Talk:Autism spectrum/Archive 10

Archive 5

←

Archive 8

Archive 9

Archive 10

Archive 11

Archive 12

→

Archive 15

New MMWR report

It looks like most of this is already in the article. Not sure if anyone would like to add anything from this Dec 18, 2009 MMWR report? [1] Doc James (talk · contribs · email) 19:30, 21 December 2009 (UTC)

Thanks for mentioning it. It's a primary study, so is a bit too much detail for this article. I added it to Epidemiology of autism #Frequency estimates. (No PMID yet! arggh. The PubMed guys must be on vacation.) Eubulides (talk) 19:48, 21 December 2009 (UTC)

New NHS report

New information to help support theory that measuring autism techniques have changed over time:

http://news.bbc.co.uk/2/hi/health/8268302.stm

"The NHS Information Centre found one in every hundred adults living in England has autism, which is identical to the rate in children.

If the vaccine was to blame, autism rates among children should be higher because the MMR has only been available since the early 1990s, the centre says" —Preceding unsigned comment added by 67.166.122.57 (talk) 07:47, 28 December 2009 (UTC)

Thanks, this looks like it would be worth adding to Epidemiology of autism and perhaps to MMR vaccine controversy. However, the actual NHS report doesn't seem to be available yet; at least I didn't find it easily available at the NHS web site. We can wait until it's actually published. Or if it is published somewhere now, can you please let us know where? Eubulides (talk) 09:19, 28 December 2009 (UTC)

This story is dated September. We discussed it then: Talk:Autism/Archive 9#Autism rate not increasing after all?. Colin°^Talk 11:13, 28 December 2009 (UTC)

Terminology and abbreviations

recent edit added a Terminology and Abbreviations section, describing about twenty abbreviations such as "AAC – Augmentative and Alternative Communication; communication through nonverbal means". This section is not helpful. The Autism article rightly doesn't discuss AAC, or the other ; the few places where it does use terms that have abbreviations, such as PDD-NOS, it explains them as it runs across them; that's enough. Anyway, it appears that the section was copied whole cloth from http://autismaspergerssyndrome.suite101.com/article.cfm/the_autism_dictionary, which we can't do (see Wikipedia:Copyright violations), so I reverted it. Eubulides (talk) 21:26, 29 December 2009 (UTC)

Salience

The article doesn't mention salience. It describes features which could be linked to salience but doesn't actually use the word. Is this because it's not a generally recognised term in this context, or is there some other reason? Nineteenthly (talk) 12:19, 30 December 2009 (UTC)

Salience is a generally recognized technical term among experts. Was there a particular wording change you'd like to propose? Please bear in mind that, other things being equal, it's better to use plain English. Eubulides (talk) 16:47, 30 December 2009 (UTC)

New statistics for autism in the US

I updated the statistics on autism in the US. The Center for Disease control has confirmed that 1 in 110 US children has autism. They also found that the rate of autism has increased 57% within four years. I wasn't sure how to add sources , so I'll list them here:

http://www.webmd.com/brain/autism/news/20091218/autism-jumps-57percent-in-just-4-years

First, that edit wasn't correct, as it confused the statistics for ASD (which is what the CDC report is about) with those for autism proper. Second, this is just one primary study, and although it's an important one, it probably doesn't need to be in Autism which cites multiple reliable secondary sources (reviews); see Wikipedia:Reliable sources (medicine-related articles) #Respect secondary sources. This new primary study is more appropriate for Epidemiology of autism #United States, where it is already cited. Eubulides (talk) 19:29, 21 January 2010 (UTC)

MMR

I assume most people have seen this? Wondering if it should be added somewere. http://www.guardian.co.uk/society/2010/jan/28/andrew-wakefield-mmr-vaccine Doc James (talk · contribs · email) 20:13, 3 February 2010 (UTC)

It's been added at MMR vaccine controversy and several related pages. It was also added to the lead here, but is way out of place, as this new report is merely a recent retraction of a primary study, and as per WP:PSTS we shouldn't be using primary studies (or retractions) when we have lots of reliable secondary sources on the same subject. Also, it's a pretty clear case of WP:RECENTISM. So I just now reverted the addition. Eubulides (talk) 03:31, 4 February 2010 (UTC)

Page loading efficiency and style

This page takes a long time to load, and part of this is due to the use of the standard Wikipedia citation templates such as {{cite journal}}. Recently developed faster & smaller Vancouver system templates such as {{vcite journal}} would make the page much faster to generate (roughly by a factor of two in my tests) as well as significantly smaller in terms of the HTML generated. Let's use them here; they're already in use in Heritability of autism, Epidemiology of autism, etc., and have resulted in major savings both for time and for the size of the generated HTML. Eubulides (talk) 08:28, 9 February 2010 (UTC)

New paper by Ozonoff

This might be useful, if used cautiously - a prospective study of early signs of autism. [2] WLU (t) (c) Wikipedia's rules:^simple/_complex 21:27, 17 February 2010 (UTC)

Autism cites a review of this area, Rogers 2009 (PMID 19582867). Rogers works with Ozonoff and Rogers's review (which Autism summarizes) captures this primary study's main point that signs begin after the age of six months. We are also citing the review by Landa 2008 (PMID 18253102) which makes similar points. Eubulides (talk) 21:59, 17 February 2010 (UTC)

those with autism have low levels of oxytocin

They appear to have some beneficial response to nasal administered oxytocin in a new study. [3]--68.35.3.66 (talk) 15:05, 18 February 2010 (UTC)

That's a press release and not suitable per WP:MEDRS. Even the original source article (though apparently published in PNAS, a high-quality source) would be problematic - new, unreplicated, very small sample size, preliminary and way too soon to represent a treatment modality. We're not a crystal ball and stuff like this should be treated with kid gloves until there's more evidence it's got a firm foothold in the academic community.

On a sidebar, I wonder how long it'll be before this one is picked up in the fringe community, a la mercury and testosterone chelation "therapy". WLU (t) (c) Wikipedia's rules:^simple/_complex 18:51, 18 February 2010 (UTC)

This thread is about this recently-published primary study:

• Andari E, Duhamel JR, Zalla T, Herbrecht E, Leboyer M, Sirigu A. Promoting social behavior with oxytocin in high-functioning autism spectrum disorders. Proc Natl Acad Sci USA. 2010. doi:10.1073/pnas.0910249107. PMID 20160081.

The fringe have long been talking about oxytocin, but they talk about everything, so that's not necessarily a minus. There is stronger scientific basis for oxytocin than for chelation (not that oxytocin is an accepted therapy or anything: it's still firmly experimental). It'd be better to cite a review, such as Bartz & Hollander 2008 (PMID 18655901), or Rossignol 2009 (PMID 19917212) (which gives oxytocin a grade B). Obviously these reviews don't cover this latest study, and that study might raise oxytocin's grade, but it's OK to wait for the next review (one will come along soon). Given the large number of experimental therapies (of which oxytocin is only one), Autism therapies is the best place to mention oxytocin; it already cites Bartz & Hollander on oxytocin and should also probably cite the newer Rossignol. Eubulides (talk) 20:00, 18 February 2010 (UTC)

Introductory sentence

"Autism is a disorder of neural development that is characterized by impaired social interaction and communication, and by restricted and repetitive behavior"

Your introductory sentence seems to privilege the atypical neural development in autism over other signs by beginning with "Autism is a disorder of neural development." It doesn't actually say "atypical neural development is the primary factor" or "the atypical behavioural development in autism is a product of atypical neural development," but by saying "autism is" this factor and it is "characterised by" these other factors, I believe it carries that implication. I believe this question is presently unanswered by the science: Is autism neurodevelopmental in origin or are the atypical brain structure and function a product of the disorder? Autism is equally a disorder of executive function characterised by atypical neural development and impaired affective contact and tolerance for change. Or, a disorder of affective contact, characterised by limited tolerance for change, impaired executive function and atypical neural development.

Is the first sentence implying what I think it is?
If so, should it? Anthony (talk) 09:24, 29 December 2009 (UTC)

Being a neurodevelopmental disorder is not a "factor" or a "sign"; it is a classification. For example, when the Common cold article says that the common cold is a "contagious, viral infectious disease of the upper respiratory system", this doesn't mean the upper respiratory system is a "factor" or a "sign" of the common cold: all that's being said is that the common cold is one of many diseases that are contagious and viral and affect the upper respiratory system. The wording is similar in Autism. It is standard practice in high-quality sources to define autism to be a neurodevelopmental disorder: for example, the introduction in Levy et al. 2009 (PMID 19819542) begins with "Autism is a neurodevelopmental disorder in the category of pervasive developmental disorders, and is characterised by severe and pervasive impairment in reciprocal socialisation, qualitative impairment in communication, and repetitive or unusual behaviour." Eubulides (talk) 09:44, 29 December 2009 (UTC)

Thank you Eubulides. How would you feel about changing the phrase from "a disorder of neural development" - 7 results on Google Scholar - to "neurodevelopmental disorder" - 6,840 results on Google Scholar? Can you tell me what classification system "neurodevelopmental disorder" is a part off? DSM IV? Anthony (talk) 13:07, 29 December 2009 (UTC) By the way, Neurodevelopmental disorder (which I visited before coming here) says it is an impairment of the growth and development of the brain or central nervous system, and lists a number of disorders that share this (sign?), but doesn't mention the term's role in taxonomy. Anthony (talk) 14:35, 29 December 2009 (UTC)

"A disorder of neural development" is easier to understand, and is less jargonish, than "neurodevelopmental disorder". Specialized jargon is appropriate for scholarly sources (which explains the Google Scholar results), but it is less appropriate for encyclopedia. The Wikipedia guideline for lead sections says that leads should avoid specialized terminology when possible, which is why it's worded the way it is. DSM-IV is about symptoms, not causes, and as such it doesn't talk about neurodevelopmental disorders. Eubulides (talk) 17:36, 29 December 2009 (UTC)

Thanks again. The present formulation is less scary. "Neurodevelopmental disorder" is the name of one of 5 "clusters" of mental disorders in DSM-V (I have just discovered). If what you are saying in the first sentence is "autism is one of that cluster of mental illnesses collectively known as the neurodevelopmental disorders", then it would be prudent to use the class name. If you are saying that disordered neural development is a feature of autism, along with impaired social interaction and communication, and restricted and repetitive behavior - that is, if you are referring to one of its features/signs then the sentence should read more like "Autism is a mental illness characterised by disordered neural development, impaired social interaction and communication, and restricted and repetitive behavior," not "privileging" one feature over the others. I realise I am being eccentric here. Everyone is taking it for granted that of course the etiology of autism will be located in the brain. I just think it's a little early for that. Anthony (talk) 19:54, 29 December 2009 (UTC)

I propose replacing the first sentence with

Autism is a psychiatric syndrome characterized by impairments in social interaction and communication, and restricted and repetitive behavior, all exhibited before a child is three years old. It is one of a cluster of conditions known collectively as the neurodevelopmental disorders.

Anthony (talk) 04:46, 31 January 2010 (UTC)

I don't see why that would be an improvement. It's much longer than what's there now. The phrase "psychiatric syndrome" gives the wrong connotation to typical readers, as it'll give them the incorrect impression of a child on a psychiatrist's couch. The idea that there's a "cluster of conditions" is not supported by reliable sources. Since reliable sources almost invariably call autism a neurodevelopmental disorder, not a psychiatric condition, why should this article do things any differently? Eubulides (talk) 04:51, 31 January 2010 (UTC)

Thanks Eubulides.

1. Length isn't an issue. Precision and clarity are.
2. It is a psychiatric syndrome. I don't accept a word of your opposition to calling it that.
3. It is one of that group known as the autism spectrum, which is a sub-group of that category called the PDDs, which belongs in the super-category called neurodevelopmental disorders... not the super-category called disorders of neural development. If you want to rename the category, run it by the APA first. Above, you say "a disorder of neural development" is easier to understand. I know what "neurodevelopmental disorder" means: it's a category in psychiatric taxonomy. What, precicely, do you mean by "a disorder of neural development"? That's not a rhetorical question. To me, it is laden with etiological implications. What meaning to you think those words to convey?
4. "Cluster" is used by the APA taxonomy taskforce I referred to above. I'm not wedded to it.

