Talk:Tourette syndrome/Archive 6

Dan Aykroyd
Speculation, most likely a joke, and thoroughly discussed at Sociological and cultural aspects of Tourette syndrome: should be reverted whenever added here, to keep this article to encyclopedic content. Sandy 02:57, 25 September 2006 (UTC)

Unvoluntary/involuntary
Although the term unvoluntary is well-defined in this article and in the literature, it is frequently changed to involuntary here, perhaps by well-meaning editors who believe it is a typo, and don't read the text.
 * Tics are described as semi-voluntary or "unvoluntary", because they are not strictly involuntary—they may be experienced as a voluntary response to an unwanted, premonitory urge.

Please restore it whenever it is removed, as it is the correct usage. I will add an inline note. Sandy (Talk) 16:39, 24 October 2006 (UTC)

Pete Bennett
Content is often added here: is included already, along with contemporaries, at Sociological and cultural aspects of Tourette syndrome, this article discusses highly notables such as Mozart and Johnson, with others summarized in the daughter article. Sandy (Talk) 20:46, 25 October 2006 (UTC)
 * In general, only very notable cultural references are included in this article, while others may be mentioned at Sociological and cultural aspects of Tourette syndrome. Sandy Georgia (Talk) 20:12, 25 June 2007 (UTC)

interesting concept, i do not know this english rule! Would anyone care to eloborate —Preceding unsigned comment added by 193.63.48.253 (talk) 14:35, 15 November 2007 (UTC)
 * Please see WP:MEDMOS. Sandy Georgia  (Talk) 14:44, 15 November 2007 (UTC)

"Epidemiology" much improved
After deliberately not looking at Epidemiology for a few days I came back to find it much improved. Thanks to everybody who labored on it! It sounds like everybody's tired of the subject now so I don't blame you for taking a break, but when you get back here are some other things to think about. Eubulides 06:28, 19 June 2007 (UTC)
 * Thanks for the compliment, but yes, we're tired of it. Sandy has now gone on indefinite leave. Stirring this up more is only likely to make it definite. I strongly believe we've polished this text enough. There are plenty other major medical topics on wiki (e.g. epilepsy) that are utterly neglected wrt epidemiology and would be vastly improved with only a fraction of the effort being expended here. Colin°Talk 13:54, 19 June 2007 (UTC)
 * Thanks for the suggestion: I rewrote Epilepsy from scratch. Reviews welcome. Eubulides 22:49, 19 June 2007 (UTC)


 * Several of these changes are not minor, and resulted in changes in meaning and subtle misinterpretations or errors; I'll be putting the article later for several hours of work to make some corrections as well as some general and MEDMOS updates.  Sandy Georgia  (Talk) 20:53, 25 June 2007 (UTC)

Minor changes
I streamlined the wording slightly in ways I hope nobody objects to, and changed two or three relatively minor things:


 * I added the date of DSM-IV-TR (i.e., 2000) since it seems relevant to the year-2000 changes so often referred to in this section.
 * Since the section mentioned Stern et al.'s upper-bound estimate of 106,000 children with TS in the UK, it should also mention their lower-bound estimate of 64,000. I added that.
 * Two wikilinks to ascertainment bias were close to each other, which was overkill: I removed one.

Eubulides 06:28, 19 June 2007 (UTC)


 * Your changes improve the text IMO by a small margin in a few places. In other places I don't detect an improvement, just a difference. It should be noted that I've failed to spot subtle errors before. Colin°Talk


