2007 (UTC)
On a related note, there is a growing body of evidence that pre-natal exposure to organophosphate pesticides, particularly chlorpyrifos, is a risk factor for pervasive developmental disorders--see the Wyatt paper referenced in the Chlorpyrifos page, and also this recent paper by Eskanazi: Environ Health Perspect 115:792–798 (2007). I'm not totally clear on the relationship between PDD and ASD, which is why I've hesitated and not added that these studies to this page. What do you all think? Yilloslime 23:18, 31 July 2007 (UTC)
The PDD diagnosis is equivalent to ASD (autism spectrum disorder) which includes autism, Asperger syndrome, and some other stuff. The Whyatt paper (actually, Rauh et al. PMID 17116700) doesn't talk about children diagnosed with PDD; it talks about children who have some PDD symptoms (the abstract calls this "pervasive developmental disorder problems" to highlight the difference). Its Table 8 shows an odds ratio of 5.39 (95% CI 1.21–24.11) for presence of PDD problems in children exposed to more than 6.17 pg/g of Chlorpyrifos. This is suggestive, but I dunno, it's a bit of a stretch to mention this here. Eskenazi et al. 2007 (PMID 17520070) has a similar problem: they measure "pervasive developmental problems" but did not measure whether the children had PDD. I think it's better to cite a review saying there are problems in this area and leave it at that. Wait, I did find one study showing indirect evidence of an association between chlorpyrifos and autism, namely D'Amelio et al. 2005 (PMID 16027737), so that might be worth a mention. I made this change. Thanks for the heads-up. Eubulides 07:59, 1 August 2007 (UTC)
As for the footnote issue, the reason I changed it is because I thought the way it orginally was it was not apparent that there were two different sources being linked--if you clicked on part of the link you got the highly technical EHP article, but if you clicked the other part you got the LA times. Also, I wanted to give the Cone article a proper cite: author name, article name, venue, date, etc. I don't agree that it's too much work for the reader--if anything this way is easier for the reader because it makes it clear what they are they are going to get before they click--a highly technical paper which most folks aren't going to understand, or a much more accessible albiet secondhand description of the research. One concern might be that having two citations makes it look like there are two studies when in fact there is only one, but the way it's phrased now it's clear that there is only one study. —Preceding unsigned comment added by Yilloslime (talk • contribs) 01:02, July 31, 2007
This is a style issue, and to some extent I'd rather defer style issues to the "cite journal" template. Maybe that template needs fixing. Still, I'd rather not have a huge amount of footnote text for what is essentially just one research result. Footnotes are big enough as it is. I like citing lay sources, but I don't like duplicating all the footnotes, and laysummary= is at least trying to address the issue using a middle ground. 131.179.64.200 05:25, 31 July 2007 (UTC)

I like the laysource parameter on cite journal; new trick for me, which I'll use in the future. But, the LA Times will go to subscription only, so a good reason not to include it. SandyGeorgia (Talk) 01:50, 31 July 2007 (UTC)

Perhaps by the time that happens, there will be some other lay source we can use that will be as good as the LA Times article. We can wait a few days and see what happens. 131.179.64.200 05:25, 31 July 2007 (UTC)
That the LA times article will soon become pay-per-view content is a good point, however, that article has been picked up by at least one other source, so it's possible that it'll be permanently available for free somewhere on the internet. Furthermore, there's no reason why we should only cite free articles, and many wikipedia entries do cite subscription-only or even print-only sources. By all means, for the ease of the reader we should try to cite free content whenever available, but if there isn't anything, then we shouldn't shy away from non-free sources. Anyways, my original objection still stands: cramming 2 different sources into the same citation is confusing, looks bad (in my opinion), gives short-shrift to the second source, and—existence of a wiki template aside—is not the conventional way to do things. Yilloslime 17:01, 31 July 2007 (UTC)

[edit] Merge proposal for environmental factors

Causes of autism and Heritability of autism #Proposed environmental triggers cover essentially the same material. Causes of autism treats Heritability of autism as a subpage for genetic factors, while covering environmental factors on its own. But Heritability of autism covers environmental factors too, in a major subsection. This subsection's coverage is shorter and less extensive than the coverage in Causes of autism, so I propose that it be merged into Causes of autism. Eubulides 07:23, 6 August 2007 (UTC)

Support, FWIW. cheers, Casliber (talk · contribs) 11:13, 1 September 2007 (UTC)
Good idea; anything that reduces redundancy in the autism-related articles will make them much easier to improve and maintain. MastCell Talk 18:04, 1 September 2007 (UTC)
Thanks. Sorry, I forgot to mention that I did the merge last month after nobody objected. Eubulides 20:15, 1 September 2007 (UTC)