Anthony (talk) 07:35, 31 January 2010 (UTC)

• "Length isn't an issue" Length is an issue. This is the lead; it's supposed to be a summary of the body, not an exposition at greater length. See WP:LEAD.
• Yes, autism is a psychiatric syndrome; it is also lots of other things, including a pervasive developmental disorder, a psychological development disorder, a diagnosis, etc., etc. But the lead shouldn't be bloated by all the pigeonholes that autism can (accurately) be slotted into. It should convey concisely the most important thing that autism is, the same thing that our best and most reliable sources say. So, for example, Courchesne et al. 2007 (PMID 17964254) start off with the definition "Autism is a genetic disorder of neural development in which the first behavioral symptoms appear early in life." This is the kind of lead sentence that Autism should have. Recent reliable sources on autism avoid definitions like "Autism is a psychiatric syndrome", because such definitions would tend to mislead non-experts.
• "Disorder of neural development" is a synonym for "neurodevelopmental disorder" that is easier for the typical reader to understand. The wikilink to Neurodevelopmental disorder should help make this clear to the expert, if it wasn't clear already. I'm not wedded to this synonym: if it confuses typical readers, we can reword it.

Eubulides (talk) 08:01, 31 January 2010 (UTC)

1. Length? What I tried to say was "accuracy should not be sacrificed for concision."
2. That article by the neurobiologist Courchesne is the only peer-reviewed article addressing autism to use disorder of neural development. Google Scholar finds 3,290 articles addressing autism that use its class name - neurodevelopmental disorder. I would like to see the class name used here and it made clear that it is a class name. Perhaps:

Autism is a (insert vague, politically acceptable term) characterized by impairments in social interaction and communication, and restricted and repetitive behavior, all exhibited before a child is three years old. It is one of those conditions collectively known as the neurodevelopmental disorders.

Anthony (talk) 09:01, 31 January 2010 (UTC)

By the way, where does this come from and what does it mean? "Recent reliable sources on autism avoid definitions like Autism is a psychiatric syndrome, because such definitions would tend to mislead non-experts." Anthony (talk) 09:08, 31 January 2010 (UTC)

It came from a comment that I wrote. In what sense is it not clear? The lead should not use vague terms. Eubulides (talk) 20:01, 31 January 2010 (UTC)

We agree. My worry is that readers will take "disorder of neural development" to mean "caused by retarded or aberrant brain development". We know that it means "neurodevelopmental in origin or to have neurodevelopmental consequences when they occur in infancy and childhood". I am concerned that "disorder of neural development" sounds so obviously like "disorder (caused by aberrant) neural development" that almost all readers will not follow the link to discover in the second paragraph of Neurodevelopmental disorder that the brain atrophy and hypertrophy, the under- and overactivity may be a a manifestation of the unknown primary cause; no more causal than the executive dysfunction or stereotypies. To share my concern here you need to assume the position that the odd brain is as causal as the odd social relations. Anthony (talk) 09:32, 1 February 2010 (UTC)

The typical reader won't distinguish between "neurodevelopmental disorder" and "disorder of neural development" in that sense; both equally connote "disorder (caused by aberrant) neural development". Furthermore, there's nothing wrong with that connotation, regardless of the wording; "Although the etiology remains undetermined, a general consensus has been that some type of aberrant neural development underlies the disorder of autism." (Rice et al. 2005, PMID 15791921). Eubulides (talk) 16:51, 1 February 2010 (UTC)

Do you have a credible review of academic and research opinion on the question of etiology? The lead and corresponding authors of the paper you cite are both brain imagers looking for autism's etiology in the brain. I can't access Bailey et al. 1996 but the other paper they cite in support of their assertion, Robert G. DeLong 1999 makes no mention at all of a consensus view, but is an interesting theory of neuropathological etiology. If neuroscientists agree abnormal brain development is the present best guess concerning etiology, then even universal consensus among neuroscientists is of low value and the assumption that cause lies in the brain must be guarded against. If there is consensus with a high degree of certainty across disciplines, then it has to be reported here with evidence.

But no matter what degree of consensus you find regarding best guess etiology, until the cause is actually found, it is reckless to call it a "disorder of neural development" or even "neurodevelopmental disorder" when you concede it will be misread by the typical reader as "disorder (caused by aberrant) neural development" while the actual meaning of the term is "caused by or causing abnormal brain development". In this article it should be called what it is: a "psychiatric syndrome" (despite your worries about the cuckoo's nest connotations), and if you use "neurodevelopmental" or its synonyms you must define it so as to avoid the erroneous meaning you say the typical reader will draw. Readers of peer-reviewed journals (should) know what this term means - "causing or caused by" - but you admit your readers will misconstrue it, and say it doesn't matter. Anthony (talk) 06:56, 2 February 2010 (UTC)

• "Do you have a credible review of academic and research opinion on the question of etiology?" Sure, several are cited in Causes.
• "If there is consensus with a high degree of certainty across disciplines, then it has to be reported here with evidence." There is, in Causes and Mechanism; this is summarized in the lead.
• "it is reckless to call it a "disorder of neural development" or even "neurodevelopmental disorder" when you concede it will be misread" No, I did not concede that. Nor is it "reckless" to describe autism using the terms that reliable sources do.
• "it should be called what it is: a 'psychiatric syndrome'" No, as discussed above, reliable sources typically do not define autism that way. And for good reason: ordinary readers are likely to misinterpret the word "psychiatric" in that context, even if the word is technically correct. Besides, the word "psychiatric" is unnecessarily broad. Let's stick with a mainstream definition rather than use an oddball one.

Eubulides (talk) 07:14, 2 February 2010 (UTC)

I think our point of contention is: I believe syndromes classed as neurodevelopmental disorders may be either the product of an obvious pathology or trauma, or a case of seemingly healthy brains behaving oddly (such as autism). You believe that if a syndrome is classified as a neurodevelopmental disorder, then the aberrant behaviour and affect is the product of the aberrant neural development.

I derive my belief from two things. I read Disorders considered to be neurodevelopmental in origin or to have neurodevelopmental consequences when they occur in infancy and childhood as saying that, and when you mentioned that it is a classification, I remembered having read many years ago that it was going to form a category in psychiatric nosology. My memory of what I read back then is that it would include childhood psychiatric syndromes associated with obvious brain disease, as well as those whose mechanism is not known but which are associated with abnormal brain function. I can't now find anything describing the inclusion criteria (but I haven't looked that hard yet). That is my reason for harrying you. If it is as I remember it and as I read Neurodevelopmental disorder, then, on its own, classification of autism as a neurodevelopmental disorder says nothing about whether the brain morphology "underlies" the disorder, it simply says "this syndrome includes unusual brain function and (hence?) morphology". And I am assuming some scholars, theorists and researchers are viewing autism in that way, too.

I could be dreaming about that paper I read, and misreading Neurodevelopmental disorder. If so, I apologise for wasting your valuable time and attention. I'll come back when I've read the ref's in Causes and Mechanism. But what I was hoping for was a study testing the assertion that "a general consensus has been that some type of aberrant neural development underlies the disorder of autism." That is, evidence of the asserted consensus. I'd be interested to know what percentage of what professions believe that, and how strongly. I didn't see anything like that in a cursory scan of the footnotes to those ref's. Anthony (talk) 10:47, 4 February 2010 (UTC)

The bottom line here is that Eubulides feels that he owns this article, and will not tolerate any tampering with it. It's not a well written article by any means, and tough and misleading reading for the laymen, but any changes will be rejected by him, just because it's his article and he will take his toys and go home. —Preceding unsigned comment added by 75.61.139.215 (talk) 05:36, 14 February 2010 (UTC)

I have no problem with Eubulides' style of argument; and every word of this important medical article should be supported by authoritative secondary sources. Anthony (talk) 02:28, 20 February 2010 (UTC)

I think it should say "We don't know what autism is, but 51% of (some kind of) researchers agree that 'Autism is a disorder of neural development that is characterized by impaired social interaction and communication, and by restricted and repetitive behavior." If "diagnosis is based on behavior, not cause or mechanism...." while "Genetic screening for autism is generally still impractical...", then in what percentage of cases do diagnoses remain purely subjective? 90-95%? Or more? 70.166.123.49 (talk) 22:27, 23 February 2010 (UTC)

Geographical Frequency

I think there should be some mention of the geographical frequency of autism in the epidemiology or some similar section. So far, most of the figures mentioned in the linked article on Epidemiology of Autism, focuses on the effects of the MMR vaccine, most of the figures from other countries are unknown or inconsistent. Maybe there is some research or a component to it that depends upon geographical or racial factors to the epidemiology, some countries have ~1% prevalence rates while some have far below even those. Since most of the research cited is from the US, at least a mention of some other (non-US) countries might be helpful with the context. --Theo10011 (talk) 01:48, 19 February 2010 (UTC)

We'd need sources to compare the geographic (really national comparison is what it sounds like you're aiming at) differences. I would guess those 'low' rates have more to do with diagnosis than environmental differences given the genetic prevalence, and it'll take reliable sources to sort it out. WLU (t) (c) Wikipedia's rules:^simple/_complex 11:49, 19 February 2010 (UTC)

I don't know of any reliable sources on national comparison, as the data simply aren't there. For example, as far as I know nobody has ever done a scientific study of autism prevalence in any country in Africa. As for race, the best I've been able to find is what's currently in the article "Most professionals believe that race, ethnicity, and socioeconomic background do not affect the occurrence of autism.", citing Bertoglio K, Hendren RL. New developments in autism. Psychiatr Clin North Am. 2009;32(1):1–14. doi:10.1016/j.psc.2008.10.004. PMID 19248913. Eubulides (talk) 00:37, 20 February 2010 (UTC)

I wanted to mention that the individual article linked to from the main page to the Epidemiology of Autism already cites some inconsistent frequency. I have been unable to locate a relevant source based on geographical/racial factors maybe because as Eubulides pointed out that a nationwide comparison might not be available, the epidemiology page however cites figures from many countries in relation to the effects of MMR vaccine, they already cite some inconsistent frequency or completely different in case of countries like Japan. Its unlikely that any stats from African countries on autism would be available but maybe there is some research backing up the MMR findings from some EU countries where prevalence is low. Maybe you can consider adding a mention or summarizing its epidemiology, based on the epidemiology of autism article as a source itself.--Theo10011 (talk) 05:48, 23 February 2010 (UTC)

Figures also differ in different countries due to their different diagnostic criteria. I don't know of any reliable source saying that the true prevalence of autism differs in different countries. I don't think we can say anything along those lines in Autism, as that would be original research. Eubulides (talk) 06:23, 23 February 2010 (UTC)

Wikipedia itself is not a reliable source; until there are transnational comparisons published in the literature, we can't add anything to that effect to the main page. WLU (t) (c) Wikipedia's rules:^simple/_complex 16:15, 23 February 2010 (UTC)

Ok thanks for all your help, I see your point. I'll keep an eye out for any relevant information until then.--Theo10011 (talk) 21:22, 23 February 2010 (UTC)

molecular autism

There is a new journal that perhaps should be referenced on the page. Molecular Autism. I notice from the above discussions, that due to control freaks, the talk page may end up being more informative than the page itself. That is not uncommon on wikipedia. I prefer a more open philosophy, where the page contains all possible information people might want to consider persuing further. Kinda the ultimate home page rather than the ultimate authority on the topic. Leave to the reader to determine whether just one or two peer review articles are enough to establish a scientific fact or medical therapy. Of course, overstatement of the evidence must be avoided, and contrary evidence should be given its say. None of this is relevant to this new journal though. I just thought it needed saying. --68.35.3.66 (talk) 22:10, 3 March 2010 (UTC)

I got a 404 when I went to that link. Anyway, we follow WP:MEDRS which likes secondary source. You also might want to look at WP:NOT. Dbrodbeck (talk) 22:46, 3 March 2010 (UTC)

Molecular Autism is a new journal that looks like it will be a good publisher of reliable sources in this area, particularly since it's open access. Currently it has published only 3 research papers, all bleeding-edge primary sources, and none of them seem appropriate for Autism. It does plan to publish a review of association studies, which should be useful either for Autism or for Heritability of autism (or both). Eubulides (talk) 00:18, 4 March 2010 (UTC)

No doubt about that, once it gets some content. Dbrodbeck (talk) 01:24, 4 March 2010 (UTC)

Being that new makes it hard (more likely impossible) to judge the impact factor for a while. I'm assuming we're only using this for reviews that don't propose anything surprising or in contradiction of much of the background information? Almost goes without saying, but extra caution should be used until it has established a reputation for being reliable, mainstream and not advocating fringe ideas. WLU (t) (c) Wikipedia's rules:^simple/_complex 18:17, 5 March 2010 (UTC)