 * Some of the tweaks don't result in an improvement, but do introduce subtle errors. I'll make corrections later when I put the article inuse for multiple updates.
 * Regarding this edit, please take note of WP:EL—links to blogs and personal web pages should be avoided, except those written by a recognized authority. Any physician can put up a blog or a webcast on Tourette's; the standard established here was physicians published and well-recognized by their TS peers.  Please don't lower the standards expected per WP:EL and established in this article, as this may lead to the introduction of lower-quality sources, blogs, webcasts, and lay literature such as seen in the sourcing at Asperger syndrome.  Only the highest quality sources, such as medical textbooks and well-recognized TS researchers, were used here.  The description of Freeman's and Walkup's credentials is not "puffery", rather the justification per WP:EL.
 * The revised version gives authoritative-enough credentials for both Freeman and Walkup. We don't need to know that many details from Freeman's resume, and it was a bit odd to see so much for Freeman and so little for Walkup. Their current posts are plenty enough credentials, and it appears to me that the revised version gives more-impressive credentials for Walkup than the original did. Eubulides 23:51, 25 June 2007 (UTC)
 * Including a blog on Wikipedia requires stronger justification than a TSA-sponsored video. (See WP:EL).  Sandy Georgia  (Talk) 01:22, 26 June 2007 (UTC)
 * Regarding the change from NIH to NINDS, the object is to pitch the article to Wiki readership. Most readers are more familiar with the NIH, so I suggest a combined version (NIH/NINDS).
 * Combining NINDS and NIH is fine. Eubulides 23:51, 25 June 2007 (UTC)
 * I used NINDS/NIH. Sandy Georgia  (Talk) 01:22, 26 June 2007 (UTC)
 * Regarding this edit: 1) Consensus of everyone else who reviewed the draft was that the first sentence was much clearer.  The tweak isn't necessary, and I don't find it more clear.  In particular, I'd like to stick as closely as possible to Robertson's wording, which says something different than you changed it to, partly because I'm curious about her reference in one of the publications to something unpublished.  2) Your change from present to past tense on the methodological issues in the studies leaves the impression that these issues no longer occur, which isn't correct.  Progress has been made, but these kinds of issues continue to affect TS studies.  3) Regarding the introduction of the lower bound, you earlier objected that we would confuse readers by introducing different ranges, but we now have Scahill's 1 to 10 and Robertson's 6 to 10.  I won't remove the Robertson range, since it's more accurate than the 1 in 1,000 (which gives undue weight to an old number), but we may be risking overburdening the reader with a lot of dates, numbers, and meaningless factoids.  Sandy Georgia  (Talk) 21:16, 25 June 2007 (UTC)
 * I assume (1) is about replacing "has been reported in all parts of the world" with "worldwide". The revision is certainly shorter, and I don't find it any less clear, but if you do then by all means revert it. I assume (2) is about replacing "have an inherent bias" with "were inherently biased". Here I thought the subject was studies "until the 1980s" so the past tense was appropriate; if not, then obviously it should be the present tense but in that case some rewording seems called for since the subject confused me at least. For (3) the inherent confusion comes from the fact that the Wikipedia article gives three different ranges for the same thing (1–10, 1–11, and 6–10 per 1,000), not from giving the lower-bound that corresponds to the 6. I continue to think that it's misleading to give just the upper bound of a range when the source gives both lower and upper bounds: Wikipedia should not suggest more confidence in an estimate than its source does. Eubulides 23:51, 25 June 2007 (UTC)
 * I think our readers are smart enough to figure out that different sources give different examples. Sandy Georgia  (Talk) 01:22, 26 June 2007 (UTC)


 * Hi, three minor changes in progress you may wish to revert or not. I moved the picture of Dr. Tourette out of the infobox to History though he may belong sooner. I found two sources in Google Scholar, and will add them though I may need help with formatting without cite templates. This should only take a few more minutes. Also I think I can fix the minor copyright problem with the naming of the syndrome although I don't know much about it. -Susanlesch 01:00, 9 September 2007 (UTC)
 * The lead doesn't need to be overcited (everything in the lead is cited in the text), the image is more appropriate in the infobox, Lesch-Nyhan is not related to Tourette syndrome, and it's not completely correct to call Tourette a neurologist by today's standards. What copyright problem with the naming of the syndrome?  Sandy Georgia  (Talk) 04:25, 9 September 2007 (UTC)
 * Well, I wrote ok to revert, so thanks for the revert. Bunch of comments below. Best wishes. -Susanlesch 06:43, 9 September 2007 (UTC)

Suggestions for further improvement in "Epidemiology"
Here are some suggestions for improvement. I can draft wording along these lines if people like. Eubulides 06:28, 19 June 2007 (UTC)


 * The biggest risk of further polishing (or the "surgery" proposed below) is that errors are introduced merely to make the text shorter, flow better or a tiny bit more precise in the opinion of one editor. Have a look at Sandy's sandbox history and the amount of text now archived. Those recent edits reflect two full days of effort by Sandy. In addition I've spent hours and hours reviewing several iterations of both your and Sandy's text (I did Sandy's sandbox reviews by email). Tony, one of Wikipedia's foremost copy editors, has reviewed and adjusted the drafts. That's surely enough. Colin°Talk


 * I spent considerable time too. I realize people want to take a break. There's no rush in making it better. Eubulides 19:45, 19 June 2007 (UTC)


 * In the 1st paragraph, the end of the second sentence "…prevalence is much higher among children than adults." is repetitive with the start of the third sentence "Children are five to twelve times more likely than adults to be identified…", and this could be reworded and streamlined. Eubulides 06:28, 19 June 2007 (UTC)
 * Extremely minor issue; not worth changing. Colin°Talk 13:54, 19 June 2007 (UTC)
 * Two different issue (tic disorders and TS), not worth changing. Better to make it clear; not all of our readers are trained in statistics, epidemiology, or any of the other issues we may all understand.  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * I agree it's only a minor annoyance. Eubulides 19:45, 19 June 2007 (UTC)
 * In the 2nd paragraph, the topic sentence "Discrepancies across prevalence estimates come from several factors: ascertainment bias in earlier samples drawn from clinically referred cases, assessment methods that may fail to detect milder cases, and differences in diagnostic criteria and thresholds." should be put at the start of the paragraph, not in the middle. The paragraph is harder to follow with the topic sentence in the middle, and could stand some surgery to make it shorter and sweeter. Eubulides 06:28, 19 June 2007 (UTC)
 * This is just your opinion of what is the "topic sentence" and the need for that to be first. Seems to me there are several related topics in this paragraph. Colin°Talk 13:54, 19 June 2007 (UTC)
 * There are many important topics in that paragraph. It's not as simple as you imply; to me, the most important concept in that paragraph is that these methodological issues continue to affect TS studies, so I disagree with any reorganization of the paragraph.  The points that need to be made were made in the draft.  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * The paragraph has three topics: (a) ascertainment bias, (b) difficulty in diagnosing milder cases accurately, and (c) changes to diagnostic criteria. The sentence I mention outlines all three topics and ties them together. The current first sentence of the paragraph abruptly launches into part of topic (a) without explaining what the paragraph is up to. The paragraph would be better-organized if its first sentence summarized to the reader what the paragraph is about. Eubulides 19:45, 19 June 2007 (UTC)
 * The paragraph has more than three topics. These tweaks aren't necessary, and move the text away from its meaning (these issues existed in the past, and continue to exist today).  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * OK, then the paragraph has four topics: the three I mentioned, plus the fact that these problems still plague TS studies. So we can revise the topic sentence to be something like "Discrepancies across prevalence estimates are continuing problems that come from several factors: ascertainment bias in samples drawn from clinically referred cases, assessment methods that may fail to detect milder cases, and differences in diagnostic criteria and thresholds." The "are continuing problems" puts the most important concept first. This particular wording is only a suggestion of course: the point is that the paragraph currently starts off with something that isn't a good topic sentence and the naive reader has no clue as to what the paragraph will be about. Eubulides 23:51, 25 June 2007 (UTC)
 * IMO, the bigger confusion was coming from the DSM-IV-TR clause, which I dropped. There are many reasons that milder cases are increasingly recognized, so along with your suggestion (below) to drop the 2000 date, dropping the 2000 ref to the DSM should help clear up all of this.  Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)
 * I ran out of steam yesterday after fighting with the pmid named refs, and failed to finish up this point. I just changed the second paragraph and very slightly reworded to bring forward the new topic sentence as you suggested.  I prefer the emphasis on "current and prior" rather than "continuing problems", since the problems are beginning to be addressed.  Please have a look at the new order; I agree it's a definite improvement in flow, from topic to 3 points to conclusion of increased estimates.  Sandy Georgia  (Talk) 14:42, 26 June 2007 (UTC)