People first language

A recent edit methodically went through the article replacing many phrases like "autistic children" with phrases like "children with autism". The edit also made a few other changes, such as replacing "mental retardation" with "developmental disabilities" and "autistic disorder" with "autism"; in some cases the change broke some wikilinks entirely. These global changes aren't appropriate. First, people-first language is controversial in the autistic community, and my impression is that it is more often opposed than supported. Second, this article should use the terminology of the sources: if the sources talk about "mental retardation" the article shouldn't change that to "developmental disability", as they are not the same thing. In particular, the article should quote the titles of sources correctly (the change also affected these titles). I undid the change; further comments are welcome here. Eubulides (talk) 00:08, 5 March 2010 (UTC)

I fully support the revert, for the reasons listed. Dbrodbeck (talk) 01:07, 5 March 2010 (UTC)

I suspect they jus used a find-replace tool to make the changes, as I wouldnt expect someone to purposely break wikilinks. —Soap— 02:08, 5 March 2010 (UTC)

Fine with me. Do we refer to the people-first language and controversy somewhere in the article? I see it here in the sociological and cultural aspects article, perhaps a note in the main page might be helpful. This is a minor point but one that could be clarified to the benefit of even the main article I think - and without adding a lot to the length. WLU (t) (c) Wikipedia's rules:^simple/_complex 18:26, 5 March 2010 (UTC)

The logical place for this sort of discussion would be the last paragraph of History, but I don't know, all the stuff that's in there right now is so much more important than the people-first tempest-in-a-teacup that I'm inclined to leave things alone. Eubulides (talk) 19:34, 5 March 2010 (UTC)

Rather than an explicit discussion in the body, I was thinking of something like this, though I know you replaced it with a footnote here. I like the asterisk-at-the-bottom-of-a-page approach rather than a footnote because it makes it obvious it's not a footnote/citation. I think it's a reasonable inclusion that, particularly given the anti-cure POV within the community, could improve the page. I'm not wedded to the idea, but it would a) head off future find-and-replace stuff and b) address how some people adopt it as an identity - all without adding significantly to the length. WLU (t) (c) Wikipedia's rules:^simple/_complex 20:27, 5 March 2010 (UTC)

Agree with Eubulides about the people-first aspect wrt autism. However, I wouldn't necessary always follow "the terminology of the sources", though we must be very careful to use equivalent and unambiguous alternatives if we differ. I suspect most of the sources are American but a UK journal may use different terminology. The "mental retardation" phrase is an WP:ENGVAR issue. See our article on Mental retardation and the hatnote of Learning disability. Colin°^Talk 20:26, 5 March 2010 (UTC)

Social deficits

Their social deficit lies in the fact that they have difficulty identifying emotions and they possess behaviors that are socially inappropriate not that they are incapable of cooperative behavior or the ability to possess a theory of mind (Downs & Smith 2004).

Downs, A., & Smith, T. (2004). Emotional understanding, cooperation, and social behavior in high-functioning children with autism. Journal of Autism and Developmental Disorders, 34(6), 625-635. —Preceding unsigned comment added by Psyjsay (talk • contribs) 05:16, 5 March 2010 (UTC)

Might be useful for Communication. WLU (t) (c) Wikipedia's rules:^simple/_complex 18:33, 5 March 2010 (UTC)

Downs & Smith 2004 (PMID 15679183) is a reasonable primary study, but it's better to use reviews for this sort of thing. The article already says "However, most studies have found no evidence of impairment in autistic individuals' ability to understand other people's basic intentions or goals; instead, data suggests that impairments are found in understanding more complex social emotions or in considering others' viewpoints.", citing the review of Hamilton 2009 (PMID 19508497); doesn't that already cover the point adequately? Eubulides (talk) 19:34, 5 March 2010 (UTC)

That still seems like something that might be better referred to in the communication section, I looked for it in communication because it would seem to fit best there. The explicit link to neuropsych might be more obvious in the source because in the text it doesn't really jump out at me. But it's certainly a judgement call. WLU (t) (c) Wikipedia's rules:^simple/_complex 20:32, 5 March 2010 (UTC)

disorder of neural development

is it or not —Preceding unsigned comment added by 65.175.247.51 (talk) 11:37, 11 March 2010 (UTC)

Vaccines: Article contradicts itself

Vaccines do not cause autism. Use & connotations of "implausible".

The article states in two places that "the vaccine hypotheses are biologically implausible and lack convincing scientific evidence," yet in Mechanism/Pathophysiology it states, "successful neurodevelopment depends on a balanced immune response. It is possible that aberrant immune activity during critical periods of neurodevelopment is part of the mechanism of some forms of ASD."

So which is it? Either tampering with the immune system is a "biologically implausible" or "possible" cause of autism.

I read the four citations under reference #7 and found them unconvincing, primarily because they are abstracts.

7.1 (Doja) is an abstract with no data and makes extremely general conclusions.
7.2 (Gerber) is available full-text, but is an overview of other studies. It offers no data, and all of the citations are abstracts.
7.3 (Gross) is highly opinionated, and all its citations are abstracts.
7.4 (Paul) is highly opinionated, and offers no data or citations.

In other words, the statement that vaccines do not cause autism is unsupported.

What's worse, the fulltext article 7.3 (Gross) [4] makes a number of revealing admissions.

First, it references a study of 124,000 children where thimerosal was found to have "significant positive association with tics," and "increased risks of language delay." (From his citation #5: Safety of thimerosal-containing vaccines [5])

Gross goes on to admit that no toxicity guidelines existed for ethylmercury while it was being used. In particular, the risks of methylmercury were well known, and the dosages for ethyl- routinely exceeded those established for methyl-.

He then goes on to lament that "Had the discovery about thimerosal come at a different time, it might have gone unnoticed." This shocking sentence implies that the author is aware of problems with thimerosal, and would rather it be kept quiet!

Next, Gross highlights a case where MMR vaccine exacerbated an infant's underlying autism-like symptoms. Fortunately for Gross, "The condition shares symptoms with ASD, but is distinct." In other words, even though the patient now has exacerbated ASD symptoms from taking MMR, it's okay because it's not actually autism.

I apologize for the length, but this is your citation. If you were willing to take into account what people are saying about these studies, you could very quickly reach an opposite conclusion. Example, look up Dr. Poul Thorsen.

Scientific abstracts are little more than unsupported opinion. It's fine to include expert opinion, as long as you say, "Scientist X claims Y and Z."

What you have done is to label unsupported opinion as fact, and failed to include the dissenting view. OrangeCatholic (talk) 09:14, 12 March 2010 (UTC)

• "Either tampering with the immune system is a "biologically implausible" or "possible" cause of autism." The two sentences do not conflict at all. The "critical periods of neurodevelopment" in question are within a few weeks of conception. Vaccines are given to children after they are born.
• "primarily because they are abstracts" "all of the citations are abstracts" The cited sources are the entire papers in question, not just the abstracts. Wikipedia articles are allowed to cite sources that are not freely readable. It's common for abstracts to be freely readable even when the full articles are not.
• "(Doja) is an abstract with no data and makes extremely general conclusions" Doja & Roberts 2006 (PMID 17168158) is a 6-page peer-reviewed medical review that goes into a lot of detail about the studies its examines. Its conclusion, which is that the evidence overwhelmingly shows no causal connection between vaccines and autism, follow from its detailed examination.
• "(Gerber) is available full-text, but is an overview of other studies." Yes, and that's exactly what's wanted for Wikipedia articles on medical topics. Please see Wikipedia:Reliable sources (medicine-related articles) #Respect secondary sources.
• "(Gross) is highly opinionated" "(Paul) is highly opinionated" These are reliable sources published in peer-reviewed biomedical journals. They are presenting the mainstream scientific view, which is what Wikipedia is supposed to do. We can't let our opinions override those of experts in the field. If there were serious scientific dispute over these matters that would be a different question, but there is not.
• Gross 'references a study of 124,000 children where thimerosal was found to have "significant positive association with tics," and "increased risks of language delay."' That's not how Gross summarizes the study. Here's what Gross says about it:

  "Meanwhile, CDC scientists continued their investigations of vaccine safety—prompting an angry backlash. After releasing a 2003 study of more than 140,000 children that showed no relationship between thimerosal and autism, the CDC received such disturbing threats that agency officials called in federal investigators [5]."

The other comments you make about Gross's article reflect similar misreadings. At any rate, the Autism article is about autism, and not about tics or language delay or mitochondrial disorders, so this isn't really the proper forum to thrash out those disagreements with what Gross says.

• 'It's fine to include expert opinion, as long as you say, "Scientist X claims Y and Z."' There is no need to write "Professor Paul Sereno says that dinosaurs evolved from archosaurs." It suffices to write "Dinosaurs evolved from archosaurs." and cite Paul Sereno's paper. This is true regardless of the fact that many people don't believe in evolution, because the opinion that evolution is hokum is a fringe opinion that is not supported by mainstream science. Things are different when there is a genuine scientific disagreement about a topic, but that's not the case here, any more than it's the case with dinosaurs and evolution.

Eubulides (talk) 18:56, 12 March 2010 (UTC)

They're not "misreadings." You call them misreadings because they conflict with his overall point. Did you read it, or am I the first? A proper scientific article asks a question, provides examples, and then follows them to a conclusion. Proper science is humble because the data may complicate or disprove the original question.

Gross' article reads like a political essay, in that he assumes a conclusion (vaccines are safe and should be taken by everyone), and then attempts to disprove a small number of counter-examples as a way of making the original conclusion seem like it was always true to begin with. Here are some of his counter-examples and the way he handled them:

• Ethylmercury is related to an unsafe chemical and was totally unstudied at the time it was being used. But, it was proven safe later on. Take that, parents!
• The discovery that an unstudied chemical was being injected into children was politically inconvenient for the pro-vaccine lobby, rather than a legitimate cause for alarm.
• A case where MMR vaccine "aggravated an underlying mitochondrial disorder" causing "regressive encephalopathy with features of autism spectrum disorder" does not conflict with a policy of vaccinating everyone.

As for dinosaurs, what if you had written that dinosaurs evolving from archosaurs was "biologically implausible and lacking in convincing scientific evidence"? You would look foolish, no? I'm old enough to remember when dinosaurs evolved from lizards, when the "bird" theory was a fringe belief. All scientific facts are likewise opinions. Whether they should be taken as "facts" depends on

• Access to data that allows a reader to confirm the opinion for himself, not just "A bunch of scientists say it's true, so it is."
• An analysis of the opposing belief, to confirm that the "fact" is not really controversial.

In the case of vaccine safety, you've done neither. You have a lot of expert opinions that it's safe, but no data, and no treatment of the opposite side.

For example, to write that evolution is fact, you would need at least some data (such as access to the butterfly study showing them turning darker during London's industrial revolution). You should also perform a cursory analysis of the opposing side (creationism), which is rife with flaws. Stating that evolution is fact "because a bunch of scientists say so", while only providing their abstracts, is not sufficient because it's just an opinion without public data to back it up.

Get it? Science isn't supposed to be secret. If evolution is so easy to prove, then at least one study should be available to the public to see for ourselves. Consider the case of a young girl growing up in a creationist town, you wouldn't be convincing her of evolution by citing only abstracts. You need evidence.

>Wikipedia articles are allowed to cite sources that are not freely readable.

Well I disagree with this policy 100%. How can a "free and open" encyclopedia cite sources that can't be verified? It's extremely naive to assume that an abstract's conclusion was obtained via proper scientific method. Rigging studies to show the "proper" conclusion is widely practiced, especially when money is involved.

I would like to read a study that compares autism rates between vaccinated and unvaccinated children. I can't find one, and I don't see that you guys have found one either. Otherwise we could read it together and have a discussion. OrangeCatholic (talk) 10:21, 13 March 2010 (UTC)

Please se WP:MEDRS, we look at secondary sources, reviews, as much better than primary ones. A single article is a data point, a review has many data points. You are indeed missing the point here, the scientific consensus is that vaccines do not cause autism. That consensus, which is referenced, shows up in the present article. Dbrodbeck (talk) 12:17, 13 March 2010 (UTC)

Because of Wikipedia's popularity (9,000 people per day consult this article) and editors' lack of accountability, there are constraints around what we can include in a medical article. With a very few exceptions, every statement is supported by at least one recent review article from a prestigious peer-reviewed journal. Often, the most recent article is only free to university students and staff, and members of professional bodies. We all lament this. (Some university libraries offer "community access" membership at low rates.) But reviews and textbooks are what our medical articles are built on. No one is hiding anything here. We're working responsibly within necessary constraints.

If a review reports that the current scientific consensus is weakening, this article will reflect that by representing the differing views: "Scientist X claims Y and Z; but scientist A claims B and C." Anthony (talk) 13:11, 13 March 2010 (UTC)

>the scientific consensus is that vaccines do not cause autism

Then say that! I'd be much more comfortable with that statement, which roughly translates to, "The opinion of the vast majority of vaccine scientists is that vaccines do not cause autism," than the current statement, which is, "The vaccine hypotheses are biologically implausible and lack convincing scientific evidence."

Here's what's wrong with the current statement:

1. Implausible? It's perfectly plausible, otherwise it wouldn't keep coming up. Mercury is a known toxin, particularly a neurotoxin. This is common knowledge. It's perfectly "plausible" for a child's neurological problems to be caused by injecting them with a neurotoxin, and Gross admitted as much in the case of Terry and Jon Poling. He even called it a "table injury," meaning it was part of a standard table of side effects.

It's also "plausible" that an infant's immune system can't handle 15 viral loads administered 32 times [6]. Here are some quotes from a 2007 Discover article that brings up the immune system repeatedly [7]:

“I no longer see autism as a disorder of the brain but as a disorder that affects the brain,” Herbert says. “It also affects the immune system and the gut. One very striking piece of evidence many of us have noticed is that when autistic children go in for certain diagnostic tests and are told not to eat or drink anything ahead of time, parents often report their child’s symptoms improve—until they start eating again after the procedure. If symptoms can improve in such a short time frame simply by avoiding exposure to foods, then we’re looking at some kind of chemically driven ‘software’—perhaps immune system signals—that can change fast. This means that at least some of autism probably comes from a kind of metabolic encephalopathy—a systemwide process that affects the brain, just like cirrhosis of the liver affects the brain.”

In his 1943 paper, “Autistic Disturbances of Affective Contact,” Kanner noted immune and digestive problems but did not include them in the diagnosis. One reads with a shiver sentences lifted out of various case histories: “large and ragged tonsils . . . she was tube-fed five times daily . . . he vomited all food from birth through the third month . . . he suffered from repeated colds and otitis media. . . .”

“Everyone was focusing on genes expressed in the brain,” says Levitt, “but this gene is important for repair of the intestine and immune function. And that’s really intriguing because a subset of autistic children have digestive and immune problems.”

2. Convincing scientific evidence? This gets back to the issue of abstracts. You can't say that there's scientific evidence to the contrary without providing such evidence. You can say scientists believe their own research, but without access to the research, who knows for sure? Scientists protect their own avenues of funding, and they can be arrogant. Einstein was famously opposed to quantum mechanics for reasons that I can only conclude were psychological (i.e. ego or religious) in nature. He must have known it was true, but played devil's advocate simply because it was over his head.

>We all lament this....But reviews and textbooks are what our medical articles are built on. No one is hiding anything here. We're working responsibly within necessary constraints.

I'm sure you're working responsibly, but the constraints are imo irresponsible. It's not enough to lament access to the data. Peer review was supposed to separate the wheat from the chaff, but in the internet age, peer review can just as easily promote the chaff and who's the wiser?

>Because of...editors' lack of accountability

Maybe true, but to me, the whole point of opening it up to the world is that many of us have college degrees and are perfectly capable of examining the methods used in a scientific study. Maybe it's time for some degree of accountability and non-anonymity. If Wikipedia took itself 100% seriously and was 100% apolitical, there's no reason for me to hide my identity or qualifications. I could be getting writing credits and academic acclaim by contributing to an encyclopedia. But you're right, I could be a 15-year-old high school kid, and that's the problem.

Anyway, the "abstracts" issue is getting beyond the scope of this talk page and I'll take it up elsewhere. In the meantime, I'll take your opinions on the "implausibility" that the immune system plays a role in autism. OrangeCatholic (talk) 14:21, 13 March 2010 (UTC)

Hey Eubulides, when you said "Things are different when there is a genuine scientific disagreement about a topic, but that's not the case here" you are seriously wrong. Lots of pediatricians and even PHDs say that thimerasol has a documented link and even the US military scientists discovered the link over 20 years ago. Google it to see the truth! You are tricked by the MSM.Bbltype 21:27, 13 March 2010 (UTC)

Find a reliable source that says that. The source must conform to WP:MEDRS Dbrodbeck (talk) 22:00, 13 March 2010 (UTC)

1. Implausible? It's perfectly plausible, otherwise it wouldn't keep coming up. By those standards, the moon landing hoax, 7 day creation, astrology, holocaust denial and a lot of other things are also perfectly plausible. --U5K0 (talk) 21:57, 13 March 2010 (UTC)

Plausibility has no necessary connection to truth, and is in the eye of the beholder.[8] One of the two authors of the cited article is a co-inventor and patent holder of the rotavirus vaccine Rotateq.[9] This article can do without the term. Anthony (talk) 08:22, 14 March 2010 (UTC)

No, the term "biologically plausible" is a standard one in the field: among other things it is used by the vaccine court to determine whether damages can be awarded. If you can show that it is biologically plausible for a vaccine to cause autism, and you can show that your child developed autism after getting the vaccine, then you will be awarded damages; it's not necessary for you to prove that the vaccine caused the autism. So, for example, it's biologically plausible for a chickenpox vaccine to cause chickenpox, and for the resulting chickenpox to aggravate a pre-existing mitochondrial condition, so if you can show that your child had a mitochondrial condition, and got chickenpox after being vaccinated by a chickenpox vaccine, and that the condition got worse, then you can collect damages. However, the vaccine court has consistently ruled that there's no biologically plausible connection between vaccines and autism, with statements such as the petitioners "failed to prove that their theory of vaccine-related causation is biologically plausible" (Moreno JA, "It's just a shot away: MMR vaccines and autism and the end of the Daubertista revolution", 35 Wm. Mitchell L. Rev. 1511 (2008–2009)). That quote was for the MMR vaccine theory, but the same goes for the mercury theories; see for example, Maugh TH II, Zajac A. 'Vaccines court' rejects mercury–autism link in 3 test cases. Los Angeles Times. 2010-03-13. Eubulides (talk) 08:55, 14 March 2010 (UTC)

Agree with keeping "implausible". The fact that this link keeps coming up despite all evidence to the contrary shows it has become a "zombie argument" (Ben Goldacre: "arguments which survive to be raised again, for eternity, no matter how many times they are shot down.") I am intrigued how the "co-inventor and patent holder of the rotavirus vaccine" fact keeps being mentioned like it is an allegation not something utterly praiseworthy. It is a topsy turvy world sometimes. Colin°^Talk 10:42, 14 March 2010 (UTC)

O.K. I've read Moreno. The vaccine court found that the petitioners “failed to prove that their theory of vaccine-related causation is biologically plausible.” What about attributing "the vaccine hypotheses are biologically implausible" to Moreno (and, hence, the vaccine court, 2009) rather than Gerber and Offit (who refer to Nelson & Bauman, 2003). My problem is not with Offit's integrity. I don't know the man. I am very conscious of an obvious potential conflict of interest, and the impact it has on parents deciding whether or not to vaccinate their children. Moreno is more recent and rests on a judge's finding. Simply more plausible. Anthony (talk) 12:02, 14 March 2010 (UTC)

I have no problem citing Offit. Show me the part of WP:MEDRS that says we must avoid sources authored by experts who are the targets of well-funded pressure groups using spurious allegations of confict-of-interest to discredit them. Perhaps the parents should more closely investigate the COI of those who peddle this rubbish rather than being quite so gullable. When we end up trusing TV personalities and actors more than scientists, it is very sad. Is there a reliable source saying Offit's paper is poor? Colin°^Talk 12:23, 14 March 2010 (UTC)

My problem is with the word "plausible". I use it to mean "believable, in the absence of proof". So, when I read "the vaccine hypotheses are biologically implausible", it jarred with that understanding. Then Eubulides pointed out I was overlooking "biological" and it seems that "biologically implausible" means "doesn't fit biological theory." Now it makes sense.

I'm not arguing the science here. It's the language, the English expression of the article, the quality of presentation, the likelihood that the maximum number of readers will come away from it better informed and understanding correctly what you are trying to say. I'm worried that an open-minded reader will read "the vaccine hypotheses are biologically implausible" and, without the benefit of Eubulides explanation, think "No they're not. Are you calling me credulous?" and turn the page. I don't want you turning people off, not even realising it, by using language, the vulgar meaning of which you have lost sight of. Anthony (talk) 16:22, 14 March 2010 (UTC) Just to be clear: I think using "implausible" is a mistake but, if you're going to keep it, referencing Offit is a rhetorical error. Anthony (talk) 16:33, 14 March 2010 (UTC)

I'm not not really following the problem with "implausible". A biological understanding prevents us believing it is possible, even without having evidence. The homoeopathy article uses similar wording for it not being "scientifically plausible".

I think I see your argument for avoiding Offit: that it weakens our argument in a debate. However, WP is an encyclopaedia, not a blog or piece of journalism. We don't have any arguments to win, just information to present. Once we choose sources or our wording for rhetorical effect, we become POV-pushers. If people choose to disbelieve a fact here because we cite Offit and because of what they've read and believed elsewhere about him, that's not really our problem. Colin°^Talk 17:56, 14 March 2010 (UTC)

This is arguing, which I am enjoying enormously, by the way; rhetoric is about communicating effectively. I just think "implausible" in common parlance has a vaguely insulting connotation that it seemingly doesn't carry in medicine. I wouldn't use it about the ideas my readers may be tentatively entertaining. But if I were going to frame my message in those terms, I'd have it delivered by a judge. I'm repeating myself. I think I have run out of things to say on this matter. Anthony (talk) 19:39, 14 March 2010 (UTC)

I'm glad you are enjoying this and are helping to improve this article and others. Perhaps there's a cultural thing -- judges in UK are often considered wig-wearing fuddy duddies, not practitioners of scientific thought. Do you think it is harmful for our readers to be told, in no uncertain terms, that an idea they had been "tentatively entertaining" was in fact implausible? I remember a chemistry PhD telling me that they had long considered homoeopathy to be just one of those alternative medicines that may or may not be worth trying and might turn out to have value. But then they looked into it and realised just how "implausible" it was. To realised a concept we are entertaining (with conviction or tentatively) is complete nonsense is a bit shocking to the system, but healthy no? Colin°^Talk 20:25, 14 March 2010 (UTC)

• Do you think it is harmful for our readers to be told, in no uncertain terms, that an idea they had been "tentatively entertaining" was in fact implausible? Harmful to their understanding if the word carries for them the "only a f***wit would believe that twaddle" connotation that "implausible" carries for me. This may be similar to a dialogue Eubulides and I had at the top of the page. I wanted him to say autism is a psychiatric syndrome, but he wouldn't touch it with a barge pole because of the connotations it carried. It's all about impressions, really. Mind you, there is no denying psychiatric syndrome is pretty loaded. This connotation I'm reading into "implausible" is subtle. And I'm not going to lose sleep over it. Speaking of which. Anthony (talk) 21:29, 14 March 2010 (UTC)

Good lord. This hypothesis has been investigated and discredited multiple times since Andrew Wakefield originally presented his hypothesis as a means of promoting his alternative vaccine for measles. The scientific consensus is that vaccines are not related to autism. The only people who keep this going are nonscientists (or discredited scientists). Any recent, high-quality reliable sources that support this claim? There are not, and all the original research in the world isn't going to change the page. WLU (t) (c) Wikipedia's rules:^simple/_complex 01:29, 15 March 2010 (UTC)

Overwhelmingly considered

A recent edit (twice made, and twice reverted) made this change to the lead:

"the vaccine hypotheses are overwhelmingly considered by the scientific community to be biologically implausible and lacking convincing scientific evidence"

This insertion is incorrect. First, there's no need for in-text attribution when there is no controversy among reliable sources, which is the case here. Second, the cited sources do not say "overwhelmingly considered by the scientific community", or anything like it; they merely say that the hypotheses are biologically implausible and that they lack convincing scientific evidence. We Wikipedia editors shouldn't be inserting our own opinions about how overwhelming things are. Eubulides (talk) 08:31, 2 March 2010 (UTC)

This kinda feeds into the belief that there's some fire behind the smoke still. I would prefer a stronger wording - to any reader with science background, the statement is unequivocal. To someone without that background, it looks like equivocation and leaves the door open still. How about something like "despite much/years of/extensive/adjective research, no association has been found." Really, are there any good studies to link them? WLU (t) (c) Wikipedia's rules:^simple/_complex 18:14, 5 March 2010 (UTC)

OK, how about changing "Although these theories lack convincing scientific evidence and are biologically implausible," to "Although these theories are biologically implausible and scientific studies have consistently found no relationship between vaccines and autism,"? Eubulides (talk) 19:34, 5 March 2010 (UTC)

That works for me, much better in my mind. Research has found against this hypothesis, it's not like we're waiting for the research to come back in. That nicely balances "frankly, we don't expect much here" with "we've looked into it and found bupkus" without veering into absolutes. WLU (t) (c) Wikipedia's rules:^simple/_complex 20:19, 5 March 2010 (UTC)

Research has found against this hypothesis? What research? You guys haven't provided a single study! The sentence should say, "Medical researchers have so far failed to convince a growing number of concerned parent groups that infant vaccines are not linked to autism." Now that would be a true statement. OrangeCatholic (talk) 09:23, 12 March 2010 (UTC)

There have been lots of studies about hypothesized vaccine–autism links, and they've come up negative. For more, please see the #Vaccines: Article contradicts itself thread below. Eubulides (talk) 18:56, 12 March 2010 (UTC)

Assuming a study is true because it says so is bad journalism. Even if I coughed up the $30, I'd still accomplish nothing unless you guys all did the same. The paywall effectively makes the data secret, and secrets do not make for public discourse.