 * The 3rd paragraph gives the mistaken impression that TS was thought to be rare before 2000 and that multiple studies since 2000 show otherwise. The paragraph should make it clearer that prevalence estimates grew by approximately 3 orders of magnitude from circa 1970 to circa 2000, and by about 1 order of magnitude since then; it's not like there was just one giant leap around 2000. Eubulides 06:28, 19 June 2007 (UTC)
 * I don't read it that way and think the jumps you mention are clear enough. Colin°Talk 13:54, 19 June 2007 (UTC)
 * I think it's clear, and the introduction of more factoids will only confuse readers. It sounds like you're trying to read too much into the text.  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * No new facts would be needed; all it needs is minor wording changes. Eubulides 23:51, 25 June 2007 (UTC)
 * The current wording is no doubt clear to experts but I think it can stand improvement for non-experts. Eubulides 19:45, 19 June 2007 (UTC)
 * In fact, I see the opposite; these past and proposed tweaks are taking the text away from what layreaders understand.   Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * OK, here's a way to fix the problem by removing a fact: remove the phrase "By 2000, a review estimated a lifetime prevalence of 0.5–1.0 per 1,000;" and its associated reference. This will make the text easier for lay readers to understand and will remove the red herring about whether 0.5–1.0 per 1,000 is "rare". Eubulides 23:51, 25 June 2007 (UTC)
 * Good suggestion, I did that (it ties to the other DSM-IV-TR issue, which isn't the only reason milder cases are now recognized - increasing awareness is also part of it, as well as better study methodology). Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)
 * In the 3rd paragraph, the 1972 NIH story is overstated. The NIH didn't believe there were fewer than 100 cases; some reviewers thought it. And reviews (presumably anonymous, and told indirectly via an anecdote) are not the kinds of sources we'd like to see here. Let's remove this reference; the point is already fully supported by the 1973 registry citation, which is much sounder than the NIH story. Eubulides 06:28, 19 June 2007 (UTC)
 * What is the harm here? The source for the 1973 "registry" is currently a book aimed at lay readers so it has flaws too. If representatives of the NIH have stated something, then it is fair to say the NIH has stated something. I don't think anybody seriously doubts this fact, so what exactly is your point? Colin°Talk 13:54, 19 June 2007 (UTC)
 * I haven't seen evidence that "representatives of the NIH have stated" fewer than 100 cases in the US in 1972. All I have seen is anecdotal evidence that some (anonymous?) reviewers (who most likely did not work for the NIH?) said that. That kind of evidence is too weak to be worth citing. I went to the library and found an old citation (Woodrow 1974) in a refereed journal that is better than either the NIH or the 1973 registry references, and substituted it for them. I hope this resolves the matter. Eubulides 19:45, 19 June 2007 (UTC)
 * Representatives would not be a correct word, but since this is named in at least three reliable sources that I have at hand (and probably many more if I went to a library), I don't see why we should exclude it. One of the most meaningful comments on the FAC—from an editor I respect—was that the article was perfectly pitched.  Many of these proposed and past changes are causing the article to lose that pitch.  Readers don't want/need an endless string of meaningless dates, numbers, percentages and factoids.  They need things they can relate to.  They can relate to the fact that the condition was once thought to be so rare that funds weren't allocated to its study.  That is exactly the kind of story that gives the article meaning to readers, which can't be conveyed by introducing more meaningless factoids and numbers.  I can't understand why you're calling the NIH issue "weak evidence"; it's well documented by multiple reliable sources. Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * OK, you talked me into the anecdote, so long as it's reworded so that it's clear we're just talking about anonymous reviewers here. Let's leave the Woodrow 1974 reference in though, though, since it's a number in a refereed journal. Eubulides 23:51, 25 June 2007 (UTC)
 * I removed Woodrow because it's too many factoids. And I'm not sure why you say we're talking about anonymous reviewers; I don't have that in any source, so it would be original research to add it.  Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)
 * The footnote for the 3:1 to 4:1 M:F ratio is overkill. It suffices to cite just one reliable source for this, and either the NIH or the TSA will do. None of the additional references are persuasive: they're mostly just mentioning the numbers as conventional wisdom (and are thus no more authoritative than the NIH or the TSA), and the exception (Leckman & Cohen) gives a range 1.6:1 to 9.3:1, which contradicts the main text of "Epidemiology". My earlier objection in this area was primarily that the section's numbers didn't match the cited sources' numbers. It's problematic to cite Leckman & Cohen without mentioning their numbers in the section. It's better to just give one good citation since that suffices. Eubulides 06:28, 19 June 2007 (UTC)
 * Yes, it may be overkill but I don't know Sandy's rationale. It does no harm.