Here's an interesting thought problem: If I purchased a study, and included quotes from it in this article under fair use, how would you guys know if I was lying? OrangeCatholic (talk) 10:30, 13 March 2010 (UTC)

We would assume good faith and get to a library if we were really suspicious. You can get, at any good university library, free access to any scientific journal they have in print version. We can request reprints from the authors. We can request PDF reprints from other editors. We could go to the websites for the NIH, CDC, AMA, WHO, Health Canada, PHAC, and dozens of other nonpartisan, highly-reliable sources for free, reliable information about the links between autism and vaccines (i.e. the lack of a link). But accusing other editors of lying and misrepresenting sources, particularly to promote such a widely-discredited hypothesis (science has concluded vaccines are unrelated to autism, DAN and Jenny McCarthy's pretences aside) won't get you any credibility. Editors (particularly new editors) who use accusations to attack other editors and claim there is a link between autism and vaccines have essentially already made up their minds. This is unfortunate as that's not how things work - good research has already been done demonstrating there is no link between vaccination and autism. And if you claimed there were well-published authors writing recent articles about a link in high-quality journals, I would first ask for the source, then check it out. Of course, since it would almost certainly be wrong, you would quickly lose all credibility.

In any case, wikipedia is an encyclopedia, it's not a place to promote theories, particularly if they are discredited. A study is assumed to be true because of the honesty of scientists, replication and peer review. This topic was a blind alley for years, and this is unfortunate. Fortunately, now it is scientifically over and dead. WLU (t) (c) Wikipedia's rules:^simple/_complex 01:41, 15 March 2010 (UTC)

FAQ

I'm surprised the page doesn't have a {{FAQ}} like talk:Evolution. Rather than re-inventing the wheel each time someone brings up the vaccine-autism chestnut, it would be nice to demonstrate consensus through a page we could point to immediately. WLU (t) (c) Wikipedia's rules:^simple/_complex 12:39, 15 March 2010 (UTC)

A tremendous idea. Dbrodbeck (talk) 13:29, 15 March 2010 (UTC)

Started at Talk:Autism/FAQ. It's light on references and doesn't have any references to previous discussions - also probably biased by my own limited knowledge to just a couple areas of autism (most notably the autism/vaccination link). Please expand. WLU (t) (c) Wikipedia's rules:^simple/_complex 14:39, 15 March 2010 (UTC)

Thanks for taking this on. One thought is that we can expand it as more discussions come in. Eubulides (talk) 19:02, 15 March 2010 (UTC)

Feel free to give suggestions for me to fill out. The biggest thing I would need help with is what the recurring issues are on the page. So far I have 7: vaccination, vaccine additives, treatments (twice), causes, person-first language (eliminating the need for a footnote!) and savant syndrome. I can start digging for resources and references in a bit. WLU (t) (c) Wikipedia's rules:^simple/_complex 19:08, 15 March 2010 (UTC)

Maybe something about the possible DSM changes in 2013? Dbrodbeck (talk) 20:49, 15 March 2010 (UTC)

Have we had a lot of questions about that? I don't monitor the talk page a lot so I don't know what questions are frequent. I see it mostly for addressing the vaccine-autism connection.

That does spark a question - has anyone on the talk page asked about increasing rates over the years? WLU (t) (c) Wikipedia's rules:^simple/_complex 21:25, 15 March 2010 (UTC)

On the DSM thing, yeah, sorry I was thinking about the Asperger's page. On the rates, I cannot remember it coming up much but I may be wrong. Dbrodbeck (talk) 21:45, 15 March 2010 (UTC)

Undent. Posting the FAQ. WLU (t) (c) Wikipedia's rules:^simple/_complex 01:12, 26 March 2010 (UTC)

MIND Institute research

Recent peer reviewed research at MIND Institute, UC Davis Medical School:

There is fairly recent research at UC Davis' MIND Institute, in the Autism Phenome Project, which indicates a particular pattern of maternal antibodies to fetal brain is found in a little over 10% of mothers of autistic children, and almost never in mothers of typically developing children.

As such, this antibody pattern is a very strong marker for autism and mothers of one autistic child need to be aware of it so, if they have the pattern, they can avoid becoming pregnant until the meaning of the pattern, and possible treatment, is known.

Disseminating this information could prevent a great many cases of autism and be a great boon to a lot of people.

After all the discussion about changing this article in the past, I am posting this here for discussion first.

The research paper is on PubMed and can be found by searching "Maternal antibodies" and "Autism". It is not a review article, as quoted below, Wikipedia rules allow peer reviewed papers as a source if properly identified as such and not given undue weight, and I suggest the finding be included in the article per those rules. —Preceding unsigned comment added by 75.61.139.215 (talk) 05:26, 14 February 2010 (UTC)

I expect that the previous comment is talking about the following, which is a review, not a primary study:

• Enstrom AM, Van de Water JA, Ashwood P. Autoimmunity in autism. Curr Opin Investig Drugs. 2009;10(5):463–73. PMID 19431079.

However, this is published in a relatively obscure journal that I don't have easy access to. Eubulides (talk) 05:56, 14 February 2010 (UTC)

I should not have been so lazy as to post without getting the proper information here for discussion. The comment above refers to this study in particular, I am copying and pasting the entire abstract from www.pubmed.gov: Neurotoxicology. 2008 Mar;29(2):226-31. Epub 2007 Nov 6.

Autism: maternally derived antibodies specific for fetal brain proteins. Braunschweig D, Ashwood P, Krakowiak P, Hertz-Picciotto I, Hansen R, Croen LA, Pessah IN, Van de Water J.

Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, CA, USA.

[See PMID 18078998 for abstract.]

I believe this research, which I am assured is being confirmed by the same researchers, should be included in this article. It could prevent a significant number of autism cases. Off the top of my head, something like : Relatively recent research has indicated that about 10% of autism cases are associated with a pattern of maternal antibodies to fetal brain not found in the mothers of typically developing children or children with developmental delays but not autistic. Researchers hope to develop a commercial test for this antibody pattern which will identify mothers at great risk of having an autistic child"

I am not sure if MIND should be mentioned by name, the test is not yet commercialized, but it's something the mothers of one autistic child really should be checked for. Even at only 10% of all autistics, and even considering that it would probably only be given to mothers of autistics, at least to start, it's still potentially a huge benefit, because it covers more people than most genetic causes that have been found.

By the way, I did not ask anyone at MIND about this, I am the father of a study participant, and my wife does not have the antibody pattern, but my reading indicates it's a big advance. —Preceding unsigned comment added by 75.61.141.31 (talk) 21:25, 16 February 2010 (UTC)

Thanks, here's a fuller cite to the study you mentioned, and I've taken the liberty of replacing the abstract in your comment (which unfortunately we can't reproduce here fully, due to copyright restrictions) with a pointer to the abstract:

• Braunschweig D, Ashwood P, Krakowiak P, et al. Autism: maternally derived antibodies specific for fetal brain proteins. Neurotoxicology. 2008;29(2):226–31. doi:10.1016/j.neuro.2007.10.010. PMID 18078998.

Unfortunately, we typically can't cite primary studies like this directly, because there are lots and lots of them and as Wikipedia editors we aren't qualified to pick and choose. For example, how are we to know that we should cite Braunschweig et al. and not the following more-recent study, which seemingly is somewhat contradictory?

• Morris CM, Zimmerman AW, Singer HS. Childhood serum anti-fetal brain antibodies do not predict autism. Pediatr Neurol. 2009;41(4):288–90. doi:10.1016/j.pediatrneurol.2009.04.014. PMID 19748049.

In order to puzzle this stuff out, we really need to cite reliable secondary sources (reviews of the research), and not primary sources (individual studies) directly, For more about this topic, please see Primary, secondary and tertiary sources and Respect secondary sources. In an area like autism, which is extremely well reviewed, we should be looking at the reviews. Enstrom et al. is one such review (and I expect that it reviews the primary study you mention) and I'll see if I can find others. Eubulides (talk) 21:56, 16 February 2010 (UTC)

Once again you are claiming the rules on primary sources are different than they actually are. In fact, primary sources, such as research papers from peer reviewed science journals, are NOT forbidden, despite the claims you make that they are. They are NOT.

In fact, the rules specifically give conditions under which they may be cited, and the proposed wording I've used for this change is permitted under those rules.

You don't own this article, and frankly I am upset that you continue to assert veto power over what other people put in when those edits are not forbidden and are good writing.

Please explain to me how this proposed change is not allowed.

Also, please explain exactly how the finding that CHILDHOOD antibodies do not correlate with autism refutes the claim that MATERNAL ANTIBODIES DURING PREGNANCY do.

I submit to you that it's you who are in conflict with the rules because you revert other people's edits on spurious grounds on a continual basis.

Also, I just followed the link you gave and reread the abstract, and they specifically do NOT question the findings from the people at MIND, read your own link before commenting on it, please. —Preceding unsigned comment added by 75.61.141.31 (talk) 01:53, 17 February 2010 (UTC)

It's not a matter of an actual contradiction, it's a matter of a seeming contradiction. And this was just one example: I'm sure that we could come up with dozens of primary studies about antibodies and autism. Why cite this one and not the others? Isn't that a WP:WEIGHT issue? And furthermore, your comment didn't address the points about respecting secondary sources. Why should this article be based on primary sources, which are lower quality, when better secondary sources are available? Eubulides (talk) 02:08, 17 February 2010 (UTC)

We shouldn't be citing primary studies, per WP:MEDRS and WP:RECENTISM. SandyGeorgia (Talk) 02:29, 17 February 2010 (UTC)

Remember when we did this before, and I proved to both of you that the links you give as justification for not using primary sources say no such thing? That in fact, by giving rules for using primary sources, they made it very clear such sources ARE allowed?

Both of you insist on misrepresenting the rules, it certainly can not be accidental. Normally I would not mind but this research, which seems to be borne out by time unless they are just plain lying about their findings, lead to a means by which mothers can avoid having autistic children, and it's simply too important that the information gets out to those who need it.

Basically, Eubulides feels he owns this article, and no one else gets to contribute, which is also of course in conflict with the rules.

So, your are both doubly wrong, and preventing potentially very helpful information from reaching the public, which, it's true, time may not bear out but that has also never been refuted in any peer reviewed journal articles I can find, and so meets the criteria for inclusion.

What's ironic to the point of being funny is that Eubulides states the article is simply too good to be messed up with what he regards as "bad" sources, when in fact it's a terrible article on many levels. —Preceding unsigned comment added by 75.61.141.31 (talk) 03:04, 17 February 2010 (UTC)

The article may be fine, I have not read it, but, SandyGeorgia is right, MEDRS and RECENTISM apply here. Oh, and if the autism article is so awful, propose changes here and get consensus. Dbrodbeck (talk) 03:26, 17 February 2010 (UTC)

I looked for other recent reviews of immune factors and autism and found this one, published in October–December 2009:

• Stigler KA, Sweeten TL, Posey DJ, McDougle CJ. Autism and immune factors: a comprehensive review. Res Autism Spectr Disord. 2009;3(4):840–60. doi:10.1016/j.rasd.2009.01.007.

This review summarizes over 100 studies, including several from the MIND group (for example, Ashwood et al. 2008, PMID 18762342). Here's how Stigler et al.'s Discussion section begins:

"Results from studies of immune function in autism to date have largely been contradictory (Table 1). As reviewed, the investigation of subgroups of patients has yielded specific abnormalities, some of which have been replicated. What is the cause of these abnormalities? Are they secondary to the disease processes that cause autism and the resultant neurobiological dysfunction?"

This seems a reasonable summary of what is known about autism and the immune system overall. No doubt the Autism article can be improved further, but since we have recent reviews on this topic there's really no need to be citing primary sources. Eubulides (talk) 10:08, 17 February 2010 (UTC)

If the primary paper is contradicted by a secondary review, we should give weight to the secondary review. If this 10% group is a real, consistent difference, then the subtype of autism this represents should be extremely easy to identify. It'll take time to sort out of course, but a genuine, consistent difference should pop out rather quickly once identified. If the study is actually being confirmed by other researchers then we'll see a second paper in the future, and review articles. Which we can then cite. WLU (t) (c) Wikipedia's rules:^simple/_complex 21:33, 17 February 2010 (UTC)

I am having trouble posting but will try:

1. No one has quoted me anything in the rules which says "no primary sources" That is because it does not exist.

2. Eubulides wants to keep control of the article, and so, using the rule that says primary sources are disfavored if a secondary source contradicts them, he bring a quote from a review article saying immune system findings have been contradictory. But E, the article seems to be referring to AUTISTICS, not MOTHERS OF AUTISTICS. It does not at all refute the paper I cited. So, it is not relevant to this discussion.

3. —Preceding unsigned comment added by 75.61.137.120 (talk) 06:51, 18 February 2010 (UTC)

3. In fact, as best I can find, there is nothing in the literature that contradicts anything in the paper from MIND. Eubulides has tried to cite two papers from other researchers which were both irrelevant because they are about the patients, not their mothers. Am I supposed to believe both his citations were honest mistakes? I do not. I think he wants to totally control this article, and he knows that he does not have the right to do that, so he comes up with spurious reasons for rejecting changes. He's trying, in very bad faith, to use the "contradicted by other researchers" rule for deleting primary sources, but it does not work, because the citations he gives do not say that. —Preceding unsigned comment added by 75.61.137.120 (talk) 06:59, 18 February 2010 (UTC)

4. I will respond to the comments about the research from MIND being either verified or refuted by other researchers:

First, if anyone else has done either thing, they seem to not have published. Although one would think that would take place eventually, the assumption that it would be done any time soon is not logical. These studies are expensive and require relatively large sample sizes. Even to do just a small study for replication of results requires dozens of autistics and dozens of controls, and to do specialized testing involving fetal brain. You can wait years for replication, but Wikipedia rules do not at all require that, they only require not giving undue weight to one study or to recent results just because they are recent. —Preceding unsigned comment added by 75.61.137.120 (talk) 07:07, 18 February 2010 (UTC)

5. Finally, I will tell you all why the article is truly terrible, since someone asked.

The article does not bring out the following points in a way which is comprehensible to a normal person:

1. Autism is a diagnosis, a condition, which can be caused by many different biological processes, and there is proven to be great heterogeneity in people diagnosed with autism. Just for example, it's very closely tied to Fragile X syndrome, but the vast majority of autistics do not have Fragile X. That by itself means there is more than one cause.

This point needs to be made in simple language, although technically the article does say there are many causes, it's so confounded with needless jargon that this is easily lost. Bad writing, really bad writing. —Preceding unsigned comment added by 75.61.137.120 (talk) 07:16, 18 February 2010 (UTC)

The point is, people are considered to be autistic when some cause or causes makes their behavior such that it fits the diagnostic criteria used by the medical profession to define autism.

Also, the fact that most autistics are mentally retarded should be emphasized more, and that autism is widely believed to be a brain condition ultimately. Certainly this is implied but again it's done poorly, when it should be the central and most clear part of the article. For example, the article says most autistics never live independently, without explaining this is due to retardation, so that a naive reader would be left wondering, "why, do they stack blocks all day and fail to feed themselves?" Again, very bad writing.

Genetics is also a part that should be rewritten. They keep looking for the causative genes but not finding them, and yet the article used Identical twin studies as proof of genetic cause. How about in utero effects? Should be balanced on that point.

Finally, it should be emphasized brain can malfunction countless ways, it's the most complex thing in universe and can screw up in the most complex ways. Very important point. —Preceding unsigned comment added by 75.61.137.120 (talk) 07:43, 18 February 2010 (UTC)

Please see #Brain can malfunction in countless ways below. Eubulides (talk) 08:27, 18 February 2010 (UTC)

Brain can malfunction in countless ways

• "it should be emphasized brain can malfunction countless ways" Yes it can, but I'm not sure how this could be said in a way that could be directly supported by a reliable review on autism. Perhaps specific wording could be suggested?
• "the article says most autistics never live independently, without explaining this is due to retardation" That's because the sources we found didn't say that it was due to retardation. If we can find good sources to support that claim, it should be added. However, for classic autism the fraction of individuals who also meet the criteria for mental retardation has been estimated at anywhere from 25% to 70%, and the fraction is much smaller for ASD in general, so we'll have to take great care in sourcing this claim.
• "Autism is a diagnosis, a condition, which can be caused by many different biological processes, and there is proven to be great heterogeneity in people diagnosed with autism." This point is covered in Causes, which says "there is increasing suspicion that autism is instead a complex disorder whose core aspects have distinct causes that often co-occur." If this isn't clear enough, perhaps you could suggest better wording? We need something that's well supported by reliable sources, but you can look at the sources for the current claim, or find other reviews.
• "How about in utero effects? Should be balanced on that point." The Causes section spends about 220 words on genetic causes, and about 180 words on in utero and other environmental causes. If anything, this is unbalanced in favor of environmental causes, since the consensus is that autism is mostly genetic (with heritability estimates like 0.7 and 0.9 being typical).
• "Genetics is also a part that should be rewritten. They keep looking for the causative genes but not finding them, and yet the article used Identical twin studies as proof of genetic cause." No, the Autism article does not do that. It says "Studies of twins suggest that heritability is 0.7 for autism and 0.9 for the broader autism phenotype, and siblings of those with autism are about 25 times more likely to be autistic than the general population." There is no suggestion here that autism is always caused by genetics, nor that the evidence for genetic cause is due entirely to identical twin studies.
• "the article seems to be referring to AUTISTICS, not MOTHERS OF AUTISTICS" No, Stigler et al. 2009 (doi:10.1016/j.rasd.2009.01.007) cites several studies on maternal immune factors, including Warren et al. 1990 (PMID 2273013), Croen et al. 2005 (PMID 15699309), Croen et al. 2008 (PMID 18554566), Singer et al. 2008 (PMID 18093664), and others (I got tired of writing them down).
• 'No one has quoted me anything in the rules which says "no primary sources"' We aren't saying primary sources should never be used. However, Wikipedia articles should be based on reliable, published secondary sources and primary sources should be avoided when these better secondary sources are available.
• "there is nothing in the literature that contradicts anything in the paper from MIND" There is nothing in the literature that contradicts anything in hundreds of other primary studies on autism either. So why is Braunschweig et al. worth promoting, over and above all those other hundreds of primary studies? Particularly when we have good reviews in this area?
• The most recently published high-quality review of autism in general that we know of is the review by Levy et al. (PMID 19819542), published in the November 2009 Lancet. It's quite a good and extensive review: 12 pages of small print. And it doesn't mention the immune-factor theories at all. In contrast, Autism is already devoting some space to these theories, supported by a reliable review, namely Stigler et al. 2009 (doi:10.1016/j.rasd.2009.01.007). Why should Autism devote even more space to immune-factor theories? Autism is already going out on a WP:WEIGHT limb to some extent, by devoting more space to it than Levy et al. and other reliable sources do.

Eubulides (talk) 08:27, 18 February 2010 (UTC)

Eubulides: All I can say is "there you go again" I looked at your links to support your claim, and for the third time, you've cited studies which are either not relevant, or do not refute the research on maternal antibodies which I wish to include in the article. The first most recent review article link does not have an abstract that tells me much and I do not have access to the full article, so I can not comment on what it says, but the others either are research on different subjects entirely, OR, in the case of the last citation, (Singer) they tend to support the general principle of maternal antibodies of molecular weight 36 and 39 kDa being tied to autism.

In other words, you repeatedly misrepresent the truth in order to find some reason to exclude valid material.

After three strikes, I say you are out and though I can't access the review article, since every other citation you've used to support your point did not do so, the odds are high that this one does not either.

I don't believe you are acting sincerely to make these repeated misrepresentations, you are just trying to find any excuse to keep out content you did not generate. That is the bottom line. —Preceding unsigned comment added by 75.61.137.120 (talk) 16:39, 18 February 2010 (UTC)

Please assume good faith, that is what we do around here. As noted, there are quite a few primary sources out there, and there is nothing currently that suggests what you want to add is better than all of those. We have a featured article here, a rather mature one at that. It is an article that relies secondary sources as WP:MEDRS suggests. Dbrodbeck (talk) 16:53, 18 February 2010 (UTC)

I am going to write up proposed changes for the other parts of the article. —Preceding unsigned comment added by 75.61.137.120 (talk) 16:49, 18 February 2010 (UTC)

Anon, I firmly agree with Dbrodbeck - your statements border on accusation, ascribe ownership issues where I can only see patient answers to your statements with referal to our policies, and fail to assume good faith. If you are going to put the effort into preparing draft changes, I would suggest getting an account and using a sub-page of your userspace. WLU (t) (c) Wikipedia's rules:^simple/_complex 18:55, 18 February 2010 (UTC)

"you've cited studies which are either not relevant, or do not refute the research on maternal antibodies" The studies I cited were not intended to refute Braunschweig et al. 2008 (PMID 18078998): they were only intended to refute the comment that Stigler et al. 2009 (doi:10.1016/j.rasd.2009.01.007) "seems to be referring to AUTISTICS, not MOTHERS OF AUTISTICS". As Stigler et al. cites several primary studies about maternal immune factors, this comment was clearly mistaken. As WP:MEDRS#Respect secondary sources advises, Autism should not attempt to cite a primary study to dispute a review, and when in doubt it's better to omit mention of a primary study. Dbrodbeck's and WLU's suggestions are good ones. Eubulides (talk) 20:00, 18 February 2010 (UTC)

Review article new to PubMed.gov today: Curr Opin Neurol. 2010 Feb 12. [Epub ahead of print] The immune system's role in the biology of autism. WP:COPYVIO redacted - WLU Goines P, Van de Water J. PMID 20160651 [PubMed - as supplied by publisher] —Preceding unsigned comment added by 75.61.137.120 (talk) 00:59, 19 February 2010 (UTC)

Please don't post lengthy quotes - it's a copyright violation and if it's just the abstract, particularly unnecessary if we have the PMID. Also note the concluding sentence includes the word may which means it's still not settled. Otherwise, as a recent review article it certainly seems like it could be included - though I'm not sure of the quality of the journal (for good or bad, I'm not familiar with the title or it's impact factor). WLU (t) (c) Wikipedia's rules:^simple/_complex 01:24, 19 February 2010 (UTC)

OK, sorry for the lengthy quotes, but unlike the text of the actual article, I thought the entire abstract should be included because so much of it had relevance to the discussion here.

I don't want to get in some kind of huge squabble with people who are acting in good faith, but, you may not know the history, I was told that only secondary sources are allowed for Wikipedia articles, and this is simply not true.

Also, studies which did not refute the primary source seemed to be used as refutation, with the intention that this would mean the primary source could not be used.

In other words, I did not feel like I was being dealt with in good faith.

Good faith would have been first of all, making an ACCURATE statement of the rules, and conceding that primary sources are allowed as long as used correctly, ie, identified as primary sources, not refuted, not given undue weight compared to other sources, not given undue weight because they are recent.

None of these rules excluded the primary source I wanted to use.

I say again, none of these rules were applicable to the source as I proposed to use it, so the objection became "how do we know this primary source is better than any other"? which is NOT the standard for inclusion, and, since autism has so many different causes, really a denial of one of the main points the article should be making.

Put another way, all of the valid causes of autism should be included in the article. Those that are merely theoretical or are in serious dispute probably should not.

Those that are based on findings that have not been refuted and are not actually disputed should be included with caveats as to them being relatively preliminary findings.