 * IMO, it should stay for two reasons. 1) You asked for it, and other readers may want this info for the same reasons; considering the work is done, and it's not burdening the reader (it's in a note), there's no good reason to lose it.  More importantly, 2) the NIH is not the highest-quality source.  Note that the NIH is single-handedly responsible for furthering (in that article) the incorrect notion that GTS first described TS (he didn't, Itard did).  I don't like to cite the NIH, did it as little as possible and in cases where their writing was the most clear, but recognizing that their writing isn't always accurate.  I also don't like to cite only the TSA, and use them for cases when their writing is the most clear.  In this case, although research numbers differ, it happens that the TSA and the NIH agree on the 3 to 4 consensus.  If we take that out, we may be chasing our tail on this issue in the future with another editor.  The documentation of the different estimates and consensusis there, no need to lose it.  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * The main harm I see is that the numbers given in the Wikipedia article disagree with the numbers given in the citation. In general, if reliable sources disagree, the main text should reflect this disagreement. Eubulides 19:45, 19 June 2007 (UTC)
 * The consensus about the various numbers is given by the citation, which provides a complete explanation. Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * It still bugs me that the "complete explanation" gives numbers that disagree with the main text. How about putting the excess detail into a Wiki comment instead? Eubulides 23:51, 25 June 2007 (UTC)
 * I could do that (add it under Notes rather than Sources you mean?), but the main note does give the 3 and 4 numbers, so I'm not clear why we need to change it. Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)
 * The section says there is an "emerging consensus" that 1–11 children per 1,000 have TS. But it cites Leckman 2002 which states "the prevalence of Tourette's syndrome is presently estimated to be between 31 and 157 cases per 1000 in children aged 13–14-years" (this in turn merely cites Hornesey et al. 2001, but Leckman makes it sound like that's the best numbers we have). The section also cites Robertson 2005 which states "seven recent studies have remarkably consistent findings and suggest a prevalence of between 0.4% and 1.76% for children aged 5–18 years old". This doesn't sound like a consensus to me; it sounds like experts in the field are giving quite different estimates. The section should not claim a consensus based on one paper. Eubulides 06:28, 19 June 2007 (UTC)
 * It does not cite Leckman 2002. It cites Scahill 2006. You also say the section cites "Robertson 2005" but I think you are looking at a different paper from the one cited. The cited paper says "a prevalence figure of between 0.6% and 1% in mainstream schoolchildren (using DSM-IV-TR criteria) is more likely to be correct than the previous lower figures that have been reported". That fits within the "consensus". It sounds to me we have a editor-stalemate on whether there is a "consensus". I'd much rather the we leave that debate to the experts than between WP editors. The text meets WP:V in this regard. Colin°Talk 13:54, 19 June 2007 (UTC)
 * The sentence saying "emerging consensus" does not have a citation. Someone who reads this section for the first time won't necessarily know that Scahill 2006 is the source for the 1–11 per 1,000 figure unless they check all the section's citations. If they read the whole section, they'll see Scahill 2006 but they'll also see the other citations I mentioned, which give numbers well outside the 1–11 per 1,000 range. But there's a more-important issue than clarifying the source here. It's not obvious that 1–11 per 1,000 is actually the emerging consensus. One paper does not make a consensus. Again, if reliable sources disagree, the Wikipedia article should reflect that disagreement: it's not our job to arbitrate disagreements among experts. Eubulides 19:45, 19 June 2007 (UTC)
 * I'll add another citation on Scahill if you think one isn't enough. Scahill is clear and in the same range as Robertson and in the same range as Hirtz; I don't see why that's not clear.  The only thing that now has undue weight in the article is the lower bound of 1 per 1,000.  If you want to extend to Robertson's upper bound of 17 now, we can do that also.  It will eventually be proven anyway, but for now, it seems to make sense to explain that most recent studies (even though they're still missing cases) are honing in on that 6 to 8 range.  The article does that now.  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * A second citation from Scahill wouldn't suffice; we'd need a second citation from an independent source giving the same 1–11 range. The ranges from Scahill, Hirtz, and Robertson overlap but it's a stretch to say that there's an "emerging consensus" on 1–11; that's Scahill's range, not the other guys'. Eubulides 23:51, 25 June 2007 (UTC)
 * Well, should we just eliminate all of this verbiage and go back to Swerdlow's 1 to 2% overall prevalence ? I'm sorry, but Scahill is about as reliable as a source gets, his numbers are in the same range as Robertson, and I don't understand why we should reject reliable sources.  Putting it together our own way would be original research.  Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)