This is science by the way, especially science of brain diseases that are not narrowly defined.

To be more specific, the fact that a secondary source, a review article, mentions related issues, but does NOT specifically refute a primary source, is not reason for exclusion under the rules, and, once again, if as many top line researchers believe, there are many different causes for autism, keeping information about any single cause out can not be justified on an intellectual basis. You are not removing chaff, or probable chaff, you are just taking out meaningful information. —Preceding unsigned comment added by 75.61.137.120 (talk) 03:52, 19 February 2010 (UTC)

Maybe I should try further to clear this up. Many, probably most, researchers in this field believe there is more than one, and possibly dozens, of causes for what is labeled autism per the diagnostic manual. If they are right, then findings of one primary source, such as "deletion on chromosome 9 causes autism" is NOT refuted, it's really important to emphasize this, is NOT REFUTED, by another paper that says, "copy number variation on chromosome 7 causes autism".

I've given an example of genetic causes, but of course it could have been "Maternal antibodies to fetal brain molecules of both 37 and 73 kDa, are almost always associated with autism" per the article I wanted to include, and that statement would not at all be refuted by mention of any other cause. It also would not be refuted by the publication of a review article which gave it short shrift, or for that matter did not mention it at all.

To make a crude analogy, suppose we were talking about causes of obesity. Someone says it's genetic, someone says it's environmental, someone says it's related to other disease processes, and all are right.

What Eubulides is doing is excluding say for example, the people who say, "Obesity is related to other disease processes" because their research is relatively recent, when in fact a huge percentage of all the research on autism is under 5 years old, and a lot is under 2.

He's not keeping it out because it's been refuted or even disputed, or any of the other valid reasons. That is what upsets me. In other words, rather than being a correct reason per the rules, it's an arbitrary over-extension of the general rule favoring secondary sources to an iron-clad prohibition of primary sources.

And the article, under his watchful eye, suffers greatly for it. —Preceding unsigned comment added by 75.61.137.120 (talk) 04:10, 19 February 2010 (UTC)

Primary sources should be used with great caution and shouldn't be used when there are secondary sources available - we also shouldn't use primary sources to 'debunk' or contradict secondary sources. Secondary sources take precedence because they are meant to synthesize and interpret primary sources into a coherent statement about some aspect of the topic they are talking about. our page on medical sources has been pointed to many times now, but I'll link to it once more - please read the whole page WP:MEDRS. It's not up to Eub or anyone to "make accurate statements about the rules" - it is up to you to review the policies and guidelines that have been cited/linked and determine if your suggestion still has merit within their scope and context. Wikipedia is a complicated place, and this is a featured article - so care should be taken in adjusting it. I haven't seen anything cited by Eub that causes me to raise an eyebrow, and I love the rules. It is up to you to try to learn and interpret them now. Given the complexity of autism and the huge amount of research that is focused on it, we can't include every new primary source, every new suggested finding, or even every new secondary source. Until there is substantial evidence that maternal antibodies has revolutionized the field or are now acknowledged as a primary driver of autism, it's inappropriate to give it a large amount of text because it is competing with a large amount of other information and speculation. However, if it's a true cause of autism, this will be determined and most scholars will begin to acknowledge it. The only thing that will bear this out is time while the results are replicated and extended outside of the MIND institute. So again, Eub's resistance to the inclusion of a primary source is appropriate per MEDRS, as well as general guidelines on primary sources, wikipedia not being a crystal ball, wikipedia not being a place to promote ideas and our efforts to avoid emphasizing new information just because it is new. The number of times a paper has been cited is a rough measure of its importance and (along with many other factors) helps inform how much emphasis we place on it. In these cases, the citation levels are zero simply because they are so new. So if this is a genuine finding, it will be replicated and extended and we can place more emphasis on it. If it revolutionizes the field, you will have the satisfaction of being right. But it may not, and we shouldn't be re-writing the article just yet. In the case of a lengthy article with a huge amount of publications on the topic, though primary sources are not necessarily prohibited, we should certainly be very cautious with them. I don't see Eub's approach as unwarranted or prejudicial, I see it as keeping an eye on the larger context and very justifiably expressing caution about a very recent publication. WLU (t) (c) Wikipedia's rules:^simple/_complex 12:05, 19 February 2010 (UTC)

I think we are down to the gist of this thing. The judgment call on whether or not a primary source is worth including taking all the points you've made, and all I have made, into consideration. This judgment call can not be made without reference to the underlying subject matter,and my point is, the subject matter actually demands inclusion of a great many findings which will actually tend to create confusion among the reader, because that IS the state of research, and, if the strong belief of many front line researchers is correct, will remain the state of research for some time.

Recent reearch, and it's true I am getting this partly from MIND but certainly not entirely form there, indicates great hetergeneity among autistics. Study after study, by finding different abnormalities, which would not seem to be causally related to each other, indicate this heterogeneity. In other words, there are certainly several causes for what is called autism by a clinician, and there may even be hundreds. HUNDREDS.

Excluding primary studies, when the pace of research is such that they are not immediately replicated or refuted, because other researchers are too busy with their own projects to do replication, or the studies are too expensive to redo, does not help the article, because you are throwing out solid information for the sake of brevity and clarity which does not exist in the real world.

IF, a big IF, the state of research was such that brevity and clariy were possible, then I would be wrong and your side would be right, but it's simply not where the research has taken us.

What really ought to happen with the article is, instead of throwing out all the primary sources people try to put in, it should be explained both that the causes of autism are subject to great controversy and emotion among lay people and parents of autistics, but that the science does not lead to any one cause either, except that genetics are suspected to be the main factor in most cases, but that "genetics" in this sense could mean epigenetics, (article is good on that point I think, though maybe could add to explanation of epigenetics) and that "genetics" certainly does not mean one gene, or even several genes, but can include all the genes that could impact on brain development and function, that is, one autistic may have defect in developing synapes, another may have defect in developing certain brain regions, another has defect in neural migration, etc, etc, etc.

Although I recognize that a laymen's article can not review all the possible ways the brain can malfunction, to knowck out solid research which could be in the article is wrong, not just from the standpoint of helping people who might otherwise have an autistic child, because they did not know of testing that could have been done, but it simply does not give a good account of the state of knowledge.

Some problems are very complex, and autism is certainly one of them. In fact, I don't pretend to know enough about medicine to say this with certainty, but it seems fairly likely if you count all the causes that are likely to be eventually discovered, it could be the most complext "disease" that exists or ever will exist. Finally though, I would like to get back to the MIND findings a little. I am not saying they will stand up to time, and I am sure many people feel strongly that some cause of autism is the cause, and that their treatment program has to be promulgated to the public, and avoiding this kind of preaching is a good thing for the article.

Also, don't misunderstand, even if that research is 100% borne out, I'm not saying it's the Holy Grail of autism, it's still only about 10%, so 90% of autistics are not impacted by it.

But, if it does bear out, it's something that can help a lot of people. I don't propose to build it up excessively, just to put it out there as one cause among many.

Sorry for going on so long and if I got too upset before. —Preceding unsigned comment added by 76.232.11.175 (talk) 14:57, 19 February 2010 (UTC)

I agree there are lots of proposed and confirmed causes of autism. So many, we have an article on it: Causes of autism which in turn links to other articles on further subtopics. As you say, this is an area of active research with lots of possible theories and preliminary results, etc. This is precisely why we should use reviews as then we get an expert to review to literature and indicate the weight that should be given to each idea. See WP:WEIGHT for why we need to consult reliable secondary sources to determine whether to include a viewpoint or issue. We have recent reviews both on the topic of autism and the specific topic of immune-related causes. I'm pretty sure we can rely on a review titled "Autism and immune factors: a comprehensive review." to be a good judge of the area and pick out the important points. Colin°^Talk 15:54, 19 February 2010 (UTC)

Excluding primary studies does help the article because it does not get filled with minutea and blind alleys; until the results are replicated and extended, this is speculation with minor confirmation - not solid information. Review articles allow a distance in time, thought and opinion that primary aritcles of necessity lack. I read the causes section as basically doing what you suggest - pointing to autism as a primarily genetic, heritable, but poorly understood condition, with no one "winning" hypothesis; accordingly, details should be suggested at talk:Causes of autism since there is no single "cause" yet known beyond a genetic contribution. You open your previous post with this being a judgement call - several other editors have now made a judgement call that it is too early to substantially rewrite, or give the findings more importance than those already explored in review articles. WLU (t) (c) Wikipedia's rules:^simple/_complex 16:20, 19 February 2010 (UTC)

Just to reemphasize, the primary research paper on maternal antibodies has been duplicated by the same researchers, and, more important perhaps from the standpoint of objective credibility, other researchers up at MIND created "autistic" monkeys by using the same antibodies to fetal brain. This was done with controls such that it seems certain something in these mothers causes brain abnormalities, in other words, not "Speculation with minor confirmation", but an extensive testing of the hypothesis through an alternate method.

I think you put too much emphasis on the idea that there are a lot of researchers out there skeptical of other researcher's findings and willing and able to test them by duplicating their efforts. What I've heard from a lot of them, not just MIND, but many others, is a firm belief that there are many causes of autism, and with that belief, it does not make sense to replicate other studies, unless you are really suspicious they are wrong, or think you can advance their findings. It makes sense to keep on with what you think will be most productive for you.

And taking it further, if these researchers are right, and there are many different causes, then it may be many primary studies will never be refuted in any secondary review articles, and frankly, a state of great confusion will reign for decades. Though the article is for layman and they can not understand the many different possible mechanisms of brain malfunction, conveying the complexity of the current state of knowledge is entirely appropriate.

At the very least, I would suggest that the article be changed to say, something like, "autism can be caused in many different ways", and use as examples Fragile X, some of the rare genetic defects, and then, with full identification as a preliminary finding, maternal antibodies. By the by, not to debate the underlying science too deeply, but the article cites high identical twin concordance as proof of genetic cause 70 to 90% of cases, but actually, identical twins are in the same mother at the same time, so in utero effects can not generally be excluded as causes, and in fact, if you think about the failure to find causative genes, in utero effects become the next place to look. I don't have easy access to full text of research papers, but there have to be many people who have expressed or admitted this same point. —Preceding unsigned comment added by 75.61.137.65 (talk) 14:46, 21 February 2010 (UTC)

Please see #Some good points below. Eubulides (talk) 00:14, 22 February 2010 (UTC)

Some good points

Some good points are made in the previous comment, but this still doesn't address the central WP:WEIGHT concern about the Autism article's promoting this particular primary study as opposed to lots of other primary studies. There are hundreds of them.

• Why not, for example, also discuss the role of fetal testosterone? After all, that is a pet project of the Cambridge group, just as maternal antibodies are a pet project of the MIND group. Or why not discuss XMRV, the preferred hypothesis at the Whittemore Peterson Institute? Or oxidative stress, the favorite at the UT Southwestern Medical Center? or maternal stress, the favorite at the Albert Einstein College of Medicine? or iodine deficiency, promoted at UT San Antonio? There are dozens more. I've purposely mentioned hypotheses that are both better- and worse-supported than the maternal antibody hypotheses: but none of these hypotheses are currently mentioned in Autism because there's not enough room to discuss all these single-research-group theories. So why pick the maternal antibody theory? It's not stronger than all these other theories, and it's weaker than some of them.
• "it does not make sense to replicate other studies" I'm afraid that it does. In a field like this there are simply too many ways to go wrong, even with experienced researchers. For example, the MIND group's widely publicized 2002 report to the California legislature concluded that the autism epidemic is real (as opposed to being an artifact of changed diagnostic criteria etc.), but these results have not been confirmed by researchers outside the group, despite multiple attempts to do so.
• "it may be many primary studies will never be refuted in any secondary review articles" That's not how reviews typically work. Reviews typically assemble the best studies in an area; bad primary studies are often simply excluded. This doesn't mean the unreviewed studies should be cited! I'm not saying that Braunschweig et al. 2008 (PMID 18078998) is a bad study—far from it—but we Wikipedia editors are emphatically not qualified to put this primary study into proper perspective without a reliable outside secondary review on the topic.
• "autism can be caused in many different ways" Yes, and this is the main point of the first paragraph of the Causes section. Improvements to the wording there will be welcome; we need reliable sources.
• "use as examples Fragile X" Yes, and that is already used as an example.
• "some of the rare genetic defects" These are less well supported. None of the genetic defects associated with autism (and this includes fragile X) have been shown to selectively cause autism.
• "then, with full identification as a preliminary finding, maternal antibodies" Unfortunately there are dozens of "preliminary findings" like that. There is simply not enough room to discuss all this sort of speculation here. Better homes for this more-speculative material are Causes of autism and (for genetic hypotheses) Heritability of autism.

Eubulides (talk) 00:14, 22 February 2010 (UTC)

Agree with all of that, and generally very uncomfortable with attempts to introduce primary sources or unreplicated research or research not yet subjected to secondary reviews. SandyGeorgia (Talk) 00:15, 22 February 2010 (UTC)

Agree. Until secondary sources indicate that this idea has general acceptance, a large volume of text on this point is inappropriate. Wikipedia follows mainstream opinion within the scholarly community, it does not set it. We don't decide what is "important", we wait until the reviews indicate the point is important. Please let it go - there are multiple editors who have indicated this is not appropriate until it shows up in a reliable, secondary source. When that happens, and when there are many sources indicating this is the or at least an important theory, then we can give it more space. It's not there yet. WLU (t) (c) Wikipedia's rules:^simple/_complex 13:22, 22 February 2010 (UTC)

I agree as well, let's wait until this shows up in a reliable secondary source. Indeed, let us move on. Dbrodbeck (talk) 13:25, 22 February 2010 (UTC)

All of you are wrong, and to be very blunt about it, if I edit the article to include this reearch, as long as it's done correctly, you will be wrong to revert it, because your action, rather than mine, will be in violation of the rules. Now, I will address these points one by one: First, testosterone during pregnancy is not forbidden from being in this article. If someone put it in, in the proper way according to the rules, anyone who removed it would be wrong to do so. —Preceding unsigned comment added by 76.232.8.186 (talk) 21:55, 26 February 2010 (UTC)

Actually, to be blunt, you are wrong. Wikipedia operates on consensus, and I see no consensus here. Well, except that the consensus is that your interpretation of policy is incorrect. Dbrodbeck (talk) 22:33, 26 February 2010 (UTC)

If "consensus" meant that the majority rules, and the majority is taken relatively randomly from editors who post in the discussion section, then you would be right, but those are not the rules, and really should not be, because of all the reasons given above. —Preceding unsigned comment added by 76.245.44.215 (talk) 21:15, 10 April 2010 (UTC)

You (assuming you are the same as the other IP above) have had this explained to you already. See WP:MEDRS. There is a consensus among everybody that has commented, other than you, that your interpretation is incorrect. As well, there is no consensus to add the material you (again, hard to tell as the IPs change, but you seem like the same person) want to. Please let us move on. Dbrodbeck (talk) 23:16, 10 April 2010 (UTC)

No, there was no consensus my interpretation of rules was wrong, in fact, after much pulling of teeth, it was finally admitted my interpretation was right and then the argument was changed to "your primary source has been disputed", and when I proved that inaccurate, "only secondary sources should be used", not because it's a rule, just because certain people felt that was better policy. Please understand: It's very likely the findings at MIND are accurate, they've been partly replicated at Kennedy Krieger, and, if people who have autistic children are informed of this research, by reading Wikipedia for example, they could get the antibody test and possibly save themselves from having a second child with autism. This is a huge potential benefit to the people who are reading this article, or about 12% of them anyway. I keep pushing this because of the possibility you can save much human misery, without even violating Wikipedia rules, and what I've gotten in response to this is repeated misrepresentation of the rules and of the findings of studies which were used in an attempt to refute this one, which actually in one case, the study from a very prominent mental retardation institution, confirmed these findings. Please reconsider this. If the findings are accurate, and remember, the other study Eubulides quoted above tends to confirm them, you are keeping really valuable information away from people who desperately need it. It's not like the other theories he mentioned, or at least not all of them, in that those theories do not lead to a test which will probably tell a subset of parents of autistic children if they are going to have another one. It's the possibility of sparing innocent people terrible problems which has me so worked up about it. Since it's allowed, and it might do a lot of good to include it, will someone please tell me the overarching reason not to?

Children bias

Everything I read about autism, including the page on Wikipedia, mentions "children" with autism all the time and virtually no mention of adults. Autism spectrum disorders don't magically disappear when the child grows up into an adult! TurboForce (talk) 15:01, 25 March 2010 (UTC)

This is a common complaint in the autistic community. I think there are several reasons for this: 1) Asperger and Kanner studied children, not adults, and the diagnostic criteria are easier to apply to children than to adults, since autistic adults are more likely to be aware of their strange behavior and in control of themselves enough to be able to modify it. 2) Generally parents of autistics will want to get them diagnosed as soon as possible, while they're still young. 3) Advertisers (for autism funds, etc) tend to be more successful portraying autism as a syndrome of children than of adults ... people are more likely to feel sympathy for a 6 year old who suffers from social problems and can't make friends than a 30 year old who still suffers from social problems and can't make friends, and more likely to believe that the 6 year old could be cured. It's just human nature. —Soap— 18:18, 25 March 2010 (UTC)

Agree with Soap. Some of the children bias is unavoidable because it is a condition that appears in childhood, is typically diagnosed in childhood and for which the most effort at interventions are aimed at children. Nearly all our sources deal with children and we (as Wikipedians) are unable to extrapolate any of those comments/findings to adults. We are at the mercy of our sources to some degree and this topic dominates paediatric healthcare and is significantly poorer in adult healthcare. I think we should make an effort cover adults more than we do (our sources are biased simply because a paper in a paediatric journal will naturally leave the adult commentary to "...and when they grow up..." tagged on the end). In each relevant section, we should consider if we can say more about adults. I have fixed one instance were we gratuitously assumed the topic was children, but didn't spot any other such cases. BTW: autism isn't unique in this regard. Colin°^Talk 19:34, 25 March 2010 (UTC)

Everything I read about autism spectrum disorders just mentions "children" at every opportunity. For the autism related pages on Wikipeida, can we please remove the constant mention of "children" and replace it with "people" with autism. This REALLY annoys me and it's about time the "children" bias was removed once and for all. People who read the pages about autism seem to think that autism "only" affects children. SORT IT OUT!!!!!!!!!!! TurboForce (talk) 01:01, 10 April 2010 (UTC)

Please see my reply above. We can't just replace "children" with "people". The research the information was based on looked at children, not adults or both. There's simply not the same degree of information about adults with autism as there is on children. If you want to change that, petition the charities and other funding bodies to do more research on adults. Colin°^Talk 09:04, 10 April 2010 (UTC)

Autism and social influence

New and interesting aritcle out from the AJS about the spatial relationship between children with autism and diagnosis of autism - [10]. The gist appears to be that living near someone with autism increases your chances of being diagnosed, due to the greater awareness and exchange of information by parents. Sciencebasedmedicine has a post on it, gotta love a lay summary - [11]. WLU (t) (c) Wikipedia's rules:^simple/_complex 13:22, 14 April 2010 (UTC)

Spironolactone and Autism

A report about the effects of Spironolactone on childen with autism. article was locked, couldn't add, putting it here instead. http://www.autismresourceconnection.com/documents/BTL-SpironolactoneandAutism_Version_7_JB.pdf —Preceding unsigned comment added by 66.21.133.205 (talk) 23:55, 8 May 2010 (UTC)

Please see WP:MEDRS and WP:RECENTISM. SandyGeorgia (Talk) 23:58, 8 May 2010 (UTC)

understood, impressed by the speed =) —Preceding unsigned comment added by 66.21.133.205 (talk) 00:20, 9 May 2010 (UTC)

Opening paragraph, clarification

"The two other autism spectrum disorders (ASD) are Asperger syndrome, which lacks delays in cognitive development and language, and PDD-NOS, diagnosed when full criteria for the other two disorders are not met.[3]" This sentence is quite confusing. I would like to change it to:

"The two other autism spectrum disorders (ASD) are Asperger syndrome, which lacks delays in cognitive development and language, and PDD-NOS, diagnosed when the full criteria for autism or Asperger syndrome are not met.[3]" Bj norge (talk) 09:09, 24 May 2010 (UTC)

Seems clearer to me. Anthony (talk) 09:23, 24 May 2010 (UTC)

Let's also add "which is" before "diagnosed when the full criteria ". Using the same wording formula for each of the disorder-defining clauses helps remove any ambiguity as to which definition applies to which disorders. Colin°^Talk 11:57, 24 May 2010 (UTC)

I agree. I made the edit and added "the full set of criteria" since "when (the) full criteria" might be a bit ungrammatical. Bj norge (talk) 15:59, 24 May 2010 (UTC)

I think we should say, "We don't know what autism is. We can only describe it in this way or a limited way...."71.166.10.215 (talk) 13:26, 11 June 2010 (UTC)

I disagree. The opening sentence is uncontroversial, and the lead clearly states that there is a lot we don't know about autism. Lova Falk talk 17:07, 11 June 2010 (UTC)

Edit Request

{{editsemiprotected}} Inconsistency between Graphic and Accompanying Text

The article makes a clear distinction between autism and ASD. However, under this section "Epidemiology" there is a graph with explanatory text that says: "Reports of autism cases per 1,000 children grew dramatically in the U.S. from 1996 to 2007. It is unknown how much, if any, growth came from changes in autism's prevalence."

This graph is meant to illustrate the text of the article which says, as the first clause in the first sentence: "Most recent reviews tend to estimate a prevalence of 1–2 per 1,000 for autism and close to 6 per 1,000 for ASD; ..."

Manifestly, the text accompanying the graph should substitute "ASD" for "autism" or just graph the increase in reported autism ending with the most recent figure of 1.2 per thousand. As it stands, the graph is both inaccurate with the facts supported by citations in the text of the article and misleading with these huge bars reaching six without referencing scale.

Well, I checked the source, and it just says "autism", but it looks like it's just a "for public consumption" release, and surely the original study that led to the figures must provide more information. I'll look at this in the morning if no one else does, but as of yet I wouldn't feel comfortable making the change. It's not necessarily a contradiction to have two different figures in the text, even if they are far apart; some studies have different critera than others. For example a psychologist can tell someone they have autism, but cannot give an official diagnosis; so does the person have autism or not? Under some studies they would be considered autistic, others not. —Soap— 02:17, 31 May 2010 (UTC)

Not done for now: The graph was self-made according to the criteria specified on its description page. The IDEA(B) criterion is worded thus: Autism means a developmental disability significantly affecting verbal and nonverbal communication and social interaction, generally evident before age three, that adversely affects a child's educational performance. Other characteristics often associated with autism are engagement in repetitive activities and stereotyped movements, resistance to environmental change or change in daily routines, and unusual responses to sensory experiences. It seems fairly ambiguous whether or not that refers specifically to autism or to the wider spectrum. What do you reckon? I've skim-read the article but all I've managed to glean is that all prevalence figures are widely variable estimates... haz (talk) 13:34, 31 May 2010 (UTC)

Autism is NOT an "emotional disorder".

Misplaced comment moved from Wikipedia talk:Protection policy 123.222.215.125 (talk) 01:46, 4 June 2010 (UTC)

It is a developmental disorder and rightly linked here to questions of theory of mind. Terming it an emotional disorder in the context of an article about attachment disorders risks subscribing to discredited and sexist theories blaming "refrigerator mothers" for their child's autism. —Preceding unsigned comment added by 86.146.255.241 (talk) 17:00, 27 May 2010 (UTC)

Copied this comment to Talk:Attachment disorder and replaced "emotional disorder" with "neurodevelopmental disorder" in Attachment disorder. Anthony (talk) 03:42, 4 June 2010 (UTC)

New Nature

New publication from Nature about autism and genetics. I can't understand anything it says :( but I think it's free full text. ['http://www.nature.com/nature/journal/vaop/ncurrent/full/nature09146.html]. WLU (t) (c) Wikipedia's rules:^simple/_complex 20:14, 9 June 2010 (UTC)

For a less technical blog about it, there's http://www.psychologytoday.com/blog/radical-behaviorist/201006/cnvs-the-true-cause-autism (of course, it isn't suitable as a source, even though the writer is definitely qualified). Also, that article isn't a peer-reviewed secondary source, but I found 3 that may be useful, albeit less recent, through PubMed:

• PMID 19910074
• PMID 19895225
• PMID 19736351

Unfortunately, none of the full texts are available without a subscription or purchase. MichaelExe (talk) 23:40, 10 June 2010 (UTC)

I've got the 1st and 3rd on my hard drive but am reading other stuff right now. I can email to someone who will read them. Anthony (talk) 13:53, 11 June 2010 (UTC)