In conclusion: further polishing introduces the risk of errors which IMO is not justified by the benefits. I've responded to your points to show they are not straightforward "improvements". You may be able to convince me that you can make the changes without causing errors, but I'd really rather not spend further time on this. Prolonging this discussion is IMO not healthy for WP. Colin°Talk 13:54, 19 June 2007 (UTC)


 * I agree that polishing is secondary. My main concern is having the text match the cited sources. The current article does a good job in that area now, but there are still a few glitches as noted above. Eubulides 19:45, 19 June 2007 (UTC)


 * We can spend all the time you want here. In the meantime, Schizophrenia passed FAR in bad need of a copyedit, Chagas disease needs a review, Epilepsy and Polio are neglected, but we can tweak here forever if you'd like.  We haven't changed anything important yet, but we are moving away from the "perfectly pitched article" that was mentioned on the FAC, and into introducing many meaningless factoids, numbers, percentages, and dates (which was, BTW, the only objection to the article on FAC).  Sandy Georgia  (Talk) 21:46, 25 June 2007 (UTC)
 * I disagree that "we haven't changed anything important"; the epidemiology section is stronger now. I take your point that other articles are weaker in this area. However, I think this discussion has been useful precisely because Tourette syndrome is so well-written: improving its epidemiology section can serve as a useful role model for the other, weaker articles. Eubulides 23:51, 25 June 2007 (UTC)
 * I agree the article is stronger now, but it would be nice if we could all get some work done on other articles. I think we're done here for now.  (I wasted a lot of time trying to implement the new MEDMOS suggestion about naming refs to agree with the PMID number, only to decide that was a bad idea—opens up too much potential for error, so I reverted myself.  Not sure if that suggestion belongs in MEDMOS at all.)  Sandy Georgia  (Talk) 01:35, 26 June 2007 (UTC)
 * However, I think this discussion has been useful precisely because Tourette syndrome is so well-written: improving its epidemiology section can serve as a useful role model for the other, weaker articles. By the way, this is the first time I've noticed you saying something positive about this article.  I'm glad to hear it; a lot of people put a lot of effort into it, and keeping it to the same standard of accuracy and citation as when it passed FAC is a constant labor.  I really am sorry that we couldn't get schizophrenia, autism or Asperger syndrome to the same level, but there's only so many hours in a day. Thanks for helping.  Sandy Georgia  (Talk) 01:45, 26 June 2007 (UTC)
 * I entitled this section '"Epidemiology" much improved'; didn't that count as saying something positive? Eubulides 04:27, 26 June 2007 (UTC)